1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289
//! This module defines methods and structures for setting up the circuit, or in
//! a more theoretical language, the "NP relation" that the circuit will be
//! related to.
//! Note that when mentioning "circuit" in this context, we are referring to
//! a specific user application in addition to the circuit used to encode the
//! verifier.
//!
//! The setup phase defines the constraints that the computation/the app must
//! satisfy, the evaluation domains, and the SRS for the polynomial commitment
//! scheme. Generally, the setup phase is an agreement between the prover and
//! the verifier on the language and the protocol parameters (cryptographic
//! primitives, security level, etc). The setup phase will also contain some
//! pre-computed values to ease both the prover's and the verifier's work.
//!
//! As part of the setup phase, the parties will also agree on a set of
//! predefined values that will shape the selectors and the computation.
//!
//! A prover will be providing a proof of a particular [IndexedRelation] created
//! during the setup phase, by encapsulating a value of this type in its
//! [crate::witness::Env] structure. The prover will then refer to the values
//! saved in the type [IndexedRelation].
//!
//! On the other side, a verifier will be instantiated with the relevant indexed
//! relation.
//!
use ark_ff::PrimeField;
use kimchi::circuits::domains::EvaluationDomains;
use log::{debug, info};
use mina_poseidon::constants::SpongeConstants;
use mvpoly::{monomials::Sparse, MVPoly};
use num_bigint::{BigInt, BigUint};
use num_integer::Integer;
use o1_utils::FieldHelpers;
use poly_commitment::{ipa::SRS, PolyComm, SRS as _};
use std::{collections::HashMap, time::Instant};
use crate::{
column::Gadget,
constraint,
curve::{ArrabbiataCurve, PlonkSpongeConstants},
interpreter::{self, VERIFIER_STARTING_INSTRUCTION},
MAXIMUM_FIELD_SIZE_IN_BITS, MAX_DEGREE, MV_POLYNOMIAL_ARITY, NUMBER_OF_COLUMNS,
NUMBER_OF_GADGETS, VERIFIER_CIRCUIT_SIZE,
};
/// An indexed relation is a structure that contains all the information needed
/// describing a specialised sub-class of the NP relation. It includes some
/// (protocol) parameters like the SRS, the evaluation domains, and the
/// constraints describing the computation.
///
/// The prover will be instantiated for a particular indexed relation, and the
/// verifier will be instantiated with (relatively) the same indexed relation.
pub struct IndexedRelation<
Fp: PrimeField,
Fq: PrimeField,
E1: ArrabbiataCurve<ScalarField = Fp, BaseField = Fq>,
E2: ArrabbiataCurve<ScalarField = Fq, BaseField = Fp>,
> where
E1::BaseField: PrimeField,
E2::BaseField: PrimeField,
{
/// Domain for Fp
pub domain_fp: EvaluationDomains<E1::ScalarField>,
/// Domain for Fq
pub domain_fq: EvaluationDomains<E2::ScalarField>,
/// SRS for the first curve
pub srs_e1: SRS<E1>,
/// SRS for the second curve
pub srs_e2: SRS<E2>,
/// The application size, i.e. the number of rows per accumulation an
/// application can use.
///
/// Note that the value is the same for both circuits. We do suppose both
/// SRS are of the same sizes and the verifier circuits are the same.
pub app_size: usize,
/// The description of the program, in terms of gadgets.
/// For now, the program is the same for both curves.
///
/// The number of elements is exactly the size of the SRS.
pub circuit_gates: Vec<Gadget>,
/// Commitments to the selectors used by both circuits
pub selectors_comm: (
[PolyComm<E1>; NUMBER_OF_GADGETS],
[PolyComm<E2>; NUMBER_OF_GADGETS],
),
/// The constraints given as multivariate polynomials using the [mvpoly]
/// library, indexed by the gadget to ease the selection of the constraints
/// while computing the cross-terms during the accumulation process.
