mirror of
https://source.quilibrium.com/quilibrium/ceremonyclient.git
synced 2024-12-27 00:55:17 +00:00
617 lines
9.3 KiB
Go
617 lines
9.3 KiB
Go
|
/*
|
||
|
* Copyright (c) 2012-2020 MIRACL UK Ltd.
|
||
|
*
|
||
|
* This file is part of MIRACL Core
|
||
|
* (see https://github.com/miracl/core).
|
||
|
*
|
||
|
* Licensed under the Apache License, Version 2.0 (the "License");
|
||
|
* you may not use this file except in compliance with the License.
|
||
|
* You may obtain a copy of the License at
|
||
|
*
|
||
|
* http://www.apache.org/licenses/LICENSE-2.0
|
||
|
*
|
||
|
* Unless required by applicable law or agreed to in writing, software
|
||
|
* distributed under the License is distributed on an "AS IS" BASIS,
|
||
|
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
||
|
* See the License for the specific language governing permissions and
|
||
|
* limitations under the License.
|
||
|
*/
|
||
|
|
||
|
/* Finite Field arithmetic Fp^16 functions */
|
||
|
|
||
|
/* FP4 elements are of the form a+ib, where i is sqrt(-1+sqrt(-1)) */
|
||
|
|
||
|
package bls48581
|
||
|
|
||
|
//import "fmt"
|
||
|
|
||
|
type FP16 struct {
|
||
|
a *FP8
|
||
|
b *FP8
|
||
|
}
|
||
|
|
||
|
func NewFP16() *FP16 {
|
||
|
F := new(FP16)
|
||
|
F.a = NewFP8()
|
||
|
F.b = NewFP8()
|
||
|
return F
|
||
|
}
|
||
|
|
||
|
/* Constructors */
|
||
|
func NewFP16int(a int) *FP16 {
|
||
|
F := new(FP16)
|
||
|
F.a = NewFP8int(a)
|
||
|
F.b = NewFP8()
|
||
|
return F
|
||
|
}
|
||
|
|
||
|
func NewFP16copy(x *FP16) *FP16 {
|
||
|
F := new(FP16)
|
||
|
F.a = NewFP8copy(x.a)
|
||
|
F.b = NewFP8copy(x.b)
|
||
|
return F
|
||
|
}
|
||
|
|
||
|
func NewFP16fp8s(c *FP8, d *FP8) *FP16 {
|
||
|
F := new(FP16)
|
||
|
F.a = NewFP8copy(c)
|
||
|
F.b = NewFP8copy(d)
|
||
|
return F
|
||
|
}
|
||
|
|
||
|
func NewFP16fp8(c *FP8) *FP16 {
|
||
|
F := new(FP16)
|
||
|
F.a = NewFP8copy(c)
|
||
|
F.b = NewFP8()
|
||
|
return F
|
||
|
}
|
||
|
|
||
|
/* reduce all components of this mod Modulus */
|
||
|
func (F *FP16) reduce() {
|
||
|
F.a.reduce()
|
||
|
F.b.reduce()
|
||
|
}
|
||
|
|
||
|
/* normalise all components of this mod Modulus */
|
||
|
