1
0
mirror of https://github.com/kubernetes-sigs/descheduler.git synced 2026-01-28 06:29:29 +01:00

Serve secure metrics at 10258

This commit is contained in:
Jan Chaloupka
2021-03-01 12:09:05 +01:00
parent 24458fb0ca
commit 701f22404b
927 changed files with 189382 additions and 7 deletions

752
vendor/golang.org/x/crypto/cryptobyte/asn1.go generated vendored Normal file
View File

@@ -0,0 +1,752 @@
// Copyright 2017 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
package cryptobyte
import (
encoding_asn1 "encoding/asn1"
"fmt"
"math/big"
"reflect"
"time"
"golang.org/x/crypto/cryptobyte/asn1"
)
// This file contains ASN.1-related methods for String and Builder.
// Builder
// AddASN1Int64 appends a DER-encoded ASN.1 INTEGER.
func (b *Builder) AddASN1Int64(v int64) {
b.addASN1Signed(asn1.INTEGER, v)
}
// AddASN1Int64WithTag appends a DER-encoded ASN.1 INTEGER with the
// given tag.
func (b *Builder) AddASN1Int64WithTag(v int64, tag asn1.Tag) {
b.addASN1Signed(tag, v)
}
// AddASN1Enum appends a DER-encoded ASN.1 ENUMERATION.
func (b *Builder) AddASN1Enum(v int64) {
b.addASN1Signed(asn1.ENUM, v)
}
func (b *Builder) addASN1Signed(tag asn1.Tag, v int64) {
b.AddASN1(tag, func(c *Builder) {
length := 1
for i := v; i >= 0x80 || i < -0x80; i >>= 8 {
length++
}
for ; length > 0; length-- {
i := v >> uint((length-1)*8) & 0xff
c.AddUint8(uint8(i))
}
})
}
// AddASN1Uint64 appends a DER-encoded ASN.1 INTEGER.
func (b *Builder) AddASN1Uint64(v uint64) {
b.AddASN1(asn1.INTEGER, func(c *Builder) {
length := 1
for i := v; i >= 0x80; i >>= 8 {
length++
}
for ; length > 0; length-- {
i := v >> uint((length-1)*8) & 0xff
c.AddUint8(uint8(i))
}
})
}
// AddASN1BigInt appends a DER-encoded ASN.1 INTEGER.
func (b *Builder) AddASN1BigInt(n *big.Int) {
if b.err != nil {
return
}
b.AddASN1(asn1.INTEGER, func(c *Builder) {
if n.Sign() < 0 {
// A negative number has to be converted to two's-complement form. So we
// invert and subtract 1. If the most-significant-bit isn't set then
// we'll need to pad the beginning with 0xff in order to keep the number
// negative.
nMinus1 := new(big.Int).Neg(n)
nMinus1.Sub(nMinus1, bigOne)
bytes := nMinus1.Bytes()
for i := range bytes {
bytes[i] ^= 0xff
}
if len(bytes) == 0 || bytes[0]&0x80 == 0 {
c.add(0xff)
}
c.add(bytes...)
} else if n.Sign() == 0 {
c.add(0)
} else {
bytes := n.Bytes()
if bytes[0]&0x80 != 0 {
c.add(0)
}
c.add(bytes...)
}
})
}
// AddASN1OctetString appends a DER-encoded ASN.1 OCTET STRING.
func (b *Builder) AddASN1OctetString(bytes []byte) {
b.AddASN1(asn1.OCTET_STRING, func(c *Builder) {
c.AddBytes(bytes)
})
}
const generalizedTimeFormatStr = "20060102150405Z0700"
// AddASN1GeneralizedTime appends a DER-encoded ASN.1 GENERALIZEDTIME.
func (b *Builder) AddASN1GeneralizedTime(t time.Time) {
if t.Year() < 0 || t.Year() > 9999 {
b.err = fmt.Errorf("cryptobyte: cannot represent %v as a GeneralizedTime", t)
return
}
b.AddASN1(asn1.GeneralizedTime, func(c *Builder) {
c.AddBytes([]byte(t.Format(generalizedTimeFormatStr)))
})
}
// AddASN1BitString appends a DER-encoded ASN.1 BIT STRING. This does not
// support BIT STRINGs that are not a whole number of bytes.
func (b *Builder) AddASN1BitString(data []byte) {
b.AddASN1(asn1.BIT_STRING, func(b *Builder) {
b.AddUint8(0)
b.AddBytes(data)
})
}
func (b *Builder) addBase128Int(n int64) {
var length int
if n == 0 {
length = 1
} else {
for i := n; i > 0; i >>= 7 {
length++
}
}
for i := length - 1; i >= 0; i-- {
o := byte(n >> uint(i*7))
o &= 0x7f
if i != 0 {
o |= 0x80
}
b.add(o)
}
}
func isValidOID(oid encoding_asn1.ObjectIdentifier) bool {
if len(oid) < 2 {
return false
}
if oid[0] > 2 || (oid[0] <= 1 && oid[1] >= 40) {
return false
}
for _, v := range oid {
if v < 0 {
return false
}
}
return true
}
func (b *Builder) AddASN1ObjectIdentifier(oid encoding_asn1.ObjectIdentifier) {
b.AddASN1(asn1.OBJECT_IDENTIFIER, func(b *Builder) {
if !isValidOID(oid) {
b.err = fmt.Errorf("cryptobyte: invalid OID: %v", oid)
return
}
b.addBase128Int(int64(oid[0])*40 + int64(oid[1]))
for _, v := range oid[2:] {
b.addBase128Int(int64(v))
}
})
}
func (b *Builder) AddASN1Boolean(v bool) {
b.AddASN1(asn1.BOOLEAN, func(b *Builder) {
if v {
b.AddUint8(0xff)
} else {
b.AddUint8(0)
}
})
}
func (b *Builder) AddASN1NULL() {
b.add(uint8(asn1.NULL), 0)
}
// MarshalASN1 calls encoding_asn1.Marshal on its input and appends the result if
// successful or records an error if one occurred.
func (b *Builder) MarshalASN1(v interface{}) {
// NOTE(martinkr): This is somewhat of a hack to allow propagation of
// encoding_asn1.Marshal errors into Builder.err. N.B. if you call MarshalASN1 with a
// value embedded into a struct, its tag information is lost.
if b.err != nil {
return
}
bytes, err := encoding_asn1.Marshal(v)
if err != nil {
b.err = err
return
}
b.AddBytes(bytes)
}
// AddASN1 appends an ASN.1 object. The object is prefixed with the given tag.
// Tags greater than 30 are not supported and result in an error (i.e.
// low-tag-number form only). The child builder passed to the
// BuilderContinuation can be used to build the content of the ASN.1 object.
func (b *Builder) AddASN1(tag asn1.Tag, f BuilderContinuation) {
if b.err != nil {
return
}
// Identifiers with the low five bits set indicate high-tag-number format
// (two or more octets), which we don't support.
if tag&0x1f == 0x1f {
b.err = fmt.Errorf("cryptobyte: high-tag number identifier octects not supported: 0x%x", tag)
return
}
b.AddUint8(uint8(tag))
b.addLengthPrefixed(1, true, f)
}
// String
// ReadASN1Boolean decodes an ASN.1 BOOLEAN and converts it to a boolean
// representation into out and advances. It reports whether the read
// was successful.
func (s *String) ReadASN1Boolean(out *bool) bool {
var bytes String
if !s.ReadASN1(&bytes, asn1.BOOLEAN) || len(bytes) != 1 {
return false
}
switch bytes[0] {
case 0:
*out = false
case 0xff:
*out = true
default:
return false
}
return true
}
var bigIntType = reflect.TypeOf((*big.Int)(nil)).Elem()
// ReadASN1Integer decodes an ASN.1 INTEGER into out and advances. If out does
// not point to an integer or to a big.Int, it panics. It reports whether the
// read was successful.
func (s *String) ReadASN1Integer(out interface{}) bool {
if reflect.TypeOf(out).Kind() != reflect.Ptr {
panic("out is not a pointer")
}
switch reflect.ValueOf(out).Elem().Kind() {
case reflect.Int, reflect.Int8, reflect.Int16, reflect.Int32, reflect.Int64:
var i int64
if !s.readASN1Int64(&i) || reflect.ValueOf(out).Elem().OverflowInt(i) {
return false
}
reflect.ValueOf(out).Elem().SetInt(i)
return true
case reflect.Uint, reflect.Uint8, reflect.Uint16, reflect.Uint32, reflect.Uint64:
var u uint64
if !s.readASN1Uint64(&u) || reflect.ValueOf(out).Elem().OverflowUint(u) {
return false
}
reflect.ValueOf(out).Elem().SetUint(u)
return true
case reflect.Struct:
if reflect.TypeOf(out).Elem() == bigIntType {
return s.readASN1BigInt(out.(*big.Int))
}
}
panic("out does not point to an integer type")
}
func checkASN1Integer(bytes []byte) bool {
if len(bytes) == 0 {
// An INTEGER is encoded with at least one octet.
return false
}
if len(bytes) == 1 {
return true
}
if bytes[0] == 0 && bytes[1]&0x80 == 0 || bytes[0] == 0xff && bytes[1]&0x80 == 0x80 {
// Value is not minimally encoded.
return false
}
return true
}
var bigOne = big.NewInt(1)
func (s *String) readASN1BigInt(out *big.Int) bool {
var bytes String
if !s.ReadASN1(&bytes, asn1.INTEGER) || !checkASN1Integer(bytes) {
return false
}
if bytes[0]&0x80 == 0x80 {
// Negative number.
neg := make([]byte, len(bytes))
for i, b := range bytes {
neg[i] = ^b
}
out.SetBytes(neg)
out.Add(out, bigOne)
out.Neg(out)
} else {
out.SetBytes(bytes)
}
return true
}
func (s *String) readASN1Int64(out *int64) bool {
var bytes String
if !s.ReadASN1(&bytes, asn1.INTEGER) || !checkASN1Integer(bytes) || !asn1Signed(out, bytes) {
return false
}
return true
}
func asn1Signed(out *int64, n []byte) bool {
length := len(n)
if length > 8 {
return false
}
for i := 0; i < length; i++ {
*out <<= 8
*out |= int64(n[i])
}
// Shift up and down in order to sign extend the result.
*out <<= 64 - uint8(length)*8
*out >>= 64 - uint8(length)*8
return true
}
func (s *String) readASN1Uint64(out *uint64) bool {
var bytes String
if !s.ReadASN1(&bytes, asn1.INTEGER) || !checkASN1Integer(bytes) || !asn1Unsigned(out, bytes) {
return false
}
return true
}
func asn1Unsigned(out *uint64, n []byte) bool {
length := len(n)
if length > 9 || length == 9 && n[0] != 0 {
// Too large for uint64.
return false
}
if n[0]&0x80 != 0 {
// Negative number.
return false
}
for i := 0; i < length; i++ {
*out <<= 8
*out |= uint64(n[i])
}
return true
}
// ReadASN1Int64WithTag decodes an ASN.1 INTEGER with the given tag into out
// and advances. It reports whether the read was successful and resulted in a
// value that can be represented in an int64.
func (s *String) ReadASN1Int64WithTag(out *int64, tag asn1.Tag) bool {
var bytes String
return s.ReadASN1(&bytes, tag) && checkASN1Integer(bytes) && asn1Signed(out, bytes)
}
// ReadASN1Enum decodes an ASN.1 ENUMERATION into out and advances. It reports
// whether the read was successful.
func (s *String) ReadASN1Enum(out *int) bool {
var bytes String
var i int64
if !s.ReadASN1(&bytes, asn1.ENUM) || !checkASN1Integer(bytes) || !asn1Signed(&i, bytes) {
return false
}
if int64(int(i)) != i {
return false
}
*out = int(i)
return true
}
func (s *String) readBase128Int(out *int) bool {
ret := 0
for i := 0; len(*s) > 0; i++ {
if i == 4 {
return false
}
ret <<= 7
b := s.read(1)[0]
ret |= int(b & 0x7f)
if b&0x80 == 0 {
*out = ret
return true
}
}
return false // truncated
}
// ReadASN1ObjectIdentifier decodes an ASN.1 OBJECT IDENTIFIER into out and
// advances. It reports whether the read was successful.
func (s *String) ReadASN1ObjectIdentifier(out *encoding_asn1.ObjectIdentifier) bool {
