core/vm, crypto/bls12381, params: add bls12-381 elliptic curve precompiles (#21018)
* crypto: add bls12-381 elliptic curve wrapper
* params: add bls12-381 precompile gas parameters
* core/vm: add bls12-381 precompiles
* core/vm: add bls12-381 precompile tests
* go.mod, go.sum: use latest bls12381 lib
* core/vm: move point encode/decode functions to base library
* crypto/bls12381: introduce bls12-381 library init function
* crypto/bls12381: import bls12381 elliptic curve implementation
* go.mod, go.sum: remove bls12-381 library
* remove unsued frobenious coeffs
supress warning for inp that used in asm
* add mappings tests for zero inputs
fix swu g2 minus z inverse constant
* crypto/bls12381: fix typo
* crypto/bls12381: better comments for bls12381 constants
* crypto/bls12381: swu, use single conditional for e2
* crypto/bls12381: utils, delete empty line
* crypto/bls12381: utils, use FromHex for string to big
* crypto/bls12381: g1, g2, strict length check for FromBytes
* crypto/bls12381: field_element, comparision changes
* crypto/bls12381: change swu, isogeny constants with hex values
* core/vm: fix point multiplication comments
* core/vm: fix multiexp gas calculation and lookup for g1 and g2
* core/vm: simpler imput length check for multiexp and pairing precompiles
* core/vm: rm empty multiexp result declarations
* crypto/bls12381: remove modulus type definition
* crypto/bls12381: use proper init function
* crypto/bls12381: get rid of new lines at fatal desciprtions
* crypto/bls12-381: fix no-adx assembly multiplication
* crypto/bls12-381: remove old config function
* crypto/bls12381: update multiplication backend
this commit changes mul backend to 6limb eip1962 backend
mul assign operations are dropped
* core/vm/contracts_tests: externalize test vectors for precompiles
* core/vm/contracts_test: externalize failure-cases for precompiles
* core/vm: linting
* go.mod: tiny up sum file
* core/vm: fix goimports linter issues
* crypto/bls12381: build tags for plain ASM or ADX implementation
Co-authored-by: Martin Holst Swende <martin@swende.se>
Co-authored-by: Péter Szilágyi <peterke@gmail.com>
2020-06-03 09:44:32 +03:00
|
|
|
// Copyright 2020 The go-ethereum Authors
|
|
|
|
// This file is part of the go-ethereum library.
|
|
|
|
//
|
|
|
|
// The go-ethereum library is free software: you can redistribute it and/or modify
|
|
|
|
// it under the terms of the GNU Lesser General Public License as published by
|
|
|
|
// the Free Software Foundation, either version 3 of the License, or
|
|
|
|
// (at your option) any later version.
|
|
|
|
//
|
|
|
|
// The go-ethereum library is distributed in the hope that it will be useful,
|
|
|
|
// but WITHOUT ANY WARRANTY; without even the implied warranty of
|
|
|
|
// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
|
|
|
// GNU Lesser General Public License for more details.
|
|
|
|
//
|
|
|
|
// You should have received a copy of the GNU Lesser General Public License
|
|
|
|
// along with the go-ethereum library. If not, see <http://www.gnu.org/licenses/>.
|
|
|
|
|
|
|
|
package bls12381
|
|
|
|
|
|
|
|
import (
|
|
|
|
"errors"
|
|
|
|
"math"
|
|
|
|
"math/big"
|
|
|
|
)
|
|
|
|
|
|
|
|
// PointG2 is type for point in G2.
|
|
|
|
// PointG2 is both used for Affine and Jacobian point representation.
|
|
|
|
// If z is equal to one the point is considered as in affine form.
|
|
|
|
type PointG2 [3]fe2
|
|
|
|
|
|
|
|
// Set copies valeus of one point to another.
|
|
|
|
func (p *PointG2) Set(p2 *PointG2) *PointG2 {
|
|
|
|
p[0].set(&p2[0])
|
|
|
|
p[1].set(&p2[1])
|
|
|
|
p[2].set(&p2[2])
|
|
|
|
return p
|
|
|
|
}
|
|
|
|
|
|
|
|
// Zero returns G2 point in point at infinity representation
|
|
|
|
func (p *PointG2) Zero() *PointG2 {
|
|
|
|
p[0].zero()
|
|
|
|
p[1].one()
|
|
|
|
p[2].zero()
|
|
|
|
return p
|
|
|
|
}
|
|
|
|
|
|
|
|
type tempG2 struct {
|
|
|
|
t [9]*fe2
|
|
|
|
}
|
|
|
|
|
|
|
|
// G2 is struct for G2 group.
|
|
|
|
type G2 struct {
|
|
|
|
f *fp2
|
|
|
|
tempG2
|
|
|
|
}
|
|
|
|
|
|
|
|
// NewG2 constructs a new G2 instance.