func (F *FP16) norm() {
|
||
|
F.a.norm()
|
||
|
F.b.norm()
|
||
|
}
|
||
|
|
||
|
/* test this==0 ? */
|
||
|
func (F *FP16) iszilch() bool {
|
||
|
return F.a.iszilch() && F.b.iszilch()
|
||
|
}
|
||
|
|
||
|
func (F *FP16) ToBytes(bf []byte) {
|
||
|
var t [8 * int(MODBYTES)]byte
|
||
|
MB := 8 * int(MODBYTES)
|
||
|
F.b.ToBytes(t[:])
|
||
|
for i := 0; i < MB; i++ {
|
||
|
bf[i] = t[i]
|
||
|
}
|
||
|
F.a.ToBytes(t[:])
|
||
|
for i := 0; i < MB; i++ {
|
||
|
bf[i+MB] = t[i]
|
||
|
}
|
||
|
}
|
||
|
|
||
|
func FP16_fromBytes(bf []byte) *FP16 {
|
||
|
var t [8 * int(MODBYTES)]byte
|
||
|
MB := 8 * int(MODBYTES)
|
||
|
for i := 0; i < MB; i++ {
|
||
|
t[i] = bf[i]
|
||
|
}
|
||
|
tb := FP8_fromBytes(t[:])
|
||
|
for i := 0; i < MB; i++ {
|
||
|
t[i] = bf[i+MB]
|
||
|
}
|
||
|
ta := FP8_fromBytes(t[:])
|
||
|
return NewFP16fp8s(ta, tb)
|
||
|
}
|
||
|
|
||
|
/* Conditional move */
|
||
|
func (F *FP16) cmove(g *FP16, d int) {
|
||
|
F.a.cmove(g.a, d)
|
||
|
F.b.cmove(g.b, d)
|
||
|
}
|
||
|
|
||
|
/* test this==1 ? */
|
||
|
func (F *FP16) isunity() bool {
|
||
|
one := NewFP8int(1)
|
||
|
return F.a.Equals(one) && F.b.iszilch()
|
||
|
}
|
||
|
|
||
|
/* test is w real? That is in a+ib test b is zero */
|
||
|
func (F *FP16) isreal() bool {
|
||
|
return F.b.iszilch()
|
||
|
}
|
||
|
|
||
|
/* extract real part a */
|
||
|
func (F *FP16) real() *FP8 {
|
||
|
return F.a
|
||
|
}
|
||
|
|
||
|
func (F *FP16) geta() *FP8 {
|
||
|
return F.a
|
||
|
}
|
||
|
|
||
|
/* extract imaginary part b */
|
||
|
func (F *FP16) getb() *FP8 {
|
||
|
return F.b
|
||
|
}
|
||
|
|
||
|
/* test this=x? */
|
||
|
func (F *FP16) Equals(x *FP16) bool {
|
||
|
return (F.a.Equals(x.a) && F.b.Equals(x.b))
|
||
|
}
|
||
|
|
||
|
/* copy this=x */
|
||
|
func (F *FP16) copy(x *FP16) {
|
||
|
F.a.copy(x.a)
|
||
|
F.b.copy(x.b)
|
||
|
}
|
||
|
|
||
|
/* set this=0 */
|
||
|
func (F *FP16) zero() {
|
||
|
F.a.zero()
|
||
|
F.b.zero()
|
||
|
}
|
||
|
|
||
|
/* set this=1 */
|
||
|
func (F *FP16) one() {
|
||
|
F.a.one()
|
||
|
F.b.zero()
|
||
|
}
|
||
|
|
||
|
/* set this=-this */
|
||
|
func (F *FP16) neg() {
|
||
|
F.norm()
|
||
|
m := NewFP8copy(F.a)
|
||