var bytes String
if !s.ReadASN1(&bytes, asn1.OBJECT_IDENTIFIER) || len(bytes) == 0 {
return false
}
// In the worst case, we get two elements from the first byte (which is
// encoded differently) and then every varint is a single byte long.
components := make([]int, len(bytes)+1)
// The first varint is 40*value1 + value2:
// According to this packing, value1 can take the values 0, 1 and 2 only.
// When value1 = 0 or value1 = 1, then value2 is <= 39. When value1 = 2,
// then there are no restrictions on value2.
var v int
if !bytes.readBase128Int(&v) {
return false
}
if v < 80 {
components[0] = v / 40
components[1] = v % 40
} else {
components[0] = 2
components[1] = v - 80
}
i := 2
for ; len(bytes) > 0; i++ {
if !bytes.readBase128Int(&v) {
return false
}
components[i] = v
}
*out = components[:i]
return true
}
// ReadASN1GeneralizedTime decodes an ASN.1 GENERALIZEDTIME into out and
// advances. It reports whether the read was successful.
func (s *String) ReadASN1GeneralizedTime(out *time.Time) bool {
var bytes String
if !s.ReadASN1(&bytes, asn1.GeneralizedTime) {
return false
}
t := string(bytes)
res, err := time.Parse(generalizedTimeFormatStr, t)
if err != nil {
return false
}
if serialized := res.Format(generalizedTimeFormatStr); serialized != t {
return false
}
*out = res
return true
}
// ReadASN1BitString decodes an ASN.1 BIT STRING into out and advances.
// It reports whether the read was successful.
func (s *String) ReadASN1BitString(out *encoding_asn1.BitString) bool {
var bytes String
if !s.ReadASN1(&bytes, asn1.BIT_STRING) || len(bytes) == 0 ||
len(bytes)*8/8 != len(bytes) {
return false
}
paddingBits := uint8(bytes[0])
bytes = bytes[1:]
if paddingBits > 7 ||
len(bytes) == 0 && paddingBits != 0 ||
len(bytes) > 0 && bytes[len(bytes)-1]&(1<<paddingBits-1) != 0 {
return false
}
out.BitLength = len(bytes)*8 - int(paddingBits)
out.Bytes = bytes
return true
}
// ReadASN1BitString decodes an ASN.1 BIT STRING into out and advances. It is
// an error if the BIT STRING is not a whole number of bytes. It reports
// whether the read was successful.
func (s *String) ReadASN1BitStringAsBytes(out *[]byte) bool {
var bytes String
if !s.ReadASN1(&bytes, asn1.BIT_STRING) || len(bytes) == 0 {
return false
}
paddingBits := uint8(bytes[0])
if paddingBits != 0 {
return false
}
*out = bytes[1:]
return true
}
// ReadASN1Bytes reads the contents of a DER-encoded ASN.1 element (not including
// tag and length bytes) into out, and advances. The element must match the
// given tag. It reports whether the read was successful.
func (s *String) ReadASN1Bytes(out *[]byte, tag asn1.Tag) bool {
return s.ReadASN1((*String)(out), tag)
}
// ReadASN1 reads the contents of a DER-encoded ASN.1 element (not including
// tag and length bytes) into out, and advances. The element must match the
// given tag. It reports whether the read was successful.
//
// Tags greater than 30 are not supported (i.e. low-tag-number format only).
func (s *String) ReadASN1(out *String, tag asn1.Tag) bool {
var t asn1.Tag
if !s.ReadAnyASN1(out, &t) || t != tag {
return false
}
return true
}
// ReadASN1Element reads the contents of a DER-encoded ASN.1 element (including
// tag and length bytes) into out, and advances. The element must match the
// given tag. It reports whether the read was successful.
//
// Tags greater than 30 are not supported (i.e. low-tag-number format only).
func (s *String) ReadASN1Element(out *String, tag asn1.Tag) bool {
var t asn1.Tag
if !s.ReadAnyASN1Element(out, &t) || t != tag {
return false
}
return true
}
// ReadAnyASN1 reads the contents of a DER-encoded ASN.1 element (not including
// tag and length bytes) into out, sets outTag to its tag, and advances.
// It reports whether the read was successful.
//
// Tags greater than 30 are not supported (i.e. low-tag-number format only).
func (s *String) ReadAnyASN1(out *String, outTag *asn1.Tag) bool {
return s.readASN1(out, outTag, true /* skip header */)
}
// ReadAnyASN1Element reads the contents of a DER-encoded ASN.1 element
// (including tag and length bytes) into out, sets outTag to is tag, and
// advances. It reports whether the read was successful.
//
// Tags greater than 30 are not supported (i.e. low-tag-number format only).
func (s *String) ReadAnyASN1Element(out *String, outTag *asn1.Tag) bool {
return s.readASN1(out, outTag, false /* include header */)
}
// PeekASN1Tag reports whether the next ASN.1 value on the string starts with
// the given tag.
func (s String) PeekASN1Tag(tag asn1.Tag) bool {
if len(s) == 0 {
return false
}
return asn1.Tag(s[0]) == tag
}
// SkipASN1 reads and discards an ASN.1 element with the given tag. It
// reports whether the operation was successful.
func (s *String) SkipASN1(tag asn1.Tag) bool {
var unused String
return s.ReadASN1(&unused, tag)
}
// ReadOptionalASN1 attempts to read the contents of a DER-encoded ASN.1
// element (not including tag and length bytes) tagged with the given tag into
// out. It stores whether an element with the tag was found in outPresent,
// unless outPresent is nil. It reports whether the read was successful.
func (s *String) ReadOptionalASN1(out *String, outPresent *bool, tag asn1.Tag) bool {
present := s.PeekASN1Tag(tag)
if outPresent != nil {
*outPresent = present
}
if present && !s.ReadASN1(out, tag) {
return false
}
return true
}
// SkipOptionalASN1 advances s over an ASN.1 element with the given tag, or
// else leaves s unchanged. It reports whether the operation was successful.
func (s *String) SkipOptionalASN1(tag asn1.Tag) bool {
if !s.PeekASN1Tag(tag) {
return true
}
var unused String
return s.ReadASN1(&unused, tag)
}
// ReadOptionalASN1Integer attempts to read an optional ASN.1 INTEGER
// explicitly tagged with tag into out and advances. If no element with a
// matching tag is present, it writes defaultValue into out instead. If out
// does not point to an integer or to a big.Int, it panics. It reports
// whether the read was successful.
func (s *String) ReadOptionalASN1Integer(out interface{}, tag asn1.Tag, defaultValue interface{}) bool {
if reflect.TypeOf(out).Kind() != reflect.Ptr {
panic("out is not a pointer")
}
var present bool
var i String
if !s.ReadOptionalASN1(&i, &present, tag) {
return false
}
if !present {
switch reflect.ValueOf(out).Elem().Kind() {
case reflect.Int, reflect.Int8, reflect.Int16, reflect.Int32, reflect.Int64,
reflect.Uint, reflect.Uint8, reflect.Uint16, reflect.Uint32, reflect.Uint64:
reflect.ValueOf(out).Elem().Set(reflect.ValueOf(defaultValue))
case reflect.Struct:
if reflect.TypeOf(out).Elem() != bigIntType {
panic("invalid integer type")
}
if reflect.TypeOf(defaultValue).Kind() != reflect.Ptr ||
reflect.TypeOf(defaultValue).Elem() != bigIntType {
panic("out points to big.Int, but defaultValue does not")
}
out.(*big.Int).Set(defaultValue.(*big.Int))
default:
panic("invalid integer type")
}
return true
}
if !i.ReadASN1Integer(out) || !i.Empty() {
return false
}
return true
}
// ReadOptionalASN1OctetString attempts to read an optional ASN.1 OCTET STRING
// explicitly tagged with tag into out and advances. If no element with a
// matching tag is present, it sets "out" to nil instead. It reports
// whether the read was successful.
func (s *String) ReadOptionalASN1OctetString(out *[]byte, outPresent *bool, tag asn1.Tag) bool {
var present bool
var child String
if !s.ReadOptionalASN1(&child, &present, tag) {
return false
}
if outPresent != nil {
*outPresent = present
}
if present {
var oct String
if !child.ReadASN1(&oct, asn1.OCTET_STRING) || !child.Empty() {
return false
}
*out = oct
} else {
*out = nil
}
return true
}
// ReadOptionalASN1Boolean sets *out to the value of the next ASN.1 BOOLEAN or,
// if the next bytes are not an ASN.1 BOOLEAN, to the value of defaultValue.
// It reports whether the operation was successful.
func (s *String) ReadOptionalASN1Boolean(out *bool, defaultValue bool) bool {
var present bool
var child String
if !s.ReadOptionalASN1(&child, &present, asn1.BOOLEAN) {
return false
}
if !present {
*out = defaultValue
return true
}
return s.ReadASN1Boolean(out)
}
func (s *String) readASN1(out *String, outTag *asn1.Tag, skipHeader bool) bool {
if len(*s) < 2 {
return false
}
tag, lenByte := (*s)[0], (*s)[1]
if tag&0x1f == 0x1f {
// ITU-T X.690 section 8.1.2
//
// An identifier octet with a tag part of 0x1f indicates a high-tag-number
// form identifier with two or more octets. We only support tags less than
// 31 (i.e. low-tag-number form, single octet identifier).
return false
}
if outTag != nil {
*outTag = asn1.Tag(tag)
}
// ITU-T X.690 section 8.1.3
//
// Bit 8 of the first length byte indicates whether the length is short- or
// long-form.
var length, headerLen uint32 // length includes headerLen
if lenByte&0x80 == 0 {
// Short-form length (section 8.1.3.4), encoded in bits 1-7.
length = uint32(lenByte) + 2
headerLen = 2
} else {
// Long-form length (section 8.1.3.5). Bits 1-7 encode the number of octets
// used to encode the length.
lenLen := lenByte & 0x7f
var len32 uint32
if lenLen == 0 || lenLen > 4 || len(*s) < int(2+lenLen) {
return false
}
lenBytes := String((*s)[2 : 2+lenLen])
if !lenBytes.readUnsigned(&len32, int(lenLen)) {
return false
}
// ITU-T X.690 section 10.1 (DER length forms) requires encoding the length
// with the minimum number of octets.
if len32 < 128 {
// Length should have used short-form encoding.
return false
}
if len32>>((lenLen-1)*8) == 0 {
// Leading octet is 0. Length should have been at least one byte shorter.
return false
}
headerLen = 2 + uint32(lenLen)
if headerLen+len32 < len32 {
// Overflow.
return false
}
length = headerLen + len32
}
if int(length) < 0 || !s.ReadBytes((*[]byte)(out), int(length)) {
return false
}
if skipHeader && !out.Skip(int(headerLen)) {
panic("cryptobyte: internal error")
}
return true
}