|
|
|
|
func NewG2() *G2 {
|
|
|
|
return newG2(nil)
|
|
|
|
}
|
|
|
|
|
|
|
|
func newG2(f *fp2) *G2 {
|
|
|
|
if f == nil {
|
|
|
|
f = newFp2()
|
|
|
|
}
|
|
|
|
t := newTempG2()
|
|
|
|
return &G2{f, t}
|
|
|
|
}
|
|
|
|
|
|
|
|
func newTempG2() tempG2 {
|
|
|
|
t := [9]*fe2{}
|
|
|
|
for i := 0; i < 9; i++ {
|
|
|
|
t[i] = &fe2{}
|
|
|
|
}
|
|
|
|
return tempG2{t}
|
|
|
|
}
|
|
|
|
|
|
|
|
// Q returns group order in big.Int.
|
|
|
|
func (g *G2) Q() *big.Int {
|
|
|
|
return new(big.Int).Set(q)
|
|
|
|
}
|
|
|
|
|
|
|
|
func (g *G2) fromBytesUnchecked(in []byte) (*PointG2, error) {
|
|
|
|
p0, err := g.f.fromBytes(in[:96])
|
|
|
|
if err != nil {
|
|
|
|
return nil, err
|
|
|
|
}
|
|
|
|
p1, err := g.f.fromBytes(in[96:])
|
|
|
|
if err != nil {
|
|
|
|
return nil, err
|
|
|
|
}
|
|
|
|
p2 := new(fe2).one()
|
|
|
|
return &PointG2{*p0, *p1, *p2}, nil
|
|
|
|
}
|
|
|
|
|
|
|
|
// FromBytes constructs a new point given uncompressed byte input.
|
|
|
|
// FromBytes does not take zcash flags into account.
|
|
|
|
// Byte input expected to be larger than 96 bytes.
|
|
|
|
// First 192 bytes should be concatenation of x and y values
|
|
|
|
// Point (0, 0) is considered as infinity.
|
|
|
|
func (g *G2) FromBytes(in []byte) (*PointG2, error) {
|
|
|
|
if len(in) != 192 {
|
|
|
|
return nil, errors.New("input string should be equal or larger than 192")
|
|
|
|
}
|
|
|
|
p0, err := g.f.fromBytes(in[:96])
|
|
|
|
if err != nil {
|
|
|
|
return nil, err
|
|
|
|
}
|
|
|
|
p1, err := g.f.fromBytes(in[96:])
|
|
|
|
if err != nil {
|
|
|
|
return nil, err
|
|
|
|
}
|
|
|
|
// check if given input points to infinity
|
|
|
|
if p0.isZero() && p1.isZero() {
|
|
|
|
return g.Zero(), nil
|
|
|
|
}
|
|
|
|
p2 := new(fe2).one()
|
|
|
|
p := &PointG2{*p0, *p1, *p2}
|
|
|
|
if !g.IsOnCurve(p) {
|
|
|
|
return nil, errors.New("point is not on curve")
|
|
|
|
}
|
|
|
|
return p, nil
|
|
|
|
}
|
|
|
|
|
2023-08-15 20:37:34 +08:00
|
|
|
// DecodePoint given encoded (x, y) coordinates in 256 bytes returns a valid G2 Point.