|
t := NewFP8()
|
||
|
m.add(F.b)
|
||
|
m.neg()
|
||
|
t.copy(m)
|
||
|
t.add(F.b)
|
||
|
F.b.copy(m)
|
||
|
F.b.add(F.a)
|
||
|
F.a.copy(t)
|
||
|
F.norm()
|
||
|
}
|
||
|
|
||
|
/* this=conjugate(this) */
|
||
|
func (F *FP16) conj() {
|
||
|
F.b.neg()
|
||
|
F.norm()
|
||
|
}
|
||
|
|
||
|
/* this=-conjugate(this) */
|
||
|
func (F *FP16) nconj() {
|
||
|
F.a.neg()
|
||
|
F.norm()
|
||
|
}
|
||
|
|
||
|
/* this+=x */
|
||
|
func (F *FP16) add(x *FP16) {
|
||
|
F.a.add(x.a)
|
||
|
F.b.add(x.b)
|
||
|
}
|
||
|
|
||
|
/* this-=x */
|
||
|
func (F *FP16) sub(x *FP16) {
|
||
|
m := NewFP16copy(x)
|
||
|
m.neg()
|
||
|
F.add(m)
|
||
|
}
|
||
|
|
||
|
/* this-=x */
|
||
|
func (F *FP16) rsub(x *FP16) {
|
||
|
F.neg()
|
||
|
F.add(x)
|
||
|
}
|
||
|
|
||
|
/* this*=s where s is FP8 */
|
||
|
func (F *FP16) pmul(s *FP8) {
|
||
|
F.a.mul(s)
|
||
|
F.b.mul(s)
|
||
|
}
|
||
|
|
||
|
/* this*=s where s is FP2 */
|
||
|
func (F *FP16) qmul(s *FP2) {
|
||
|
F.a.qmul(s)
|
||
|
F.b.qmul(s)
|
||
|
}
|
||
|
|
||
|
/* this*=s where s is FP */
|
||
|
func (F *FP16) tmul(s *FP) {
|
||
|
F.a.tmul(s)
|
||
|
F.b.tmul(s)
|
||
|
}
|
||
|
|
||
|
/* this*=c where c is int */
|
||
|
func (F *FP16) imul(c int) {
|
||
|
F.a.imul(c)
|
||
|
F.b.imul(c)
|
||
|
}
|
||
|
|
||
|
/* this*=this */
|
||
|
func (F *FP16) sqr() {
|
||
|
t1 := NewFP8copy(F.a)
|
||
|
t2 := NewFP8copy(F.b)
|
||
|
t3 := NewFP8copy(F.a)
|
||
|
|
||
|
t3.mul(F.b)
|
||
|
t1.add(F.b)
|
||
|
t2.times_i()
|
||
|
|
||
|
t2.add(F.a)
|
||
|
|
||
|
t1.norm()
|
||
|
t2.norm()
|
||
|
|
||
|
F.a.copy(t1)
|
||
|
F.a.mul(t2)
|
||
|
|
||
|
t2.copy(t3)
|
||
|
t2.times_i()
|
||
|
t2.add(t3)
|
||
|
t2.norm()
|
||
|
t2.neg()
|
||
|
F.a.add(t2)
|
||
|
|
||
|
F.b.copy(t3)
|
||
|
F.b.add(t3)
|
||
|
|
||
|
F.norm()
|
||
|
}
|
||
|
|
||
|
/* this*=y */
|
||
|
func (F *FP16) mul(y *FP16) {
|
||
|
t1 := NewFP8copy(F.a)
|
||
|
t2 := NewFP8copy(F.b)
|
||
|
t3 := NewFP8()
|
||
|
t4 := NewFP8copy(F.b)
|
||
|
|
||
|
t1.mul(y.a)
|
||
|
t2.mul(y.b)
|
||
|
t3.copy(y.b)
|
||
|
t3.add(y.a)
|
||
|
t4.add(F.a)
|
||
|
|
||
|
t3.norm()
|
||
|
t4.norm()
|
||
|
|
||
|
t4.mul(t3)
|
||
|
|
||
|
t3.copy(t1)
|
||
|
t3.neg()