46
vendor/golang.org/x/crypto/cryptobyte/asn1/asn1.go generated vendored Normal file
View File

@@ -0,0 +1,46 @@
// Copyright 2017 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
// Package asn1 contains supporting types for parsing and building ASN.1
// messages with the cryptobyte package.
package asn1 // import "golang.org/x/crypto/cryptobyte/asn1"
// Tag represents an ASN.1 identifier octet, consisting of a tag number
// (indicating a type) and class (such as context-specific or constructed).
//
// Methods in the cryptobyte package only support the low-tag-number form, i.e.
// a single identifier octet with bits 7-8 encoding the class and bits 1-6
// encoding the tag number.
type Tag uint8
const (
classConstructed = 0x20
classContextSpecific = 0x80
)
// Constructed returns t with the constructed class bit set.
func (t Tag) Constructed() Tag { return t | classConstructed }
// ContextSpecific returns t with the context-specific class bit set.
func (t Tag) ContextSpecific() Tag { return t | classContextSpecific }
// The following is a list of standard tag and class combinations.
const (
BOOLEAN = Tag(1)
INTEGER = Tag(2)
BIT_STRING = Tag(3)
OCTET_STRING = Tag(4)
NULL = Tag(5)
OBJECT_IDENTIFIER = Tag(6)
ENUM = Tag(10)
UTF8String = Tag(12)
SEQUENCE = Tag(16 | classConstructed)
SET = Tag(17 | classConstructed)
PrintableString = Tag(19)
T61String = Tag(20)
IA5String = Tag(22)
UTCTime = Tag(23)
GeneralizedTime = Tag(24)
GeneralString = Tag(27)
)

337
vendor/golang.org/x/crypto/cryptobyte/builder.go generated vendored Normal file
View File

@@ -0,0 +1,337 @@
// Copyright 2017 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
package cryptobyte
import (
"errors"
"fmt"
)
// A Builder builds byte strings from fixed-length and length-prefixed values.
// Builders either allocate space as needed, or are fixed, which means that
// they write into a given buffer and produce an error if it's exhausted.
//
// The zero value is a usable Builder that allocates space as needed.
//
// Simple values are marshaled and appended to a Builder using methods on the
// Builder. Length-prefixed values are marshaled by providing a
// BuilderContinuation, which is a function that writes the inner contents of
// the value to a given Builder. See the documentation for BuilderContinuation
// for details.
type Builder struct {
err error
result []byte
fixedSize bool
child *Builder
offset int
pendingLenLen int
pendingIsASN1 bool
inContinuation *bool
}
// NewBuilder creates a Builder that appends its output to the given buffer.
// Like append(), the slice will be reallocated if its capacity is exceeded.
// Use Bytes to get the final buffer.
func NewBuilder(buffer []byte) *Builder {
return &Builder{
result: buffer,
}
}
// NewFixedBuilder creates a Builder that appends its output into the given
// buffer. This builder does not reallocate the output buffer. Writes that
// would exceed the buffer's capacity are treated as an error.
func NewFixedBuilder(buffer []byte) *Builder {
return &Builder{
result: buffer,
fixedSize: true,
}
}
// SetError sets the value to be returned as the error from Bytes. Writes
// performed after calling SetError are ignored.
func (b *Builder) SetError(err error) {
b.err = err
}
// Bytes returns the bytes written by the builder or an error if one has
// occurred during building.
func (b *Builder) Bytes() ([]byte, error) {
if b.err != nil {
return nil, b.err
}
return b.result[b.offset:], nil
}
// BytesOrPanic returns the bytes written by the builder or panics if an error
// has occurred during building.
func (b *Builder) BytesOrPanic() []byte {
if b.err != nil {
panic(b.err)
}
return b.result[b.offset:]
}
// AddUint8 appends an 8-bit value to the byte string.
func (b *Builder) AddUint8(v uint8) {
b.add(byte(v))
}
// AddUint16 appends a big-endian, 16-bit value to the byte string.
func (b *Builder) AddUint16(v uint16) {
b.add(byte(v>>8), byte(v))
}
// AddUint24 appends a big-endian, 24-bit value to the byte string. The highest
// byte of the 32-bit input value is silently truncated.
func (b *Builder) AddUint24(v uint32) {
b.add(byte(v>>16), byte(v>>8), byte(v))
}
// AddUint32 appends a big-endian, 32-bit value to the byte string.
func (b *Builder) AddUint32(v uint32) {
b.add(byte(v>>24), byte(v>>16), byte(v>>8), byte(v))
}
// AddBytes appends a sequence of bytes to the byte string.
func (b *Builder) AddBytes(v []byte) {
b.add(v...)
}
// BuilderContinuation is a continuation-passing interface for building
// length-prefixed byte sequences. Builder methods for length-prefixed
// sequences (AddUint8LengthPrefixed etc) will invoke the BuilderContinuation
// supplied to them. The child builder passed to the continuation can be used
// to build the content of the length-prefixed sequence. For example:
//
// parent := cryptobyte.NewBuilder()
// parent.AddUint8LengthPrefixed(func (child *Builder) {
// child.AddUint8(42)
// child.AddUint8LengthPrefixed(func (grandchild *Builder) {
// grandchild.AddUint8(5)
// })
// })
//
// It is an error to write more bytes to the child than allowed by the reserved
// length prefix. After the continuation returns, the child must be considered
// invalid, i.e. users must not store any copies or references of the child
// that outlive the continuation.
//
// If the continuation panics with a value of type BuildError then the inner
// error will be returned as the error from Bytes. If the child panics
// otherwise then Bytes will repanic with the same value.
type BuilderContinuation func(child *Builder)
// BuildError wraps an error. If a BuilderContinuation panics with this value,
// the panic will be recovered and the inner error will be returned from
// Builder.Bytes.
type BuildError struct {
Err error
}
// AddUint8LengthPrefixed adds a 8-bit length-prefixed byte sequence.
func (b *Builder) AddUint8LengthPrefixed(f BuilderContinuation) {
b.addLengthPrefixed(1, false, f)
}
// AddUint16LengthPrefixed adds a big-endian, 16-bit length-prefixed byte sequence.
func (b *Builder) AddUint16LengthPrefixed(f BuilderContinuation) {
b.addLengthPrefixed(2, false, f)
}
// AddUint24LengthPrefixed adds a big-endian, 24-bit length-prefixed byte sequence.
func (b *Builder) AddUint24LengthPrefixed(f BuilderContinuation) {
b.addLengthPrefixed(3, false, f)
}
// AddUint32LengthPrefixed adds a big-endian, 32-bit length-prefixed byte sequence.
func (b *Builder) AddUint32LengthPrefixed(f BuilderContinuation) {
b.addLengthPrefixed(4, false, f)
}
func (b *Builder) callContinuation(f BuilderContinuation, arg *Builder) {
if !*b.inContinuation {
*b.inContinuation = true
defer func() {
*b.inContinuation = false
r := recover()
if r == nil {
return
}
if buildError, ok := r.(BuildError); ok {
b.err = buildError.Err
} else {
panic(r)
}
}()
}
f(arg)
}
func (b *Builder) addLengthPrefixed(lenLen int, isASN1 bool, f BuilderContinuation) {
// Subsequent writes can be ignored if the builder has encountered an error.
if b.err != nil {
return
}
offset := len(b.result)
b.add(make([]byte, lenLen)...)
if b.inContinuation == nil {
b.inContinuation = new(bool)
}
b.child = &Builder{
result: b.result,
fixedSize: b.fixedSize,
offset: offset,
pendingLenLen: lenLen,
pendingIsASN1: isASN1,
inContinuation: b.inContinuation,
}
b.callContinuation(f, b.child)
b.flushChild()
if b.child != nil {
panic("cryptobyte: internal error")
}
}
func (b *Builder) flushChild() {
if b.child == nil {
return
}
b.child.flushChild()
child := b.child
b.child = nil
if child.err != nil {
b.err = child.err
return
}
length := len(child.result) - child.pendingLenLen - child.offset
if length < 0 {
panic("cryptobyte: internal error") // result unexpectedly shrunk
}
if child.pendingIsASN1 {
// For ASN.1, we reserved a single byte for the length. If that turned out
// to be incorrect, we have to move the contents along in order to make
// space.
if child.pendingLenLen != 1 {
panic("cryptobyte: internal error")
}
var lenLen, lenByte uint8
if int64(length) > 0xfffffffe {
b.err = errors.New("pending ASN.1 child too long")
return
} else if length > 0xffffff {
lenLen = 5
lenByte = 0x80 | 4
} else if length > 0xffff {
lenLen = 4
lenByte = 0x80 | 3
} else if length > 0xff {
lenLen = 3
lenByte = 0x80 | 2
} else if length > 0x7f {
lenLen = 2
lenByte = 0x80 | 1
} else {
lenLen = 1
lenByte = uint8(length)
length = 0
}
// Insert the initial length byte, make space for successive length bytes,
// and adjust the offset.
child.result[child.offset] = lenByte
extraBytes := int(lenLen - 1)
if extraBytes != 0 {
child.add(make([]byte, extraBytes)...)
childStart := child.offset + child.pendingLenLen
copy(child.result[childStart+extraBytes:], child.result[childStart:])
}
child.offset++
child.pendingLenLen = extraBytes
}
l := length
for i := child.pendingLenLen - 1; i >= 0; i-- {
child.result[child.offset+i] = uint8(l)
l >>= 8
}
if l != 0 {
b.err = fmt.Errorf("cryptobyte: pending child length %d exceeds %d-byte length prefix", length, child.pendingLenLen)
return
}
if b.fixedSize && &b.result[0] != &child.result[0] {
panic("cryptobyte: BuilderContinuation reallocated a fixed-size buffer")
}
b.result = child.result
}
func (b *Builder) add(bytes ...byte) {
if b.err != nil {
return
}
if b.child != nil {
panic("cryptobyte: attempted write while child is pending")
}
if len(b.result)+len(bytes) < len(bytes) {
b.err = errors.New("cryptobyte: length overflow")
}
if b.fixedSize && len(b.result)+len(bytes) > cap(b.result) {
b.err = errors.New("cryptobyte: Builder is exceeding its fixed-size buffer")
return
}
b.result = append(b.result, bytes...)
}
// Unwrite rolls back n bytes written directly to the Builder. An attempt by a
// child builder passed to a continuation to unwrite bytes from its parent will
// panic.
func (b *Builder) Unwrite(n int) {
if b.err != nil {
return
}
if b.child != nil {
panic("cryptobyte: attempted unwrite while child is pending")
}
length := len(b.result) - b.pendingLenLen - b.offset
if length < 0 {
panic("cryptobyte: internal error")
}
if n > length {
panic("cryptobyte: attempted to unwrite more than was written")
}
b.result = b.result[:len(b.result)-n]
}
// A MarshalingValue marshals itself into a Builder.
type MarshalingValue interface {
// Marshal is called by Builder.AddValue. It receives a pointer to a builder
// to marshal itself into. It may return an error that occurred during
// marshaling, such as unset or invalid values.
Marshal(b *Builder) error
}
// AddValue calls Marshal on v, passing a pointer to the builder to append to.
// If Marshal returns an error, it is set on the Builder so that subsequent
// appends don't have an effect.
func (b *Builder) AddValue(v MarshalingValue) {
err := v.Marshal(b)
if err != nil {
b.err = err
}
}