|
core/vm, crypto/bls12381, params: add bls12-381 elliptic curve precompiles (#21018)
* crypto: add bls12-381 elliptic curve wrapper
* params: add bls12-381 precompile gas parameters
* core/vm: add bls12-381 precompiles
* core/vm: add bls12-381 precompile tests
* go.mod, go.sum: use latest bls12381 lib
* core/vm: move point encode/decode functions to base library
* crypto/bls12381: introduce bls12-381 library init function
* crypto/bls12381: import bls12381 elliptic curve implementation
* go.mod, go.sum: remove bls12-381 library
* remove unsued frobenious coeffs
supress warning for inp that used in asm
* add mappings tests for zero inputs
fix swu g2 minus z inverse constant
* crypto/bls12381: fix typo
* crypto/bls12381: better comments for bls12381 constants
* crypto/bls12381: swu, use single conditional for e2
* crypto/bls12381: utils, delete empty line
* crypto/bls12381: utils, use FromHex for string to big
* crypto/bls12381: g1, g2, strict length check for FromBytes
* crypto/bls12381: field_element, comparision changes
* crypto/bls12381: change swu, isogeny constants with hex values
* core/vm: fix point multiplication comments
* core/vm: fix multiexp gas calculation and lookup for g1 and g2
* core/vm: simpler imput length check for multiexp and pairing precompiles
* core/vm: rm empty multiexp result declarations
* crypto/bls12381: remove modulus type definition
* crypto/bls12381: use proper init function
* crypto/bls12381: get rid of new lines at fatal desciprtions
* crypto/bls12-381: fix no-adx assembly multiplication
* crypto/bls12-381: remove old config function
* crypto/bls12381: update multiplication backend
this commit changes mul backend to 6limb eip1962 backend
mul assign operations are dropped
* core/vm/contracts_tests: externalize test vectors for precompiles
* core/vm/contracts_test: externalize failure-cases for precompiles
* core/vm: linting
* go.mod: tiny up sum file
* core/vm: fix goimports linter issues
* crypto/bls12381: build tags for plain ASM or ADX implementation
Co-authored-by: Martin Holst Swende <martin@swende.se>
Co-authored-by: Péter Szilágyi <peterke@gmail.com>
2020-06-03 09:44:32 +03:00
|
|
|
func (g *G2) DecodePoint(in []byte) (*PointG2, error) {
|
|
|
|
if len(in) != 256 {
|
|
|
|
return nil, errors.New("invalid g2 point length")
|
|
|
|
}
|
|
|
|
pointBytes := make([]byte, 192)
|
|
|
|
x0Bytes, err := decodeFieldElement(in[:64])
|
|
|
|
if err != nil {
|
|
|
|
return nil, err
|
|
|
|
}
|
|
|
|
x1Bytes, err := decodeFieldElement(in[64:128])
|
|
|
|
if err != nil {
|
|
|
|
return nil, err
|
|
|
|
}
|
|
|
|
y0Bytes, err := decodeFieldElement(in[128:192])
|
|
|
|
if err != nil {
|
|
|
|
return nil, err
|
|
|
|
}
|
|
|
|
y1Bytes, err := decodeFieldElement(in[192:])
|
|
|
|
if err != nil {
|
|
|
|
return nil, err
|
|
|
|
}
|
|
|
|
copy(pointBytes[:48], x1Bytes)
|
|
|
|
copy(pointBytes[48:96], x0Bytes)
|
|
|
|
copy(pointBytes[96:144], y1Bytes)
|
|
|
|
copy(pointBytes[144:192], y0Bytes)
|
|
|
|
return g.FromBytes(pointBytes)
|
|
|
|
}
|
|
|
|
|
|
|
|
// ToBytes serializes a point into bytes in uncompressed form,
|
|
|
|
// does not take zcash flags into account,
|
|
|
|
// returns (0, 0) if point is infinity.
|
|
|
|
func (g *G2) ToBytes(p *PointG2) []byte {
|
|
|
|
out := make([]byte, 192)
|
|
|
|
if g.IsZero(p) {
|
|
|
|
return out
|
|
|
|
}
|
|
|
|
g.Affine(p)
|
|
|
|
copy(out[:96], g.f.toBytes(&p[0]))
|
|
|
|
copy(out[96:], g.f.toBytes(&p[1]))
|
|
|
|
return out
|
|
|
|
}
|
|
|
|
|
2020-06-08 09:53:19 +02:00
|
|
|
// EncodePoint encodes a point into 256 bytes.
|
core/vm, crypto/bls12381, params: add bls12-381 elliptic curve precompiles (#21018)
* crypto: add bls12-381 elliptic curve wrapper
* params: add bls12-381 precompile gas parameters
* core/vm: add bls12-381 precompiles
* core/vm: add bls12-381 precompile tests
* go.mod, go.sum: use latest bls12381 lib
* core/vm: move point encode/decode functions to base library
* crypto/bls12381: introduce bls12-381 library init function
* crypto/bls12381: import bls12381 elliptic curve implementation
* go.mod, go.sum: remove bls12-381 library
* remove unsued frobenious coeffs
supress warning for inp that used in asm
* add mappings tests for zero inputs
fix swu g2 minus z inverse constant
* crypto/bls12381: fix typo
* crypto/bls12381: better comments for bls12381 constants
* crypto/bls12381: swu, use single conditional for e2
* crypto/bls12381: utils, delete empty line
* crypto/bls12381: utils, use FromHex for string to big
* crypto/bls12381: g1, g2, strict length check for FromBytes
* crypto/bls12381: field_element, comparision changes
* crypto/bls12381: change swu, isogeny constants with hex values
* core/vm: fix point multiplication comments
* core/vm: fix multiexp gas calculation and lookup for g1 and g2
* core/vm: simpler imput length check for multiexp and pairing precompiles
* core/vm: rm empty multiexp result declarations
* crypto/bls12381: remove modulus type definition
* crypto/bls12381: use proper init function
* crypto/bls12381: get rid of new lines at fatal desciprtions
* crypto/bls12-381: fix no-adx assembly multiplication
* crypto/bls12-381: remove old config function
* crypto/bls12381: update multiplication backend
this commit changes mul backend to 6limb eip1962 backend
mul assign operations are dropped
* core/vm/contracts_tests: externalize test vectors for precompiles
* core/vm/contracts_test: externalize failure-cases for precompiles
* core/vm: linting
* go.mod: tiny up sum file
* core/vm: fix goimports linter issues
* crypto/bls12381: build tags for plain ASM or ADX implementation
Co-authored-by: Martin Holst Swende <martin@swende.se>
Co-authored-by: Péter Szilágyi <peterke@gmail.com>
2020-06-03 09:44:32 +03:00
|
|
|
func (g *G2) EncodePoint(p *PointG2) []byte {
|
|
|
|
// outRaw is 96 bytes
|
|
|
|
outRaw := g.ToBytes(p)
|
|
|
|
out := make([]byte, 256)
|
|
|
|
// encode x
|
|
|
|
copy(out[16:16+48], outRaw[48:96])
|
|
|
|
copy(out[80:80+48], outRaw[:48])
|
|
|
|
// encode y
|
|
|
|
copy(out[144:144+48], outRaw[144:])
|
|
|
|
copy(out[208:208+48], outRaw[96:144])
|
|
|
|
return out
|
|
|
|
}
|
|
|
|
|
|
|
|
// New creates a new G2 Point which is equal to zero in other words point at infinity.