|
||
|
t4.add(t3)
|
||
|
t4.norm()
|
||
|
|
||
|
t3.copy(t2)
|
||
|
t3.neg()
|
||
|
F.b.copy(t4)
|
||
|
F.b.add(t3)
|
||
|
|
||
|
t2.times_i()
|
||
|
F.a.copy(t2)
|
||
|
F.a.add(t1)
|
||
|
|
||
|
F.norm()
|
||
|
}
|
||
|
|
||
|
/* convert this to hex string */
|
||
|
func (F *FP16) toString() string {
|
||
|
return ("[" + F.a.toString() + "," + F.b.toString() + "]")
|
||
|
}
|
||
|
|
||
|
/* this=1/this */
|
||
|
func (F *FP16) inverse() {
|
||
|
t1 := NewFP8copy(F.a)
|
||
|
t2 := NewFP8copy(F.b)
|
||
|
|
||
|
t1.sqr()
|
||
|
t2.sqr()
|
||
|
t2.times_i()
|
||
|
t2.norm()
|
||
|
t1.sub(t2)
|
||
|
t1.norm()
|
||
|
|
||
|
t1.inverse(nil)
|
||
|
|
||
|
F.a.mul(t1)
|
||
|
t1.neg()
|
||
|
t1.norm()
|
||
|
F.b.mul(t1)
|
||
|
}
|
||
|
|
||
|
/* this*=i where i = sqrt(sqrt(-1+sqrt(-1))) */
|
||
|
func (F *FP16) times_i() {
|
||
|
s := NewFP8copy(F.b)
|
||
|
t := NewFP8copy(F.a)
|
||
|
s.times_i()
|
||
|
F.a.copy(s)
|
||
|
F.b.copy(t)
|
||
|
F.norm()
|
||
|
}
|
||
|
|
||
|
func (F *FP16) times_i2() {
|
||
|
F.a.times_i()
|
||
|
F.b.times_i()
|
||
|
}
|
||
|
|
||
|
func (F *FP16) times_i4() {
|
||
|
F.a.times_i2()
|
||
|
F.b.times_i2()
|
||
|
}
|
||
|
|
||
|
/* this=this^p using Frobenius */
|
||
|
func (F *FP16) frob(f *FP2) {
|
||
|
ff := NewFP2copy(f)
|
||
|
ff.sqr()
|
||
|
ff.norm()
|
||
|
|
||
|
F.a.frob(ff)
|
||
|
F.b.frob(ff)
|
||
|
F.b.qmul(f)
|
||
|
F.b.times_i()
|
||
|
|
||
|
}
|
||
|
|
||
|
/* this=this^e */
|
||
|
func (F *FP16) pow(e *BIG) *FP16 {
|
||
|
w := NewFP16copy(F)
|
||
|
w.norm()
|
||
|
z := NewBIGcopy(e)
|
||
|
r := NewFP16int(1)
|
||
|
z.norm()
|
||
|
for true {
|
||
|
bt := z.parity()
|
||
|
z.fshr(1)
|
||
|
if bt == 1 {
|
||
|
r.mul(w)
|
||
|
}
|
||
|
if z.iszilch() {
|
||
|
break
|
||
|
}
|
||
|
w.sqr()
|
||
|
}
|
||
|
r.reduce()
|
||
|
return r
|
||
|
}
|
||
|
|
||
|
/* XTR xtr_a function */
|
||
|
/*
|
||
|
func (F *FP16) xtr_A(w *FP16, y *FP16, z *FP16) {
|
||
|
r := NewFP16copy(w)
|
||
|
t := NewFP16copy(w)
|
||
|
r.sub(y)
|
||
|
r.norm()
|
||
|
r.pmul(F.a)
|
||
|
t.add(y)
|
||
|
t.norm()
|
||
|
t.pmul(F.b)
|
||
|
t.times_i()
|
||
|
|
||
|
F.copy(r)
|
||
|
F.add(t)
|
||
|
F.add(z)
|
||
|
|
||
|