161
vendor/golang.org/x/crypto/cryptobyte/string.go generated vendored Normal file
View File

@@ -0,0 +1,161 @@
// Copyright 2017 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
// Package cryptobyte contains types that help with parsing and constructing
// length-prefixed, binary messages, including ASN.1 DER. (The asn1 subpackage
// contains useful ASN.1 constants.)
//
// The String type is for parsing. It wraps a []byte slice and provides helper
// functions for consuming structures, value by value.
//
// The Builder type is for constructing messages. It providers helper functions
// for appending values and also for appending length-prefixed submessages
// without having to worry about calculating the length prefix ahead of time.
//
// See the documentation and examples for the Builder and String types to get
// started.
package cryptobyte // import "golang.org/x/crypto/cryptobyte"
// String represents a string of bytes. It provides methods for parsing
// fixed-length and length-prefixed values from it.
type String []byte
// read advances a String by n bytes and returns them. If less than n bytes
// remain, it returns nil.
func (s *String) read(n int) []byte {
if len(*s) < n || n < 0 {
return nil
}
v := (*s)[:n]
*s = (*s)[n:]
return v
}
// Skip advances the String by n byte and reports whether it was successful.
func (s *String) Skip(n int) bool {
return s.read(n) != nil
}
// ReadUint8 decodes an 8-bit value into out and advances over it.
// It reports whether the read was successful.
func (s *String) ReadUint8(out *uint8) bool {
v := s.read(1)
if v == nil {
return false
}
*out = uint8(v[0])
return true
}
// ReadUint16 decodes a big-endian, 16-bit value into out and advances over it.
// It reports whether the read was successful.
func (s *String) ReadUint16(out *uint16) bool {
v := s.read(2)
if v == nil {
return false
}
*out = uint16(v[0])<<8 | uint16(v[1])
return true
}
// ReadUint24 decodes a big-endian, 24-bit value into out and advances over it.
// It reports whether the read was successful.
func (s *String) ReadUint24(out *uint32) bool {
v := s.read(3)
if v == nil {
return false
}
*out = uint32(v[0])<<16 | uint32(v[1])<<8 | uint32(v[2])
return true
}
// ReadUint32 decodes a big-endian, 32-bit value into out and advances over it.
// It reports whether the read was successful.
func (s *String) ReadUint32(out *uint32) bool {
v := s.read(4)
if v == nil {
return false
}
*out = uint32(v[0])<<24 | uint32(v[1])<<16 | uint32(v[2])<<8 | uint32(v[3])
return true
}
func (s *String) readUnsigned(out *uint32, length int) bool {
v := s.read(length)
if v == nil {
return false
}
var result uint32
for i := 0; i < length; i++ {
result <<= 8
result |= uint32(v[i])
}
*out = result
return true
}
func (s *String) readLengthPrefixed(lenLen int, outChild *String) bool {
lenBytes := s.read(lenLen)
if lenBytes == nil {
return false
}
var length uint32
for _, b := range lenBytes {
length = length << 8
length = length | uint32(b)
}
v := s.read(int(length))
if v == nil {
return false
}
*outChild = v
return true
}
// ReadUint8LengthPrefixed reads the content of an 8-bit length-prefixed value
// into out and advances over it. It reports whether the read was successful.
func (s *String) ReadUint8LengthPrefixed(out *String) bool {
return s.readLengthPrefixed(1, out)
}
// ReadUint16LengthPrefixed reads the content of a big-endian, 16-bit
// length-prefixed value into out and advances over it. It reports whether the
// read was successful.
func (s *String) ReadUint16LengthPrefixed(out *String) bool {
return s.readLengthPrefixed(2, out)
}
// ReadUint24LengthPrefixed reads the content of a big-endian, 24-bit
// length-prefixed value into out and advances over it. It reports whether
// the read was successful.
func (s *String) ReadUint24LengthPrefixed(out *String) bool {
return s.readLengthPrefixed(3, out)
}
// ReadBytes reads n bytes into out and advances over them. It reports
// whether the read was successful.
func (s *String) ReadBytes(out *[]byte, n int) bool {
v := s.read(n)
if v == nil {
return false
}
*out = v
return true
}
// CopyBytes copies len(out) bytes into out and advances over them. It reports
// whether the copy operation was successful
func (s *String) CopyBytes(out []byte) bool {
n := len(out)
v := s.read(n)
if v == nil {
return false
}
return copy(out, v) == n
}
// Empty reports whether the string does not contain any bytes.
func (s String) Empty() bool {
return len(s) == 0
}

32
vendor/golang.org/x/crypto/internal/subtle/aliasing.go generated vendored Normal file
View File

@@ -0,0 +1,32 @@
// Copyright 2018 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
// +build !appengine
// Package subtle implements functions that are often useful in cryptographic
// code but require careful thought to use correctly.
package subtle // import "golang.org/x/crypto/internal/subtle"
import "unsafe"
// AnyOverlap reports whether x and y share memory at any (not necessarily
// corresponding) index. The memory beyond the slice length is ignored.
func AnyOverlap(x, y []byte) bool {
return len(x) > 0 && len(y) > 0 &&
uintptr(unsafe.Pointer(&x[0])) <= uintptr(unsafe.Pointer(&y[len(y)-1])) &&
uintptr(unsafe.Pointer(&y[0])) <= uintptr(unsafe.Pointer(&x[len(x)-1]))
}
// InexactOverlap reports whether x and y share memory at any non-corresponding
// index. The memory beyond the slice length is ignored. Note that x and y can
// have different lengths and still not have any inexact overlap.
//
// InexactOverlap can be used to implement the requirements of the crypto/cipher
// AEAD, Block, BlockMode and Stream interfaces.
func InexactOverlap(x, y []byte) bool {
if len(x) == 0 || len(y) == 0 || &x[0] == &y[0] {
return false
}
return AnyOverlap(x, y)
}