|
|
|
|
func (g *G2) New() *PointG2 {
|
|
|
|
return new(PointG2).Zero()
|
|
|
|
}
|
|
|
|
|
|
|
|
// Zero returns a new G2 Point which is equal to point at infinity.
|
|
|
|
func (g *G2) Zero() *PointG2 {
|
|
|
|
return new(PointG2).Zero()
|
|
|
|
}
|
|
|
|
|
|
|
|
// One returns a new G2 Point which is equal to generator point.
|
|
|
|
func (g *G2) One() *PointG2 {
|
|
|
|
p := &PointG2{}
|
|
|
|
return p.Set(&g2One)
|
|
|
|
}
|
|
|
|
|
|
|
|
// IsZero returns true if given point is equal to zero.
|
|
|
|
func (g *G2) IsZero(p *PointG2) bool {
|
|
|
|
return p[2].isZero()
|
|
|
|
}
|
|
|
|
|
|
|
|
// Equal checks if given two G2 point is equal in their affine form.
|
|
|
|
func (g *G2) Equal(p1, p2 *PointG2) bool {
|
|
|
|
if g.IsZero(p1) {
|
|
|
|
return g.IsZero(p2)
|
|
|
|
}
|
|
|
|
if g.IsZero(p2) {
|
|
|
|
return g.IsZero(p1)
|
|
|
|
}
|
|
|
|
t := g.t
|
|
|
|
g.f.square(t[0], &p1[2])
|
|
|
|
g.f.square(t[1], &p2[2])
|
|
|
|
g.f.mul(t[2], t[0], &p2[0])
|
|
|
|
g.f.mul(t[3], t[1], &p1[0])
|
|
|
|
g.f.mul(t[0], t[0], &p1[2])
|
|
|
|
g.f.mul(t[1], t[1], &p2[2])
|
|
|
|
g.f.mul(t[1], t[1], &p1[1])
|
|
|
|
g.f.mul(t[0], t[0], &p2[1])
|
|
|
|
return t[0].equal(t[1]) && t[2].equal(t[3])
|
|
|
|
}
|
|
|
|
|
|
|
|
// InCorrectSubgroup checks whether given point is in correct subgroup.
|
|
|
|
func (g *G2) InCorrectSubgroup(p *PointG2) bool {
|
|
|
|
tmp := &PointG2{}
|
|
|
|
g.MulScalar(tmp, p, q)
|
|
|
|
return g.IsZero(tmp)
|
|
|
|
}
|
|
|
|
|
|
|
|
// IsOnCurve checks a G2 point is on curve.
|
|
|
|
func (g *G2) IsOnCurve(p *PointG2) bool {
|
|
|
|
if g.IsZero(p) {
|
|
|
|
return true
|
|
|
|
}
|
|
|
|
t := g.t
|
|
|
|
g.f.square(t[0], &p[1])
|
|
|
|
g.f.square(t[1], &p[0])
|
|
|
|
g.f.mul(t[1], t[1], &p[0])
|
|
|
|
g.f.square(t[2], &p[2])
|
|
|
|
g.f.square(t[3], t[2])
|
|
|
|
g.f.mul(t[2], t[2], t[3])
|
|
|
|
g.f.mul(t[2], b2, t[2])
|
|
|
|
g.f.add(t[1], t[1], t[2])
|
|
|
|
return t[0].equal(t[1])
|
|
|
|
}
|
|
|
|
|
|
|
|
// IsAffine checks a G2 point whether it is in affine form.