F.norm()
|
||
|
}
|
||
|
*/
|
||
|
/* XTR xtr_d function */
|
||
|
/*
|
||
|
func (F *FP16) xtr_D() {
|
||
|
w := NewFP16copy(F)
|
||
|
F.sqr()
|
||
|
w.conj()
|
||
|
w.add(w)
|
||
|
w.norm()
|
||
|
F.sub(w)
|
||
|
F.reduce()
|
||
|
}
|
||
|
*/
|
||
|
/* r=x^n using XTR method on traces of FP48s */
|
||
|
/*
|
||
|
func (F *FP16) xtr_pow(n *BIG) *FP16 {
|
||
|
sf := NewFP16copy(F)
|
||
|
sf.norm()
|
||
|
a := NewFP16int(3)
|
||
|
b := NewFP16copy(sf)
|
||
|
c := NewFP16copy(b)
|
||
|
c.xtr_D()
|
||
|
t := NewFP16()
|
||
|
r := NewFP16()
|
||
|
|
||
|
par := n.parity()
|
||
|
v := NewBIGcopy(n)
|
||
|
v.norm()
|
||
|
v.fshr(1)
|
||
|
if par == 0 {
|
||
|
v.dec(1)
|
||
|
v.norm()
|
||
|
}
|
||
|
|
||
|
nb := v.nbits()
|
||
|
for i := nb - 1; i >= 0; i-- {
|
||
|
if v.bit(i) != 1 {
|
||
|
t.copy(b)
|
||
|
sf.conj()
|
||
|
c.conj()
|
||
|
b.xtr_A(a, sf, c)
|
||
|
sf.conj()
|
||
|
c.copy(t)
|
||
|
c.xtr_D()
|
||
|
a.xtr_D()
|
||
|
} else {
|
||
|
t.copy(a)
|
||
|
t.conj()
|
||
|
a.copy(b)
|
||
|
a.xtr_D()
|
||
|
b.xtr_A(c, sf, t)
|
||
|
c.xtr_D()
|
||
|
}
|
||
|
}
|
||
|
if par == 0 {
|
||
|
r.copy(c)
|
||
|
} else {
|
||
|
r.copy(b)
|
||
|
}
|
||
|
r.reduce()
|
||
|
return r
|
||
|
}
|
||
|
*/
|
||
|
/* r=ck^a.cl^n using XTR double exponentiation method on traces of FP48s. See Stam thesis. */
|
||
|
/*
|
||
|
func (F *FP16) xtr_pow2(ck *FP16, ckml *FP16, ckm2l *FP16, a *BIG, b *BIG) *FP16 {
|
||
|
|
||
|
e := NewBIGcopy(a)
|
||
|
d := NewBIGcopy(b)
|
||
|
w := NewBIGint(0)
|
||
|
e.norm()
|
||
|
d.norm()
|
||
|
cu := NewFP16copy(ck) // can probably be passed in w/o copying
|
||
|
cv := NewFP16copy(F)
|
||
|
cumv := NewFP16copy(ckml)
|
||
|
cum2v := NewFP16copy(ckm2l)
|
||
|
r := NewFP16()
|
||
|
t := NewFP16()
|
||
|
|
||
|
f2 := 0
|
||
|
for d.parity() == 0 && e.parity() == 0 {
|
||
|
d.fshr(1)
|
||
|
e.fshr(1)
|
||
|
f2++
|
||
|
}
|
||
|
|
||
|
for Comp(d, e) != 0 {
|
||
|
if Comp(d, e) > 0 {
|
||
|
w.copy(e)
|
||
|
w.imul(4)
|
||
|
w.norm()
|
||
|
if Comp(d, w) <= 0 {
|
||
|
w.copy(d)
|
||
|
d.copy(e)
|
||
|
e.rsub(w)
|
||
|
e.norm()
|
||
|
|
||
|
t.copy(cv)
|
||
|
t.xtr_A(cu, cumv, cum2v)
|
||
|
cum2v.copy(cumv)
|
||