View File

@@ -0,0 +1,35 @@
// Copyright 2018 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
// +build appengine
// Package subtle implements functions that are often useful in cryptographic
// code but require careful thought to use correctly.
package subtle // import "golang.org/x/crypto/internal/subtle"
// This is the Google App Engine standard variant based on reflect
// because the unsafe package and cgo are disallowed.
import "reflect"
// AnyOverlap reports whether x and y share memory at any (not necessarily
// corresponding) index. The memory beyond the slice length is ignored.
func AnyOverlap(x, y []byte) bool {
return len(x) > 0 && len(y) > 0 &&
reflect.ValueOf(&x[0]).Pointer() <= reflect.ValueOf(&y[len(y)-1]).Pointer() &&
reflect.ValueOf(&y[0]).Pointer() <= reflect.ValueOf(&x[len(x)-1]).Pointer()
}
// InexactOverlap reports whether x and y share memory at any non-corresponding
// index. The memory beyond the slice length is ignored. Note that x and y can
// have different lengths and still not have any inexact overlap.
//
// InexactOverlap can be used to implement the requirements of the crypto/cipher
// AEAD, Block, BlockMode and Stream interfaces.
func InexactOverlap(x, y []byte) bool {
if len(x) == 0 || len(y) == 0 || &x[0] == &y[0] {
return false
}
return AnyOverlap(x, y)
}

173
vendor/golang.org/x/crypto/nacl/secretbox/secretbox.go generated vendored Normal file
View File

@@ -0,0 +1,173 @@
// Copyright 2012 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
/*
Package secretbox encrypts and authenticates small messages.
Secretbox uses XSalsa20 and Poly1305 to encrypt and authenticate messages with
secret-key cryptography. The length of messages is not hidden.
It is the caller's responsibility to ensure the uniqueness of nonces—for
example, by using nonce 1 for the first message, nonce 2 for the second
message, etc. Nonces are long enough that randomly generated nonces have
negligible risk of collision.
Messages should be small because:
1. The whole message needs to be held in memory to be processed.
2. Using large messages pressures implementations on small machines to decrypt
and process plaintext before authenticating it. This is very dangerous, and
this API does not allow it, but a protocol that uses excessive message sizes
might present some implementations with no other choice.
3. Fixed overheads will be sufficiently amortised by messages as small as 8KB.
4. Performance may be improved by working with messages that fit into data caches.
Thus large amounts of data should be chunked so that each message is small.
(Each message still needs a unique nonce.) If in doubt, 16KB is a reasonable
chunk size.
This package is interoperable with NaCl: https://nacl.cr.yp.to/secretbox.html.
*/
package secretbox // import "golang.org/x/crypto/nacl/secretbox"
import (
"golang.org/x/crypto/internal/subtle"
"golang.org/x/crypto/poly1305"
"golang.org/x/crypto/salsa20/salsa"
)
// Overhead is the number of bytes of overhead when boxing a message.
const Overhead = poly1305.TagSize
// setup produces a sub-key and Salsa20 counter given a nonce and key.
func setup(subKey *[32]byte, counter *[16]byte, nonce *[24]byte, key *[32]byte) {
// We use XSalsa20 for encryption so first we need to generate a
// key and nonce with HSalsa20.
var hNonce [16]byte
copy(hNonce[:], nonce[:])
salsa.HSalsa20(subKey, &hNonce, key, &salsa.Sigma)
// The final 8 bytes of the original nonce form the new nonce.
copy(counter[:], nonce[16:])
}
// sliceForAppend takes a slice and a requested number of bytes. It returns a
// slice with the contents of the given slice followed by that many bytes and a
// second slice that aliases into it and contains only the extra bytes. If the
// original slice has sufficient capacity then no allocation is performed.
func sliceForAppend(in []byte, n int) (head, tail []byte) {
if total := len(in) + n; cap(in) >= total {
head = in[:total]
} else {
head = make([]byte, total)
copy(head, in)
}
tail = head[len(in):]
return
}
// Seal appends an encrypted and authenticated copy of message to out, which
// must not overlap message. The key and nonce pair must be unique for each
// distinct message and the output will be Overhead bytes longer than message.
func Seal(out, message []byte, nonce *[24]byte, key *[32]byte) []byte {
var subKey [32]byte
var counter [16]byte
setup(&subKey, &counter, nonce, key)
// The Poly1305 key is generated by encrypting 32 bytes of zeros. Since
// Salsa20 works with 64-byte blocks, we also generate 32 bytes of
// keystream as a side effect.
var firstBlock [64]byte
salsa.XORKeyStream(firstBlock[:], firstBlock[:], &counter, &subKey)
var poly1305Key [32]byte
copy(poly1305Key[:], firstBlock[:])
ret, out := sliceForAppend(out, len(message)+poly1305.TagSize)
if subtle.AnyOverlap(out, message) {
panic("nacl: invalid buffer overlap")
}
// We XOR up to 32 bytes of message with the keystream generated from
// the first block.
firstMessageBlock := message
if len(firstMessageBlock) > 32 {
firstMessageBlock = firstMessageBlock[:32]
}
tagOut := out
out = out[poly1305.TagSize:]
for i, x := range firstMessageBlock {
out[i] = firstBlock[32+i] ^ x
}
message = message[len(firstMessageBlock):]
ciphertext := out
out = out[len(firstMessageBlock):]
// Now encrypt the rest.
counter[8] = 1
salsa.XORKeyStream(out, message, &counter, &subKey)
var tag [poly1305.TagSize]byte
poly1305.Sum(&tag, ciphertext, &poly1305Key)
copy(tagOut, tag[:])
return ret
}
// Open authenticates and decrypts a box produced by Seal and appends the
// message to out, which must not overlap box. The output will be Overhead
// bytes smaller than box.
func Open(out, box []byte, nonce *[24]byte, key *[32]byte) ([]byte, bool) {
if len(box) < Overhead {
return nil, false
}
var subKey [32]byte
var counter [16]byte
setup(&subKey, &counter, nonce, key)
// The Poly1305 key is generated by encrypting 32 bytes of zeros. Since
// Salsa20 works with 64-byte blocks, we also generate 32 bytes of
// keystream as a side effect.
var firstBlock [64]byte
salsa.XORKeyStream(firstBlock[:], firstBlock[:], &counter, &subKey)
var poly1305Key [32]byte
copy(poly1305Key[:], firstBlock[:])
var tag [poly1305.TagSize]byte
copy(tag[:], box)
if !poly1305.Verify(&tag, box[poly1305.TagSize:], &poly1305Key) {
return nil, false
}
ret, out := sliceForAppend(out, len(box)-Overhead)
if subtle.AnyOverlap(out, box) {
panic("nacl: invalid buffer overlap")
}
// We XOR up to 32 bytes of box with the keystream generated from
// the first block.
box = box[Overhead:]
firstMessageBlock := box
if len(firstMessageBlock) > 32 {
firstMessageBlock = firstMessageBlock[:32]
}
for i, x := range firstMessageBlock {
out[i] = firstBlock[32+i] ^ x
}
box = box[len(firstMessageBlock):]
out = out[len(firstMessageBlock):]
// Now decrypt the rest.
counter[8] = 1
salsa.XORKeyStream(out, box, &counter, &subKey)
return ret, true
}

39
vendor/golang.org/x/crypto/poly1305/bits_compat.go generated vendored Normal file
View File

@@ -0,0 +1,39 @@
// Copyright 2019 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
// +build !go1.13
package poly1305
// Generic fallbacks for the math/bits intrinsics, copied from
// src/math/bits/bits.go. They were added in Go 1.12, but Add64 and Sum64 had
// variable time fallbacks until Go 1.13.
func bitsAdd64(x, y, carry uint64) (sum, carryOut uint64) {
sum = x + y + carry
carryOut = ((x & y) | ((x | y) &^ sum)) >> 63
return
}
func bitsSub64(x, y, borrow uint64) (diff, borrowOut uint64) {
diff = x - y - borrow
borrowOut = ((^x & y) | (^(x ^ y) & diff)) >> 63
return
}
func bitsMul64(x, y uint64) (hi, lo uint64) {
const mask32 = 1<<32 - 1
x0 := x & mask32
x1 := x >> 32
y0 := y & mask32
y1 := y >> 32
w0 := x0 * y0
t := x1*y0 + w0>>32
w1 := t & mask32
w2 := t >> 32
w1 += x0 * y1
hi = x1*y1 + w2 + w1>>32
lo = x * y
return
}

21
vendor/golang.org/x/crypto/poly1305/bits_go1.13.go generated vendored Normal file
View File

@@ -0,0 +1,21 @@
// Copyright 2019 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
// +build go1.13
package poly1305
import "math/bits"
func bitsAdd64(x, y, carry uint64) (sum, carryOut uint64) {
return bits.Add64(x, y, carry)
}
func bitsSub64(x, y, borrow uint64) (diff, borrowOut uint64) {
return bits.Sub64(x, y, borrow)
}
func bitsMul64(x, y uint64) (hi, lo uint64) {
return bits.Mul64(x, y)
}

9
vendor/golang.org/x/crypto/poly1305/mac_noasm.go generated vendored Normal file
View File

@@ -0,0 +1,9 @@
// Copyright 2018 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
// +build !amd64,!ppc64le,!s390x gccgo purego
package poly1305
type mac struct{ macGeneric }

99
vendor/golang.org/x/crypto/poly1305/poly1305.go generated vendored Normal file
View File