|
|
|
|
func (g *G2) IsAffine(p *PointG2) bool {
|
|
|
|
return p[2].isOne()
|
|
|
|
}
|
|
|
|
|
|
|
|
// Affine calculates affine form of given G2 point.
|
|
|
|
func (g *G2) Affine(p *PointG2) *PointG2 {
|
|
|
|
if g.IsZero(p) {
|
|
|
|
return p
|
|
|
|
}
|
|
|
|
if !g.IsAffine(p) {
|
|
|
|
t := g.t
|
|
|
|
g.f.inverse(t[0], &p[2])
|
|
|
|
g.f.square(t[1], t[0])
|
|
|
|
g.f.mul(&p[0], &p[0], t[1])
|
|
|
|
g.f.mul(t[0], t[0], t[1])
|
|
|
|
g.f.mul(&p[1], &p[1], t[0])
|
|
|
|
p[2].one()
|
|
|
|
}
|
|
|
|
return p
|
|
|
|
}
|
|
|
|
|
|
|
|
// Add adds two G2 points p1, p2 and assigns the result to point at first argument.
|
|
|
|
func (g *G2) Add(r, p1, p2 *PointG2) *PointG2 {
|
2022-11-03 00:48:13 -07:00
|
|
|
// http://www.hyperelliptic.org/EFD/g1p/auto-shortw-jacobian-0.html#addition-add-2007-bl
|
core/vm, crypto/bls12381, params: add bls12-381 elliptic curve precompiles (#21018)
* crypto: add bls12-381 elliptic curve wrapper
* params: add bls12-381 precompile gas parameters
* core/vm: add bls12-381 precompiles
* core/vm: add bls12-381 precompile tests
* go.mod, go.sum: use latest bls12381 lib
* core/vm: move point encode/decode functions to base library
* crypto/bls12381: introduce bls12-381 library init function
* crypto/bls12381: import bls12381 elliptic curve implementation
* go.mod, go.sum: remove bls12-381 library
* remove unsued frobenious coeffs
supress warning for inp that used in asm
* add mappings tests for zero inputs
fix swu g2 minus z inverse constant
* crypto/bls12381: fix typo
* crypto/bls12381: better comments for bls12381 constants
* crypto/bls12381: swu, use single conditional for e2
* crypto/bls12381: utils, delete empty line
* crypto/bls12381: utils, use FromHex for string to big
* crypto/bls12381: g1, g2, strict length check for FromBytes
* crypto/bls12381: field_element, comparision changes
* crypto/bls12381: change swu, isogeny constants with hex values
* core/vm: fix point multiplication comments
* core/vm: fix multiexp gas calculation and lookup for g1 and g2
* core/vm: simpler imput length check for multiexp and pairing precompiles
* core/vm: rm empty multiexp result declarations
* crypto/bls12381: remove modulus type definition
* crypto/bls12381: use proper init function
* crypto/bls12381: get rid of new lines at fatal desciprtions
* crypto/bls12-381: fix no-adx assembly multiplication
* crypto/bls12-381: remove old config function
* crypto/bls12381: update multiplication backend
this commit changes mul backend to 6limb eip1962 backend
mul assign operations are dropped
* core/vm/contracts_tests: externalize test vectors for precompiles
* core/vm/contracts_test: externalize failure-cases for precompiles
* core/vm: linting
* go.mod: tiny up sum file
* core/vm: fix goimports linter issues
* crypto/bls12381: build tags for plain ASM or ADX implementation
Co-authored-by: Martin Holst Swende <martin@swende.se>
Co-authored-by: Péter Szilágyi <peterke@gmail.com>
2020-06-03 09:44:32 +03:00
|
|
|
if g.IsZero(p1) {
|
|
|
|
return r.Set(p2)
|
|
|
|
}
|
|
|
|
if g.IsZero(p2) {
|
|
|
|
return r.Set(p1)
|
|
|
|
}
|
|
|
|
t := g.t
|
|
|
|
g.f.square(t[7], &p1[2])
|
|
|
|
g.f.mul(t[1], &p2[0], t[7])
|
|
|
|
g.f.mul(t[2], &p1[2], t[7])
|
|
|
|
g.f.mul(t[0], &p2[1], t[2])
|
|
|
|
g.f.square(t[8], &p2[2])