|
cum2v.conj()
|
||
|
cumv.copy(cv)
|
||
|
cv.copy(cu)
|
||
|
cu.copy(t)
|
||
|
} else {
|
||
|
if d.parity() == 0 {
|
||
|
d.fshr(1)
|
||
|
r.copy(cum2v)
|
||
|
r.conj()
|
||
|
t.copy(cumv)
|
||
|
t.xtr_A(cu, cv, r)
|
||
|
cum2v.copy(cumv)
|
||
|
cum2v.xtr_D()
|
||
|
cumv.copy(t)
|
||
|
cu.xtr_D()
|
||
|
} else {
|
||
|
if e.parity() == 1 {
|
||
|
d.sub(e)
|
||
|
d.norm()
|
||
|
d.fshr(1)
|
||
|
t.copy(cv)
|
||
|
t.xtr_A(cu, cumv, cum2v)
|
||
|
cu.xtr_D()
|
||
|
cum2v.copy(cv)
|
||
|
cum2v.xtr_D()
|
||
|
cum2v.conj()
|
||
|
cv.copy(t)
|
||
|
} else {
|
||
|
w.copy(d)
|
||
|
d.copy(e)
|
||
|
d.fshr(1)
|
||
|
e.copy(w)
|
||
|
t.copy(cumv)
|
||
|
t.xtr_D()
|
||
|
cumv.copy(cum2v)
|
||
|
cumv.conj()
|
||
|
cum2v.copy(t)
|
||
|
cum2v.conj()
|
||
|
t.copy(cv)
|
||
|
t.xtr_D()
|
||
|
cv.copy(cu)
|
||
|
cu.copy(t)
|
||
|
}
|
||
|
}
|
||
|
}
|
||
|
}
|
||
|
if Comp(d, e) < 0 {
|
||
|
w.copy(d)
|
||
|
w.imul(4)
|
||
|
w.norm()
|
||
|
if Comp(e, w) <= 0 {
|
||
|
e.sub(d)
|
||
|
e.norm()
|
||
|
t.copy(cv)
|
||
|
t.xtr_A(cu, cumv, cum2v)
|
||
|
cum2v.copy(cumv)
|
||
|
cumv.copy(cu)
|
||
|
cu.copy(t)
|
||
|
} else {
|
||
|
if e.parity() == 0 {
|
||
|
w.copy(d)
|
||
|
d.copy(e)
|
||
|
d.fshr(1)
|
||
|
e.copy(w)
|
||
|
t.copy(cumv)
|
||
|
t.xtr_D()
|
||
|
cumv.copy(cum2v)
|
||
|
cumv.conj()
|
||
|
cum2v.copy(t)
|
||
|
cum2v.conj()
|
||
|
t.copy(cv)
|
||
|
t.xtr_D()
|
||
|
cv.copy(cu)
|
||
|
cu.copy(t)
|
||
|
} else {
|
||
|
if d.parity() == 1 {
|
||
|
w.copy(e)
|
||
|
e.copy(d)
|
||
|
w.sub(d)
|
||
|
w.norm()
|
||
|
d.copy(w)
|
||
|
d.fshr(1)
|
||
|
t.copy(cv)
|
||
|
t.xtr_A(cu, cumv, cum2v)
|
||
|
cumv.conj()
|
||
|
cum2v.copy(cu)
|
||
|
cum2v.xtr_D()
|
||
|
cum2v.conj()
|
||
|
cu.copy(cv)
|
||
|
cu.xtr_D()
|
||
|
cv.copy(t)
|
||
|
} else {
|
||
|
d.fshr(1)
|
||
|
r.copy(cum2v)
|
||
|
r.conj()
|
||
|
t.copy(cumv)
|
||
|
t.xtr_A(cu, cv, r)
|
||
|
cum2v.copy(cumv)
|
||
|
cum2v.xtr_D()
|
||
|
cumv.copy(t)
|
||
|
cu.xtr_D()
|
||
|
}
|
||
|
}
|
||
|
}
|
||
|
}
|
||
|
}
|
||
|
r.copy(cv)
|
||
|
r.xtr_A(cu, cumv, cum2v)
|
||
|
for i := 0; i < f2; i++ {
|
||
|
r.xtr_D()
|
||
|
}
|
||
|
r = r.xtr_pow(d)
|
||
|
return r
|
||
|
}
|
||
|
*/
|