@@ -0,0 +1,99 @@
// Copyright 2012 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
// Package poly1305 implements Poly1305 one-time message authentication code as
// specified in https://cr.yp.to/mac/poly1305-20050329.pdf.
//
// Poly1305 is a fast, one-time authentication function. It is infeasible for an
// attacker to generate an authenticator for a message without the key. However, a
// key must only be used for a single message. Authenticating two different
// messages with the same key allows an attacker to forge authenticators for other
// messages with the same key.
//
// Poly1305 was originally coupled with AES in order to make Poly1305-AES. AES was
// used with a fixed key in order to generate one-time keys from an nonce.
// However, in this package AES isn't used and the one-time key is specified
// directly.
package poly1305 // import "golang.org/x/crypto/poly1305"
import "crypto/subtle"
// TagSize is the size, in bytes, of a poly1305 authenticator.
const TagSize = 16
// Sum generates an authenticator for msg using a one-time key and puts the
// 16-byte result into out. Authenticating two different messages with the same
// key allows an attacker to forge messages at will.
func Sum(out *[16]byte, m []byte, key *[32]byte) {
h := New(key)
h.Write(m)
h.Sum(out[:0])
}
// Verify returns true if mac is a valid authenticator for m with the given key.
func Verify(mac *[16]byte, m []byte, key *[32]byte) bool {
var tmp [16]byte
Sum(&tmp, m, key)
return subtle.ConstantTimeCompare(tmp[:], mac[:]) == 1
}
// New returns a new MAC computing an authentication
// tag of all data written to it with the given key.
// This allows writing the message progressively instead
// of passing it as a single slice. Common users should use
// the Sum function instead.
//
// The key must be unique for each message, as authenticating
// two different messages with the same key allows an attacker
// to forge messages at will.
func New(key *[32]byte) *MAC {
m := &MAC{}
initialize(key, &m.macState)
return m
}
// MAC is an io.Writer computing an authentication tag
// of the data written to it.
//
// MAC cannot be used like common hash.Hash implementations,
// because using a poly1305 key twice breaks its security.
// Therefore writing data to a running MAC after calling
// Sum or Verify causes it to panic.
type MAC struct {
mac // platform-dependent implementation
finalized bool
}
// Size returns the number of bytes Sum will return.
func (h *MAC) Size() int { return TagSize }
// Write adds more data to the running message authentication code.
// It never returns an error.
//
// It must not be called after the first call of Sum or Verify.
func (h *MAC) Write(p []byte) (n int, err error) {
if h.finalized {
panic("poly1305: write to MAC after Sum or Verify")
}
return h.mac.Write(p)
}
// Sum computes the authenticator of all data written to the
// message authentication code.
func (h *MAC) Sum(b []byte) []byte {
var mac [TagSize]byte
h.mac.Sum(&mac)
h.finalized = true
return append(b, mac[:]...)
}
// Verify returns whether the authenticator of all data written to
// the message authentication code matches the expected value.
func (h *MAC) Verify(expected []byte) bool {
var mac [TagSize]byte
h.mac.Sum(&mac)
h.finalized = true
return subtle.ConstantTimeCompare(expected, mac[:]) == 1
}

47
vendor/golang.org/x/crypto/poly1305/sum_amd64.go generated vendored Normal file
View File

@@ -0,0 +1,47 @@
// Copyright 2012 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
// +build !gccgo,!purego
package poly1305
//go:noescape
func update(state *macState, msg []byte)
// mac is a wrapper for macGeneric that redirects calls that would have gone to
// updateGeneric to update.
//
// Its Write and Sum methods are otherwise identical to the macGeneric ones, but
// using function pointers would carry a major performance cost.
type mac struct{ macGeneric }
func (h *mac) Write(p []byte) (int, error) {
nn := len(p)
if h.offset > 0 {
n := copy(h.buffer[h.offset:], p)
if h.offset+n < TagSize {
h.offset += n
return nn, nil
}
p = p[n:]
h.offset = 0
update(&h.macState, h.buffer[:])
}
if n := len(p) - (len(p) % TagSize); n > 0 {
update(&h.macState, p[:n])
p = p[n:]
}
if len(p) > 0 {
h.offset += copy(h.buffer[h.offset:], p)
}
return nn, nil
}
func (h *mac) Sum(out *[16]byte) {
state := h.macState
if h.offset > 0 {
update(&state, h.buffer[:h.offset])
}
finalize(out, &state.h, &state.s)
}

108
vendor/golang.org/x/crypto/poly1305/sum_amd64.s generated vendored Normal file
View File

@@ -0,0 +1,108 @@
// Copyright 2012 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
// +build !gccgo,!purego
#include "textflag.h"
#define POLY1305_ADD(msg, h0, h1, h2) \
ADDQ 0(msg), h0; \
ADCQ 8(msg), h1; \
ADCQ $1, h2; \
LEAQ 16(msg), msg
#define POLY1305_MUL(h0, h1, h2, r0, r1, t0, t1, t2, t3) \
MOVQ r0, AX; \
MULQ h0; \
MOVQ AX, t0; \
MOVQ DX, t1; \
MOVQ r0, AX; \
MULQ h1; \
ADDQ AX, t1; \
ADCQ $0, DX; \
MOVQ r0, t2; \
IMULQ h2, t2; \
ADDQ DX, t2; \
\
MOVQ r1, AX; \
MULQ h0; \
ADDQ AX, t1; \
ADCQ $0, DX; \
MOVQ DX, h0; \
MOVQ r1, t3; \
IMULQ h2, t3; \
MOVQ r1, AX; \
MULQ h1; \
ADDQ AX, t2; \
ADCQ DX, t3; \
ADDQ h0, t2; \
ADCQ $0, t3; \
\
MOVQ t0, h0; \
MOVQ t1, h1; \
MOVQ t2, h2; \
ANDQ $3, h2; \
MOVQ t2, t0; \
ANDQ $0xFFFFFFFFFFFFFFFC, t0; \
ADDQ t0, h0; \
ADCQ t3, h1; \
ADCQ $0, h2; \
SHRQ $2, t3, t2; \
SHRQ $2, t3; \
ADDQ t2, h0; \
ADCQ t3, h1; \
ADCQ $0, h2
// func update(state *[7]uint64, msg []byte)
TEXT ·update(SB), $0-32
MOVQ state+0(FP), DI
MOVQ msg_base+8(FP), SI
MOVQ msg_len+16(FP), R15
MOVQ 0(DI), R8 // h0
MOVQ 8(DI), R9 // h1
MOVQ 16(DI), R10 // h2
MOVQ 24(DI), R11 // r0
MOVQ 32(DI), R12 // r1
CMPQ R15, $16
JB bytes_between_0_and_15
loop:
POLY1305_ADD(SI, R8, R9, R10)
multiply:
POLY1305_MUL(R8, R9, R10, R11, R12, BX, CX, R13, R14)
SUBQ $16, R15
CMPQ R15, $16
JAE loop
bytes_between_0_and_15:
TESTQ R15, R15
JZ done
MOVQ $1, BX
XORQ CX, CX
XORQ R13, R13
ADDQ R15, SI
flush_buffer:
SHLQ $8, BX, CX
SHLQ $8, BX
MOVB -1(SI), R13
XORQ R13, BX
DECQ SI
DECQ R15
JNZ flush_buffer
ADDQ BX, R8
ADCQ CX, R9
ADCQ $0, R10
MOVQ $16, R15
JMP multiply
done:
MOVQ R8, 0(DI)
MOVQ R9, 8(DI)
MOVQ R10, 16(DI)
RET

310
vendor/golang.org/x/crypto/poly1305/sum_generic.go generated vendored Normal file
View File