|
|
|
|
g.f.mul(t[3], &p1[0], t[8])
|
|
|
|
g.f.mul(t[4], &p2[2], t[8])
|
|
|
|
g.f.mul(t[2], &p1[1], t[4])
|
|
|
|
if t[1].equal(t[3]) {
|
|
|
|
if t[0].equal(t[2]) {
|
|
|
|
return g.Double(r, p1)
|
|
|
|
}
|
2020-11-25 09:24:50 +01:00
|
|
|
return r.Zero()
|
core/vm, crypto/bls12381, params: add bls12-381 elliptic curve precompiles (#21018)
* crypto: add bls12-381 elliptic curve wrapper
* params: add bls12-381 precompile gas parameters
* core/vm: add bls12-381 precompiles
* core/vm: add bls12-381 precompile tests
* go.mod, go.sum: use latest bls12381 lib
* core/vm: move point encode/decode functions to base library
* crypto/bls12381: introduce bls12-381 library init function
* crypto/bls12381: import bls12381 elliptic curve implementation
* go.mod, go.sum: remove bls12-381 library
* remove unsued frobenious coeffs
supress warning for inp that used in asm
* add mappings tests for zero inputs
fix swu g2 minus z inverse constant
* crypto/bls12381: fix typo
* crypto/bls12381: better comments for bls12381 constants
* crypto/bls12381: swu, use single conditional for e2
* crypto/bls12381: utils, delete empty line
* crypto/bls12381: utils, use FromHex for string to big
* crypto/bls12381: g1, g2, strict length check for FromBytes
* crypto/bls12381: field_element, comparision changes
* crypto/bls12381: change swu, isogeny constants with hex values
* core/vm: fix point multiplication comments
* core/vm: fix multiexp gas calculation and lookup for g1 and g2
* core/vm: simpler imput length check for multiexp and pairing precompiles
* core/vm: rm empty multiexp result declarations
* crypto/bls12381: remove modulus type definition
* crypto/bls12381: use proper init function
* crypto/bls12381: get rid of new lines at fatal desciprtions
* crypto/bls12-381: fix no-adx assembly multiplication
* crypto/bls12-381: remove old config function
* crypto/bls12381: update multiplication backend
this commit changes mul backend to 6limb eip1962 backend
mul assign operations are dropped
* core/vm/contracts_tests: externalize test vectors for precompiles
* core/vm/contracts_test: externalize failure-cases for precompiles
* core/vm: linting
* go.mod: tiny up sum file
* core/vm: fix goimports linter issues
* crypto/bls12381: build tags for plain ASM or ADX implementation
Co-authored-by: Martin Holst Swende <martin@swende.se>
Co-authored-by: Péter Szilágyi <peterke@gmail.com>
2020-06-03 09:44:32 +03:00
|
|
|
}
|
|
|
|
g.f.sub(t[1], t[1], t[3])
|
|
|
|
g.f.double(t[4], t[1])
|
|
|
|
g.f.square(t[4], t[4])
|
|
|
|
g.f.mul(t[5], t[1], t[4])
|
|
|
|
g.f.sub(t[0], t[0], t[2])
|
|
|
|
g.f.double(t[0], t[0])
|
|
|
|
g.f.square(t[6], t[0])
|
|
|
|
g.f.sub(t[6], t[6], t[5])
|
|
|
|
g.f.mul(t[3], t[3], t[4])
|
|
|
|
g.f.double(t[4], t[3])
|
|
|
|
g.f.sub(&r[0], t[6], t[4])
|
|
|
|
g.f.sub(t[4], t[3], &r[0])
|
|
|
|
g.f.mul(t[6], t[2], t[5])
|
|
|
|
g.f.double(t[6], t[6])
|
|
|
|
g.f.mul(t[0], t[0], t[4])
|
|
|
|
g.f.sub(&r[1], t[0], t[6])
|
|
|
|
g.f.add(t[0], &p1[2], &p2[2])
|
|
|
|
g.f.square(t[0], t[0])
|
|
|
|
g.f.sub(t[0], t[0], t[7])
|
|
|
|
g.f.sub(t[0], t[0], t[8])
|
|
|
|
g.f.mul(&r[2], t[0], t[1])
|
|
|
|
return r
|
|
|
|
}
|
|
|
|
|
|
|
|
// Double doubles a G2 point p and assigns the result to the point at first argument.