///
/// When the accumulation scheme is implemented, this structure will
/// probably be subject to changes as the SNARK used for the accumulation
/// scheme will probably work over expressions used in
/// [kimchi::circuits::expr]. We leave that for the future, and focus
/// on the accumulation scheme implementation.
///
/// We keep two sets of constraints for the time being as we might want in
/// the future to have different circuits for one of the curves, as inspired
/// by [CycleFold](https://eprint.iacr.org/2023/1192).
/// In the current design, both circuits are the same and the prover will do
/// the same job over both curves.
pub constraints_fp: HashMap<Gadget, Vec<Sparse<Fp, { MV_POLYNOMIAL_ARITY }, { MAX_DEGREE }>>>,
pub constraints_fq: HashMap<Gadget, Vec<Sparse<Fq, { MV_POLYNOMIAL_ARITY }, { MAX_DEGREE }>>>,
/// Initial state of the sponge, containing circuit specific
/// information.
// FIXME: setup correctly with the initial transcript.
// The sponge must be initialized correctly with all the information
// related to the actual relation being accumulated/proved.
// Note that it must include the information of both circuits!
pub initial_sponge: [BigInt; PlonkSpongeConstants::SPONGE_WIDTH],
}
impl<
Fp: PrimeField,
Fq: PrimeField,
E1: ArrabbiataCurve<ScalarField = Fp, BaseField = Fq>,
E2: ArrabbiataCurve<ScalarField = Fq, BaseField = Fp>,
> IndexedRelation<Fp, Fq, E1, E2>
where
E1::BaseField: PrimeField,
E2::BaseField: PrimeField,
{
pub fn new(srs_log2_size: usize) -> Self {
assert!(E1::ScalarField::MODULUS_BIT_SIZE <= MAXIMUM_FIELD_SIZE_IN_BITS.try_into().unwrap(), "The size of the field Fp is too large, it should be less than {MAXIMUM_FIELD_SIZE_IN_BITS}");
assert!(Fq::MODULUS_BIT_SIZE <= MAXIMUM_FIELD_SIZE_IN_BITS.try_into().unwrap(), "The size of the field Fq is too large, it should be less than {MAXIMUM_FIELD_SIZE_IN_BITS}");
let modulus_fp = E1::ScalarField::modulus_biguint();
let alpha = PlonkSpongeConstants::PERM_SBOX;
assert!(
(modulus_fp - BigUint::from(1_u64)).gcd(&BigUint::from(alpha)) == BigUint::from(1_u64),
"The modulus of Fp should be coprime with {alpha}"
);
let modulus_fq = E2::ScalarField::modulus_biguint();
let alpha = PlonkSpongeConstants::PERM_SBOX;
assert!(
(modulus_fq - BigUint::from(1_u64)).gcd(&BigUint::from(alpha)) == BigUint::from(1_u64),
"The modulus of Fq should be coprime with {alpha}"
);
let srs_size = 1 << srs_log2_size;
let domain_fp = EvaluationDomains::<E1::ScalarField>::create(srs_size).unwrap();
let domain_fq = EvaluationDomains::<E2::ScalarField>::create(srs_size).unwrap();
info!("Create an SRS of size {srs_log2_size} for the first curve");
let srs_e1: SRS<E1> = {
let start = Instant::now();
let srs = SRS::create(srs_size);
debug!("SRS for E1 created in {:?}", start.elapsed());
let start = Instant::now();
srs.get_lagrange_basis(domain_fp.d1);
debug!("Lagrange basis for E1 added in {:?}", start.elapsed());
srs
};
info!("Create an SRS of size {srs_log2_size} for the second curve");
let srs_e2: SRS<E2> = {
let start = Instant::now();
let srs = SRS::create(srs_size);
debug!("SRS for E2 created in {:?}", start.elapsed());
