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

use super::{G1Affine, G2Affine, PublicParameters, GT};
use crate::base::impl_serde_for_ark_serde_unchecked;
use ark_ec::pairing::{Pairing, PairingOutput};
use ark_serialize::{CanonicalDeserialize, CanonicalSerialize};
use itertools::MultiUnzip;
use num_traits::One;

/// The transparent setup information that the prover must know to create a proof.
/// This is public knowledge and must match with the verifier's setup information.
/// See Section 3.3 of https://eprint.iacr.org/2020/1274.pdf for details.
///
///
/// Note:
/// We use nu = m and k = m-i or m-j.
/// This indexing is more convenient for coding because lengths of the arrays used are typically 2^k rather than 2^i or 2^j.
pub struct ProverSetup<'a> {
    /// `Gamma_1[k]` = Γ_1,(m-k) in the Dory paper.
    pub(super) Gamma_1: Vec<&'a [G1Affine]>,
    /// `Gamma_2[k]` = Γ_2,(m-k) in the Dory paper.
    pub(super) Gamma_2: Vec<&'a [G2Affine]>,
    /// `H_1` = H_1 in the Dory paper. This could be used for blinding, but is currently only used in the Fold-Scalars algorithm.
    pub(super) H_1: G1Affine,
    /// `H_2` = H_2 in the Dory paper. This could be used for blinding, but is currently only used in the Fold-Scalars algorithm.
    pub(super) H_2: G2Affine,
    /// `Gamma_2_fin` = Gamma_2,fin in the Dory paper.
    pub(super) Gamma_2_fin: G2Affine,
    /// `max_nu` is the maximum nu that this setup will work for
    pub(super) max_nu: usize,
    /// The handle to the `blitzar` Gamma_1 instances.
    #[cfg(feature = "blitzar")]
    blitzar_handle:
        blitzar::compute::MsmHandle<blitzar::compute::ElementP2<ark_bls12_381::g1::Config>>,
}

impl<'a> ProverSetup<'a> {
    /// Create a new `ProverSetup` from the public parameters.
    pub(super) fn new(
        Gamma_1: &'a [G1Affine],
        Gamma_2: &'a [G2Affine],
        H_1: G1Affine,
        H_2: G2Affine,
        Gamma_2_fin: G2Affine,
        max_nu: usize,
    ) -> Self {
        assert_eq!(Gamma_1.len(), 1 << max_nu);
        assert_eq!(Gamma_2.len(), 1 << max_nu);
        #[cfg(feature = "blitzar")]
        let blitzar_handle = blitzar::compute::MsmHandle::new(&Vec::from_iter(
            Gamma_1.iter().copied().map(Into::into),
        ));
        let (Gamma_1, Gamma_2): (Vec<_>, Vec<_>) = (0..max_nu + 1)
            .map(|k| (&Gamma_1[..1 << k], &Gamma_2[..1 << k]))
            .unzip();
        ProverSetup {
            Gamma_1,
            Gamma_2,
            H_1,
            H_2,
            Gamma_2_fin,
            max_nu,
            #[cfg(feature = "blitzar")]
            blitzar_handle,
        }
    }

    #[cfg(feature = "blitzar")]
    #[tracing::instrument(name = "ProverSetup::blitzar_msm", level = "debug", skip_all)]
    pub(super) fn blitzar_msm(
        &self,
        res: &mut [blitzar::compute::ElementP2<ark_bls12_381::g1::Config>],
        element_num_bytes: u32,
        scalars: &[u8],
    ) {
        self.blitzar_handle.msm(res, element_num_bytes, scalars)
    }
}

impl<'a> From<&'a PublicParameters> for ProverSetup<'a> {
    fn from(value: &'a PublicParameters) -> Self {
        Self::new(
            &value.Gamma_1,
            &value.Gamma_2,
            value.H_1,
            value.H_2,
            value.Gamma_2_fin,
            value.max_nu,
        )
    }
}