|
|
|
|
func (g *G2) Double(r, p *PointG2) *PointG2 {
|
2022-11-03 00:48:13 -07:00
|
|
|
// http://www.hyperelliptic.org/EFD/g1p/auto-shortw-jacobian-0.html#doubling-dbl-2009-l
|
core/vm, crypto/bls12381, params: add bls12-381 elliptic curve precompiles (#21018)
* crypto: add bls12-381 elliptic curve wrapper
* params: add bls12-381 precompile gas parameters
* core/vm: add bls12-381 precompiles
* core/vm: add bls12-381 precompile tests
* go.mod, go.sum: use latest bls12381 lib
* core/vm: move point encode/decode functions to base library
* crypto/bls12381: introduce bls12-381 library init function
* crypto/bls12381: import bls12381 elliptic curve implementation
* go.mod, go.sum: remove bls12-381 library
* remove unsued frobenious coeffs
supress warning for inp that used in asm
* add mappings tests for zero inputs
fix swu g2 minus z inverse constant
* crypto/bls12381: fix typo
* crypto/bls12381: better comments for bls12381 constants
* crypto/bls12381: swu, use single conditional for e2
* crypto/bls12381: utils, delete empty line
* crypto/bls12381: utils, use FromHex for string to big
* crypto/bls12381: g1, g2, strict length check for FromBytes
* crypto/bls12381: field_element, comparision changes
* crypto/bls12381: change swu, isogeny constants with hex values
* core/vm: fix point multiplication comments
* core/vm: fix multiexp gas calculation and lookup for g1 and g2
* core/vm: simpler imput length check for multiexp and pairing precompiles
* core/vm: rm empty multiexp result declarations
* crypto/bls12381: remove modulus type definition
* crypto/bls12381: use proper init function
* crypto/bls12381: get rid of new lines at fatal desciprtions
* crypto/bls12-381: fix no-adx assembly multiplication
* crypto/bls12-381: remove old config function
* crypto/bls12381: update multiplication backend
this commit changes mul backend to 6limb eip1962 backend
mul assign operations are dropped
* core/vm/contracts_tests: externalize test vectors for precompiles
* core/vm/contracts_test: externalize failure-cases for precompiles
* core/vm: linting
* go.mod: tiny up sum file
* core/vm: fix goimports linter issues
* crypto/bls12381: build tags for plain ASM or ADX implementation
Co-authored-by: Martin Holst Swende <martin@swende.se>
Co-authored-by: Péter Szilágyi <peterke@gmail.com>
2020-06-03 09:44:32 +03:00
|
|
|
if g.IsZero(p) {
|
|
|
|
return r.Set(p)
|
|
|
|
}
|
|
|
|
t := g.t
|
|
|
|
g.f.square(t[0], &p[0])
|
|
|
|
g.f.square(t[1], &p[1])
|
|
|
|
g.f.square(t[2], t[1])
|
|
|
|
g.f.add(t[1], &p[0], t[1])
|
|
|
|
g.f.square(t[1], t[1])
|
|
|
|
g.f.sub(t[1], t[1], t[0])
|
|
|
|
g.f.sub(t[1], t[1], t[2])
|
|
|
|
g.f.double(t[1], t[1])
|
|
|
|
g.f.double(t[3], t[0])
|
|
|
|
g.f.add(t[0], t[3], t[0])
|
|
|
|
g.f.square(t[4], t[0])
|
|
|
|
g.f.double(t[3], t[1])
|
|
|
|
g.f.sub(&r[0], t[4], t[3])
|
|
|
|
g.f.sub(t[1], t[1], &r[0])
|
|
|
|
g.f.double(t[2], t[2])
|
|
|
|
g.f.double(t[2], t[2])
|
|
|
|
g.f.double(t[2], t[2])
|
|
|
|
g.f.mul(t[0], t[0], t[1])
|
|
|
|
g.f.sub(t[1], t[0], t[2])
|
|
|
|
g.f.mul(t[0], &p[1], &p[2])
|
|
|
|
r[1].set(t[1])
|
|
|
|
g.f.double(&r[2], t[0])
|
|
|
|
return r
|
|
|
|
}
|
|
|
|
|
|
|
|
// Neg negates a G2 point p and assigns the result to the point at first argument.
|
|
|
|
func (g *G2) Neg(r, p *PointG2) *PointG2 {
|
|
|
|
r[0].set(&p[0])
|
|
|
|
g.f.neg(&r[1], &p[1])
|
|
|
|
r[2].set(&p[2])
|
|
|
|
return r
|
|
|
|
}
|
|
|
|
|
|
|
|
// Sub subtracts two G2 points p1, p2 and assigns the result to point at first argument.
|
|
|
|
func (g *G2) Sub(c, a, b *PointG2) *PointG2 {
|
|
|
|
d := &PointG2{}
|
|
|
|
g.Neg(d, b)
|
|
|
|
g.Add(c, a, d)
|
|
|
|
return c
|
|
|
|
}
|
|
|
|
|
|
|
|
// MulScalar multiplies a point by given scalar value in big.Int and assigns the result to point at first argument.