let start = Instant::now();
srs.get_lagrange_basis(domain_fq.d1);
debug!("Lagrange basis for E2 added in {:?}", start.elapsed());
srs
};
let constraints_fp: HashMap<
Gadget,
Vec<Sparse<E1::ScalarField, { MV_POLYNOMIAL_ARITY }, { MAX_DEGREE }>>,
> = {
let env: constraint::Env<E1> = constraint::Env::new();
let constraints = env.get_all_constraints_indexed_by_gadget();
constraints
.into_iter()
.map(|(k, polynomials)| {
(
k,
polynomials
.into_iter()
.map(|p| Sparse::from_expr(p, Some(NUMBER_OF_COLUMNS)))
.collect(),
)
})
.collect()
};
let constraints_fq: HashMap<
Gadget,
Vec<Sparse<E2::ScalarField, { MV_POLYNOMIAL_ARITY }, { MAX_DEGREE }>>,
> = {
let env: constraint::Env<E2> = constraint::Env::new();
let constraints = env.get_all_constraints_indexed_by_gadget();
constraints
.into_iter()
.map(|(k, polynomials)| {
(
k,
polynomials
.into_iter()
.map(|p| Sparse::from_expr(p, Some(NUMBER_OF_COLUMNS)))
.collect(),
)
})
.collect()
};
// FIXME: note that the app size can be different for both curves. We
// suppose we have the same circuit on both curves for now.
let app_size = srs_size - VERIFIER_CIRCUIT_SIZE;
// Build the selectors for both circuits.
// FIXME: we suppose we have the same circuit on both curve for now.
let circuit_gates: Vec<Gadget> = {
let mut v: Vec<Gadget> = Vec::with_capacity(srs_size);
// The first [app_size] rows are for the application
v.extend([Gadget::App].repeat(app_size));
// Verifier circuit structure
{
let mut curr_instruction = VERIFIER_STARTING_INSTRUCTION;
for _i in 0..VERIFIER_CIRCUIT_SIZE - 1 {
v.push(Gadget::from(curr_instruction));
curr_instruction = interpreter::fetch_next_instruction(curr_instruction);
}
// Additional row for the Poseidon hash
v.push(Gadget::NoOp);
}
assert_eq!(v.len(), srs_size);
v
};
let selectors_comm: (
[PolyComm<E1>; NUMBER_OF_GADGETS],
[PolyComm<E2>; NUMBER_OF_GADGETS],
) = {
// We initialize to the blinder to avoid the identity element.
let init: (
[PolyComm<E1>; NUMBER_OF_GADGETS],
[PolyComm<E2>; NUMBER_OF_GADGETS],
) = (
std::array::from_fn(|_| PolyComm::new(vec![srs_e1.h])),
std::array::from_fn(|_| PolyComm::new(vec![srs_e2.h])),
);
// We commit to the selectors using evaluations.
// As they are supposed to be one or zero, each row adds a small
// contribution to the commitment.
// We explicity commit to all gadgets, even if they are not used in
// this indexed relation.
circuit_gates
.iter()
.enumerate()
.fold(init, |mut acc, (row, g)| {
let i = usize::from(*g);
acc.0[i] = &acc.0[i] + &srs_e1.get_lagrange_basis(domain_fp.d1)[row];
acc.1[i] = &acc.1[i] + &srs_e2.get_lagrange_basis(domain_fq.d1)[row];
acc
})
};
// FIXME: setup correctly the initial sponge state
let sponge_e1: [BigInt; PlonkSpongeConstants::SPONGE_WIDTH] =
std::array::from_fn(|_i| BigInt::from(42u64));
Self {
domain_fp,
domain_fq,
srs_e1,
srs_e2,
app_size,
circuit_gates,
selectors_comm,
constraints_fp,
constraints_fq,
initial_sponge: sponge_e1,
}
}
pub fn get_srs_size(&self) -> usize {
self.domain_fp.d1.size as usize
}
pub fn get_srs_blinders(&self) -> (E1, E2) {
(self.srs_e1.h, self.srs_e2.h)
}
}