/// The transparent setup information that the verifier must know to verify a proof.
/// This is public knowledge and must match with the prover's setup information.
/// See Section 3.3 of https://eprint.iacr.org/2020/1274.pdf for details.
///
///
/// Note:
/// We use nu = m and k = m-i or m-j.
/// This indexing is more convenient for coding because lengths of the arrays used are typically 2^k rather than 2^i or 2^j.
#[derive(CanonicalSerialize, CanonicalDeserialize, PartialEq, Eq, Debug, Clone)]
pub struct VerifierSetup {
    /// `Delta_1L[k]` = Δ_1L,(m-k) in the Dory paper, so `Delta_1L[0]` is unused. Note, this is the same as `Delta_2L`.
    pub(super) Delta_1L: Vec<GT>,
    /// `Delta_1R[k]` = Δ_1R,(m-k) in the Dory paper, so `Delta_1R[0]` is unused.
    pub(super) Delta_1R: Vec<GT>,
    /// `Delta_2L[k]` = Δ_2L,(m-k) in the Dory paper, so `Delta_2L[0]` is unused. Note, this is the same as `Delta_1L`.
    pub(super) Delta_2L: Vec<GT>,
    /// `Delta_2R[k]` = Δ_2R,(m-k) in the Dory paper, so `Delta_2R[0]` is unused.
    pub(super) Delta_2R: Vec<GT>,
    /// `chi[k]` = χ,(m-k) in the Dory paper.
    pub(super) chi: Vec<GT>,
    /// `Gamma_1_0` is the Γ_1 used in Scalar-Product algorithm in the Dory paper.
    pub(super) Gamma_1_0: G1Affine,
    /// `Gamma_2_0` is the Γ_2 used in Scalar-Product algorithm in the Dory paper.
    pub(super) Gamma_2_0: G2Affine,
    /// `H_1` = H_1 in the Dory paper. This could be used for blinding, but is currently only used in the Fold-Scalars algorithm.
    pub(super) H_1: G1Affine,
    /// `H_2` = H_2 in the Dory paper. This could be used for blinding, but is currently only used in the Fold-Scalars algorithm.
    pub(super) H_2: G2Affine,
    /// `H_T` = H_T in the Dory paper.
    pub(super) H_T: GT,
    /// `Gamma_2_fin` = Gamma_2,fin in the Dory paper.
    pub(super) Gamma_2_fin: G2Affine,
    /// `max_nu` is the maximum nu that this setup will work for
    pub(super) max_nu: usize,
}

impl_serde_for_ark_serde_unchecked!(VerifierSetup);

impl VerifierSetup {
    /// Create a new `VerifierSetup` from the public parameters.
    pub(super) fn new(
        Gamma_1_nu: &[G1Affine],
        Gamma_2_nu: &[G2Affine],
        H_1: G1Affine,
        H_2: G2Affine,
        Gamma_2_fin: G2Affine,
        max_nu: usize,
    ) -> Self {
        assert_eq!(Gamma_1_nu.len(), 1 << max_nu);
        assert_eq!(Gamma_2_nu.len(), 1 << max_nu);
        let (Delta_1L_2L, Delta_1R, Delta_2R, chi): (Vec<_>, Vec<_>, Vec<_>, Vec<_>) = (0..max_nu
            + 1)
            .map(|k| {
                if k == 0 {
                    (
                        PairingOutput(One::one()),
                        PairingOutput(One::one()),
                        PairingOutput(One::one()),
                        Pairing::pairing(Gamma_1_nu[0], Gamma_2_nu[0]),
                    )
                } else {
                    (
                        Pairing::multi_pairing(
                            &Gamma_1_nu[..1 << (k - 1)],
                            &Gamma_2_nu[..1 << (k - 1)],
                        ),
                        Pairing::multi_pairing(
                            &Gamma_1_nu[1 << (k - 1)..1 << k],
                            &Gamma_2_nu[..1 << (k - 1)],
                        ),
                        Pairing::multi_pairing(
                            &Gamma_1_nu[..1 << (k - 1)],
                            &Gamma_2_nu[1 << (k - 1)..1 << k],
                        ),
                        Pairing::multi_pairing(&Gamma_1_nu[..1 << k], &Gamma_2_nu[..1 << k]),
                    )
                }
            })
            .multiunzip();
        Self {
            Delta_1L: Delta_1L_2L.clone(),
            Delta_1R,
            Delta_2L: Delta_1L_2L,
            Delta_2R,
            chi,
            Gamma_1_0: Gamma_1_nu[0],
            Gamma_2_0: Gamma_2_nu[0],
            H_1,
            H_2,
            H_T: Pairing::pairing(H_1, H_2),
            Gamma_2_fin,
            max_nu,
        }
    }
}

impl From<&PublicParameters> for VerifierSetup {
    fn from(value: &PublicParameters) -> Self {
        Self::new(
            &value.Gamma_1,
            &value.Gamma_2,
            value.H_1,
            value.H_2,
            value.Gamma_2_fin,
            value.max_nu,
        )
    }
}