|
|
|
|
func (g *G2) MulScalar(c, p *PointG2, e *big.Int) *PointG2 {
|
|
|
|
q, n := &PointG2{}, &PointG2{}
|
|
|
|
n.Set(p)
|
|
|
|
l := e.BitLen()
|
|
|
|
for i := 0; i < l; i++ {
|
|
|
|
if e.Bit(i) == 1 {
|
|
|
|
g.Add(q, q, n)
|
|
|
|
}
|
|
|
|
g.Double(n, n)
|
|
|
|
}
|
|
|
|
return c.Set(q)
|
|
|
|
}
|
|
|
|
|
|
|
|
// ClearCofactor maps given a G2 point to correct subgroup
|
|
|
|
func (g *G2) ClearCofactor(p *PointG2) {
|
|
|
|
g.MulScalar(p, p, cofactorEFFG2)
|
|
|
|
}
|
|
|
|
|
|
|
|
// MultiExp calculates multi exponentiation. Given pairs of G2 point and scalar values
|
|
|
|
// (P_0, e_0), (P_1, e_1), ... (P_n, e_n) calculates r = e_0 * P_0 + e_1 * P_1 + ... + e_n * P_n
|
|
|
|
// Length of points and scalars are expected to be equal, otherwise an error is returned.
|
|
|
|
// Result is assigned to point at first argument.
|
|
|
|
func (g *G2) MultiExp(r *PointG2, points []*PointG2, powers []*big.Int) (*PointG2, error) {
|
|
|
|
if len(points) != len(powers) {
|
|
|
|
return nil, errors.New("point and scalar vectors should be in same length")
|
|
|
|
}
|
|
|
|
var c uint32 = 3
|
|
|
|
if len(powers) >= 32 {
|
|
|
|
c = uint32(math.Ceil(math.Log10(float64(len(powers)))))
|
|
|
|
}
|
|
|
|
bucketSize, numBits := (1<<c)-1, uint32(g.Q().BitLen())
|
|
|
|
windows := make([]*PointG2, numBits/c+1)
|
|
|
|
bucket := make([]*PointG2, bucketSize)
|
|
|
|
acc, sum := g.New(), g.New()
|
|
|
|
for i := 0; i < bucketSize; i++ {
|
|
|
|
bucket[i] = g.New()
|
|
|
|
}
|
|
|
|
mask := (uint64(1) << c) - 1
|
|
|
|
j := 0
|
|
|
|
var cur uint32
|
|
|
|
for cur <= numBits {
|
|
|
|
acc.Zero()
|
|
|
|
bucket = make([]*PointG2, (1<<c)-1)
|
|
|
|
for i := 0; i < len(bucket); i++ {
|
|
|
|
bucket[i] = g.New()
|
|
|
|
}
|
|
|
|
for i := 0; i < len(powers); i++ {
|
|
|
|
s0 := powers[i].Uint64()
|
|
|
|
index := uint(s0 & mask)
|
|
|
|
if index != 0 {
|
|
|
|
g.Add(bucket[index-1], bucket[index-1], points[i])
|
|
|
|
}
|
|
|
|
powers[i] = new(big.Int).Rsh(powers[i], uint(c))
|
|
|
|
}
|
|
|
|
sum.Zero()
|
|
|
|
for i := len(bucket) - 1; i >= 0; i-- {
|
|
|
|
g.Add(sum, sum, bucket[i])
|
|
|
|
g.Add(acc, acc, sum)
|
|
|
|
}
|
|
|
|
windows[j] = g.New()
|
|
|
|
windows[j].Set(acc)
|
|
|
|
j++
|
|
|
|
cur += c
|
|
|
|
}
|
|
|
|
acc.Zero()
|
|
|
|
for i := len(windows) - 1; i >= 0; i-- {
|
|
|
|
for j := uint32(0); j < c; j++ {
|
|
|
|
g.Double(acc, acc)
|
|
|
|
}
|
|
|
|
g.Add(acc, acc, windows[i])
|
|
|
|
}
|
|
|
|
return r.Set(acc), nil
|
|
|
|
}
|
|
|
|
|
|
|
|
// MapToCurve given a byte slice returns a valid G2 point.
|
|
|
|
// This mapping function implements the Simplified Shallue-van de Woestijne-Ulas method.
|
|
|
|
// https://tools.ietf.org/html/draft-irtf-cfrg-hash-to-curve-05#section-6.6.2
|
|
|
|
// Input byte slice should be a valid field element, otherwise an error is returned.
|
|
|
|
func (g *G2) MapToCurve(in []byte) (*PointG2, error) {
|
|
|
|
fp2 := g.f
|
|
|
|
u, err := fp2.fromBytes(in)
|
|
|
|
if err != nil {
|
|
|
|
return nil, err
|
|
|
|
}
|
|
|
|
x, y := swuMapG2(fp2, u)
|
|
|
|
isogenyMapG2(fp2, x, y)
|
|
|
|
z := new(fe2).one()
|
|
|
|
q := &PointG2{*x, *y, *z}
|
|
|
|
g.ClearCofactor(q)
|
|
|
|
return g.Affine(q), nil
|
|
|
|
}
|