@@ -0,0 +1,310 @@
// Copyright 2018 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
// This file provides the generic implementation of Sum and MAC. Other files
// might provide optimized assembly implementations of some of this code.
package poly1305
import "encoding/binary"
// Poly1305 [RFC 7539] is a relatively simple algorithm: the authentication tag
// for a 64 bytes message is approximately
//
// s + m[0:16] * r⁴ + m[16:32] * r³ + m[32:48] * r² + m[48:64] * r mod 2¹³⁰ - 5
//
// for some secret r and s. It can be computed sequentially like
//
// for len(msg) > 0:
// h += read(msg, 16)
// h *= r
// h %= 2¹³⁰ - 5
// return h + s
//
// All the complexity is about doing performant constant-time math on numbers
// larger than any available numeric type.
func sumGeneric(out *[TagSize]byte, msg []byte, key *[32]byte) {
h := newMACGeneric(key)
h.Write(msg)
h.Sum(out)
}
func newMACGeneric(key *[32]byte) macGeneric {
m := macGeneric{}
initialize(key, &m.macState)
return m
}
// macState holds numbers in saturated 64-bit little-endian limbs. That is,
// the value of [x0, x1, x2] is x[0] + x[1] * 2⁶⁴ + x[2] * 2¹²⁸.
type macState struct {
// h is the main accumulator. It is to be interpreted modulo 2¹³⁰ - 5, but
// can grow larger during and after rounds. It must, however, remain below
// 2 * (2¹³⁰ - 5).
h [3]uint64
// r and s are the private key components.
r [2]uint64
s [2]uint64
}
type macGeneric struct {
macState
buffer [TagSize]byte
offset int
}
// Write splits the incoming message into TagSize chunks, and passes them to
// update. It buffers incomplete chunks.
func (h *macGeneric) Write(p []byte) (int, error) {
nn := len(p)
if h.offset > 0 {
n := copy(h.buffer[h.offset:], p)
if h.offset+n < TagSize {
h.offset += n
return nn, nil
}
p = p[n:]
h.offset = 0
updateGeneric(&h.macState, h.buffer[:])
}
if n := len(p) - (len(p) % TagSize); n > 0 {
updateGeneric(&h.macState, p[:n])
p = p[n:]
}
if len(p) > 0 {
h.offset += copy(h.buffer[h.offset:], p)
}
return nn, nil
}
// Sum flushes the last incomplete chunk from the buffer, if any, and generates
// the MAC output. It does not modify its state, in order to allow for multiple
// calls to Sum, even if no Write is allowed after Sum.
func (h *macGeneric) Sum(out *[TagSize]byte) {
state := h.macState
if h.offset > 0 {
updateGeneric(&state, h.buffer[:h.offset])
}
finalize(out, &state.h, &state.s)
}
// [rMask0, rMask1] is the specified Poly1305 clamping mask in little-endian. It
// clears some bits of the secret coefficient to make it possible to implement
// multiplication more efficiently.
const (
rMask0 = 0x0FFFFFFC0FFFFFFF
rMask1 = 0x0FFFFFFC0FFFFFFC
)
// initialize loads the 256-bit key into the two 128-bit secret values r and s.
func initialize(key *[32]byte, m *macState) {
m.r[0] = binary.LittleEndian.Uint64(key[0:8]) & rMask0
m.r[1] = binary.LittleEndian.Uint64(key[8:16]) & rMask1
m.s[0] = binary.LittleEndian.Uint64(key[16:24])
m.s[1] = binary.LittleEndian.Uint64(key[24:32])
}
// uint128 holds a 128-bit number as two 64-bit limbs, for use with the
// bits.Mul64 and bits.Add64 intrinsics.
type uint128 struct {
lo, hi uint64
}
func mul64(a, b uint64) uint128 {
hi, lo := bitsMul64(a, b)
return uint128{lo, hi}
}
func add128(a, b uint128) uint128 {
lo, c := bitsAdd64(a.lo, b.lo, 0)
hi, c := bitsAdd64(a.hi, b.hi, c)
if c != 0 {
panic("poly1305: unexpected overflow")
}
return uint128{lo, hi}
}
func shiftRightBy2(a uint128) uint128 {
a.lo = a.lo>>2 | (a.hi&3)<<62
a.hi = a.hi >> 2
return a
}
// updateGeneric absorbs msg into the state.h accumulator. For each chunk m of
// 128 bits of message, it computes
//
// h₊ = (h + m) * r mod 2¹³⁰ - 5
//
// If the msg length is not a multiple of TagSize, it assumes the last
// incomplete chunk is the final one.
func updateGeneric(state *macState, msg []byte) {
h0, h1, h2 := state.h[0], state.h[1], state.h[2]
r0, r1 := state.r[0], state.r[1]
for len(msg) > 0 {
var c uint64
// For the first step, h + m, we use a chain of bits.Add64 intrinsics.
// The resulting value of h might exceed 2¹³⁰ - 5, but will be partially
// reduced at the end of the multiplication below.
//
// The spec requires us to set a bit just above the message size, not to
// hide leading zeroes. For full chunks, that's 1 << 128, so we can just
// add 1 to the most significant (2¹²⁸) limb, h2.
if len(msg) >= TagSize {
h0, c = bitsAdd64(h0, binary.LittleEndian.Uint64(msg[0:8]), 0)
h1, c = bitsAdd64(h1, binary.LittleEndian.Uint64(msg[8:16]), c)
h2 += c + 1
msg = msg[TagSize:]
} else {
var buf [TagSize]byte
copy(buf[:], msg)
buf[len(msg)] = 1
h0, c = bitsAdd64(h0, binary.LittleEndian.Uint64(buf[0:8]), 0)
h1, c = bitsAdd64(h1, binary.LittleEndian.Uint64(buf[8:16]), c)
h2 += c
msg = nil
}
// Multiplication of big number limbs is similar to elementary school
// columnar multiplication. Instead of digits, there are 64-bit limbs.
//
// We are multiplying a 3 limbs number, h, by a 2 limbs number, r.
//
// h2 h1 h0 x
// r1 r0 =
// ----------------
// h2r0 h1r0 h0r0 <-- individual 128-bit products
// + h2r1 h1r1 h0r1
// ------------------------
// m3 m2 m1 m0 <-- result in 128-bit overlapping limbs
// ------------------------
// m3.hi m2.hi m1.hi m0.hi <-- carry propagation
// + m3.lo m2.lo m1.lo m0.lo
// -------------------------------
// t4 t3 t2 t1 t0 <-- final result in 64-bit limbs
//
// The main difference from pen-and-paper multiplication is that we do
// carry propagation in a separate step, as if we wrote two digit sums
// at first (the 128-bit limbs), and then carried the tens all at once.
h0r0 := mul64(h0, r0)
h1r0 := mul64(h1, r0)
h2r0 := mul64(h2, r0)
h0r1 := mul64(h0, r1)
h1r1 := mul64(h1, r1)
h2r1 := mul64(h2, r1)
// Since h2 is known to be at most 7 (5 + 1 + 1), and r0 and r1 have their
// top 4 bits cleared by rMask{0,1}, we know that their product is not going
// to overflow 64 bits, so we can ignore the high part of the products.
//
// This also means that the product doesn't have a fifth limb (t4).
if h2r0.hi != 0 {
panic("poly1305: unexpected overflow")
}
if h2r1.hi != 0 {
panic("poly1305: unexpected overflow")
}
m0 := h0r0
m1 := add128(h1r0, h0r1) // These two additions don't overflow thanks again
m2 := add128(h2r0, h1r1) // to the 4 masked bits at the top of r0 and r1.
m3 := h2r1
t0 := m0.lo
t1, c := bitsAdd64(m1.lo, m0.hi, 0)
t2, c := bitsAdd64(m2.lo, m1.hi, c)
t3, _ := bitsAdd64(m3.lo, m2.hi, c)
// Now we have the result as 4 64-bit limbs, and we need to reduce it
// modulo 2¹³⁰ - 5. The special shape of this Crandall prime lets us do
// a cheap partial reduction according to the reduction identity
//
// c * 2¹³⁰ + n = c * 5 + n mod 2¹³⁰ - 5
//
// because 2¹³⁰ = 5 mod 2¹³⁰ - 5. Partial reduction since the result is
// likely to be larger than 2¹³⁰ - 5, but still small enough to fit the
// assumptions we make about h in the rest of the code.
//
// See also https://speakerdeck.com/gtank/engineering-prime-numbers?slide=23
// We split the final result at the 2¹³⁰ mark into h and cc, the carry.
// Note that the carry bits are effectively shifted left by 2, in other
// words, cc = c * 4 for the c in the reduction identity.
h0, h1, h2 = t0, t1, t2&maskLow2Bits
cc := uint128{t2 & maskNotLow2Bits, t3}
// To add c * 5 to h, we first add cc = c * 4, and then add (cc >> 2) = c.
h0, c = bitsAdd64(h0, cc.lo, 0)
h1, c = bitsAdd64(h1, cc.hi, c)
h2 += c
cc = shiftRightBy2(cc)
h0, c = bitsAdd64(h0, cc.lo, 0)
h1, c = bitsAdd64(h1, cc.hi, c)
h2 += c
// h2 is at most 3 + 1 + 1 = 5, making the whole of h at most
//
// 5 * 2¹²⁸ + (2¹²⁸ - 1) = 6 * 2¹²⁸ - 1
}
state.h[0], state.h[1], state.h[2] = h0, h1, h2
}
const (
maskLow2Bits uint64 = 0x0000000000000003
maskNotLow2Bits uint64 = ^maskLow2Bits
)
// select64 returns x if v == 1 and y if v == 0, in constant time.
func select64(v, x, y uint64) uint64 { return ^(v-1)&x | (v-1)&y }
// [p0, p1, p2] is 2¹³⁰ - 5 in little endian order.
const (
p0 = 0xFFFFFFFFFFFFFFFB
p1 = 0xFFFFFFFFFFFFFFFF
p2 = 0x0000000000000003
)
// finalize completes the modular reduction of h and computes
//
// out = h + s mod 2¹²⁸
//
func finalize(out *[TagSize]byte, h *[3]uint64, s *[2]uint64) {
h0, h1, h2 := h[0], h[1], h[2]
// After the partial reduction in updateGeneric, h might be more than
// 2¹³⁰ - 5, but will be less than 2 * (2¹³⁰ - 5). To complete the reduction
// in constant time, we compute t = h - (2¹³⁰ - 5), and select h as the
// result if the subtraction underflows, and t otherwise.
hMinusP0, b := bitsSub64(h0, p0, 0)
hMinusP1, b := bitsSub64(h1, p1, b)
_, b = bitsSub64(h2, p2, b)
// h = h if h < p else h - p
h0 = select64(b, h0, hMinusP0)
h1 = select64(b, h1, hMinusP1)
// Finally, we compute the last Poly1305 step
//
// tag = h + s mod 2¹²⁸
//
// by just doing a wide addition with the 128 low bits of h and discarding
// the overflow.
h0, c := bitsAdd64(h0, s[0], 0)
h1, _ = bitsAdd64(h1, s[1], c)
binary.LittleEndian.PutUint64(out[0:8], h0)
binary.LittleEndian.PutUint64(out[8:16], h1)
}

47
vendor/golang.org/x/crypto/poly1305/sum_ppc64le.go generated vendored Normal file
View File

@@ -0,0 +1,47 @@
// Copyright 2019 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
// +build !gccgo,!purego
package poly1305
//go:noescape
func update(state *macState, msg []byte)
// mac is a wrapper for macGeneric that redirects calls that would have gone to
// updateGeneric to update.
//
// Its Write and Sum methods are otherwise identical to the macGeneric ones, but
// using function pointers would carry a major performance cost.
type mac struct{ macGeneric }
func (h *mac) Write(p []byte) (int, error) {
nn := len(p)
if h.offset > 0 {
n := copy(h.buffer[h.offset:], p)
if h.offset+n < TagSize {
h.offset += n
return nn, nil
}
p = p[n:]
h.offset = 0
update(&h.macState, h.buffer[:])
}
if n := len(p) - (len(p) % TagSize); n > 0 {
update(&h.macState, p[:n])
p = p[n:]
}
if len(p) > 0 {
h.offset += copy(h.buffer[h.offset:], p)
}
return nn, nil
}
func (h *mac) Sum(out *[16]byte) {
state := h.macState
if h.offset > 0 {
update(&state, h.buffer[:h.offset])
}
finalize(out, &state.h, &state.s)
}

181
vendor/golang.org/x/crypto/poly1305/sum_ppc64le.s generated vendored Normal file
View File

@@ -0,0 +1,181 @@
// Copyright 2019 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
// +build !gccgo,!purego
#include "textflag.h"
// This was ported from the amd64 implementation.
#define POLY1305_ADD(msg, h0, h1, h2, t0, t1, t2) \
MOVD (msg), t0; \
MOVD 8(msg), t1; \
MOVD $1, t2; \
ADDC t0, h0, h0; \
ADDE t1, h1, h1; \
ADDE t2, h2; \
ADD $16, msg
#define POLY1305_MUL(h0, h1, h2, r0, r1, t0, t1, t2, t3, t4, t5) \
MULLD r0, h0, t0; \
MULLD r0, h1, t4; \
MULHDU r0, h0, t1; \
MULHDU r0, h1, t5; \
ADDC t4, t1, t1; \
MULLD r0, h2, t2; \
ADDZE t5; \
MULHDU r1, h0, t4; \
MULLD r1, h0, h0; \
ADD t5, t2, t2; \
ADDC h0, t1, t1; \
MULLD h2, r1, t3; \
ADDZE t4, h0; \
MULHDU r1, h1, t5; \
MULLD r1, h1, t4; \
ADDC t4, t2, t2; \
ADDE t5, t3, t3; \
ADDC h0, t2, t2; \
MOVD $-4, t4; \
MOVD t0, h0; \
MOVD t1, h1; \
ADDZE t3; \
ANDCC $3, t2, h2; \
AND t2, t4, t0; \
ADDC t0, h0, h0; \
ADDE t3, h1, h1; \
SLD $62, t3, t4; \
SRD $2, t2; \
ADDZE h2; \
OR t4, t2, t2; \
SRD $2, t3; \
ADDC t2, h0, h0; \
ADDE t3, h1, h1; \
ADDZE h2
DATA ·poly1305Mask<>+0x00(SB)/8, $0x0FFFFFFC0FFFFFFF
DATA ·poly1305Mask<>+0x08(SB)/8, $0x0FFFFFFC0FFFFFFC
GLOBL ·poly1305Mask<>(SB), RODATA, $16
// func update(state *[7]uint64, msg []byte)
TEXT ·update(SB), $0-32
MOVD state+0(FP), R3
MOVD msg_base+8(FP), R4
MOVD msg_len+16(FP), R5
MOVD 0(R3), R8 // h0
MOVD 8(R3), R9 // h1
MOVD 16(R3), R10 // h2
MOVD 24(R3), R11 // r0
MOVD 32(R3), R12 // r1
CMP R5, $16
BLT bytes_between_0_and_15
loop:
POLY1305_ADD(R4, R8, R9, R10, R20, R21, R22)
multiply:
POLY1305_MUL(R8, R9, R10, R11, R12, R16, R17, R18, R14, R20, R21)
ADD $-16, R5
CMP R5, $16
BGE loop
bytes_between_0_and_15:
CMP $0, R5
BEQ done
MOVD $0, R16 // h0
MOVD $0, R17 // h1
flush_buffer:
CMP R5, $8
BLE just1
MOVD $8, R21
SUB R21, R5, R21
// Greater than 8 -- load the rightmost remaining bytes in msg
// and put into R17 (h1)
MOVD (R4)(R21), R17
MOVD $16, R22
// Find the offset to those bytes
SUB R5, R22, R22
SLD $3, R22
// Shift to get only the bytes in msg
SRD R22, R17, R17
// Put 1 at high end
MOVD $1, R23
SLD $3, R21
SLD R21, R23, R23
OR R23, R17, R17
// Remainder is 8
MOVD $8, R5
just1:
CMP R5, $8
BLT less8
// Exactly 8
MOVD (R4), R16
CMP $0, R17
// Check if we've already set R17; if not
// set 1 to indicate end of msg.
BNE carry
MOVD $1, R17
BR carry
less8:
MOVD $0, R16 // h0
MOVD $0, R22 // shift count
CMP R5, $4
BLT less4
MOVWZ (R4), R16
ADD $4, R4
ADD $-4, R5
MOVD $32, R22
less4:
CMP R5, $2
BLT less2
MOVHZ (R4), R21
SLD R22, R21, R21
OR R16, R21, R16
ADD $16, R22
ADD $-2, R5
ADD $2, R4
less2:
CMP $0, R5
BEQ insert1
MOVBZ (R4), R21
SLD R22, R21, R21
OR R16, R21, R16
ADD $8, R22
insert1:
// Insert 1 at end of msg
MOVD $1, R21
SLD R22, R21, R21
OR R16, R21, R16
carry:
// Add new values to h0, h1, h2
ADDC R16, R8
ADDE R17, R9
ADDE $0, R10
MOVD $16, R5
ADD R5, R4
BR multiply
done:
// Save h0, h1, h2 in state
MOVD R8, 0(R3)
MOVD R9, 8(R3)
MOVD R10, 16(R3)
RET

75
vendor/golang.org/x/crypto/poly1305/sum_s390x.go generated vendored Normal file
View File

@@ -0,0 +1,75 @@
// Copyright 2018 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
// +build !gccgo,!purego
package poly1305
import (
"golang.org/x/sys/cpu"
)
// updateVX is an assembly implementation of Poly1305 that uses vector
// instructions. It must only be called if the vector facility (vx) is
// available.
//go:noescape
func updateVX(state *macState, msg []byte)
// mac is a replacement for macGeneric that uses a larger buffer and redirects
// calls that would have gone to updateGeneric to updateVX if the vector
// facility is installed.
//
// A larger buffer is required for good performance because the vector
// implementation has a higher fixed cost per call than the generic
// implementation.
type mac struct {
macState
buffer [16 * TagSize]byte // size must be a multiple of block size (16)
offset int
}
func (h *mac) Write(p []byte) (int, error) {
nn := len(p)
if h.offset > 0 {
n := copy(h.buffer[h.offset:], p)
if h.offset+n < len(h.buffer) {
h.offset += n
return nn, nil
}
p = p[n:]
h.offset = 0
if cpu.S390X.HasVX {
updateVX(&h.macState, h.buffer[:])
} else {
updateGeneric(&h.macState, h.buffer[:])
}
}
tail := len(p) % len(h.buffer) // number of bytes to copy into buffer
body := len(p) - tail // number of bytes to process now
if body > 0 {
if cpu.S390X.HasVX {
updateVX(&h.macState, p[:body])
} else {
updateGeneric(&h.macState, p[:body])
}
}
h.offset = copy(h.buffer[:], p[body:]) // copy tail bytes - can be 0
return nn, nil
}
func (h *mac) Sum(out *[TagSize]byte) {
state := h.macState
remainder := h.buffer[:h.offset]
// Use the generic implementation if we have 2 or fewer blocks left
// to sum. The vector implementation has a higher startup time.
if cpu.S390X.HasVX && len(remainder) > 2*TagSize {
updateVX(&state, remainder)
} else if len(remainder) > 0 {
updateGeneric(&state, remainder)
}
finalize(out, &state.h, &state.s)
}

503
vendor/golang.org/x/crypto/poly1305/sum_s390x.s generated vendored Normal file
View File

@@ -0,0 +1,503 @@
// Copyright 2018 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
// +build !gccgo,!purego
#include "textflag.h"
// This implementation of Poly1305 uses the vector facility (vx)
// to process up to 2 blocks (32 bytes) per iteration using an
// algorithm based on the one described in:
//
// NEON crypto, Daniel J. Bernstein & Peter Schwabe
// https://cryptojedi.org/papers/neoncrypto-20120320.pdf
//
// This algorithm uses 5 26-bit limbs to represent a 130-bit
// value. These limbs are, for the most part, zero extended and
// placed into 64-bit vector register elements. Each vector
// register is 128-bits wide and so holds 2 of these elements.
// Using 26-bit limbs allows us plenty of headroom to accomodate
// accumulations before and after multiplication without
// overflowing either 32-bits (before multiplication) or 64-bits
// (after multiplication).
//
// In order to parallelise the operations required to calculate
// the sum we use two separate accumulators and then sum those
// in an extra final step. For compatibility with the generic
// implementation we perform this summation at the end of every
// updateVX call.
//
// To use two accumulators we must multiply the message blocks
// by r² rather than r. Only the final message block should be
// multiplied by r.
//
// Example:
//
// We want to calculate the sum (h) for a 64 byte message (m):
//
// h = m[0:16]r + m[16:32]r³ + m[32:48]r² + m[48:64]r
//
// To do this we split the calculation into the even indices
// and odd indices of the message. These form our SIMD 'lanes':
//
// h = m[ 0:16]r + m[32:48]r² + <- lane 0
// m[16:32]r³ + m[48:64]r <- lane 1
//
// To calculate this iteratively we refactor so that both lanes
// are written in terms of r² and r:
//
// h = (m[ 0:16]r² + m[32:48])r² + <- lane 0
// (m[16:32]r² + m[48:64])r <- lane 1
// ^ ^
// | coefficients for second iteration
// coefficients for first iteration
//
// So in this case we would have two iterations. In the first
// both lanes are multiplied by r². In the second only the
// first lane is multiplied by r² and the second lane is
// instead multiplied by r. This gives use the odd and even
// powers of r that we need from the original equation.
//
// Notation:
//
// h - accumulator
// r - key
// m - message
//
// [a, b] - SIMD register holding two 64-bit values
// [a, b, c, d] - SIMD register holding four 32-bit values
// x[n] - limb n of variable x with bit width i
//
// Limbs are expressed in little endian order, so for 26-bit
// limbs x[4] will be the most significant limb and x[0]
// will be the least significant limb.
// masking constants
#define MOD24 V0 // [0x0000000000ffffff, 0x0000000000ffffff] - mask low 24-bits
#define MOD26 V1 // [0x0000000003ffffff, 0x0000000003ffffff] - mask low 26-bits
// expansion constants (see EXPAND macro)
#define EX0 V2
#define EX1 V3
#define EX2 V4
// key (r², r or 1 depending on context)
#define R_0 V5
#define R_1 V6
#define R_2 V7
#define R_3 V8
#define R_4 V9
// precalculated coefficients (5r², 5r or 0 depending on context)
#define R5_1 V10
#define R5_2 V11
#define R5_3 V12
#define R5_4 V13
// message block (m)
#define M_0 V14
#define M_1 V15
#define M_2 V16
#define M_3 V17
#define M_4 V18
// accumulator (h)
#define H_0 V19
#define H_1 V20
#define H_2 V21
#define H_3 V22
#define H_4 V23
// temporary registers (for short-lived values)
#define T_0 V24
#define T_1 V25
#define T_2 V26
#define T_3 V27
#define T_4 V28
GLOBL ·constants<>(SB), RODATA, $0x30
// EX0
DATA ·constants<>+0x00(SB)/8, $0x0006050403020100
DATA ·constants<>+0x08(SB)/8, $0x1016151413121110
// EX1
DATA ·constants<>+0x10(SB)/8, $0x060c0b0a09080706
DATA ·constants<>+0x18(SB)/8, $0x161c1b1a19181716
// EX2
DATA ·constants<>+0x20(SB)/8, $0x0d0d0d0d0d0f0e0d
DATA ·constants<>+0x28(SB)/8, $0x1d1d1d1d1d1f1e1d
// MULTIPLY multiplies each lane of f and g, partially reduced
// modulo 2¹³ - 5. The result, h, consists of partial products
// in each lane that need to be reduced further to produce the
// final result.
//
// h = (fg) % 2¹³ + (5fg) / 2¹³
//
// Note that the multiplication by 5 of the high bits is
// achieved by precalculating the multiplication of four of the
// g coefficients by 5. These are g51-g54.
#define MULTIPLY(f0, f1, f2, f3, f4, g0, g1, g2, g3, g4, g51, g52, g53, g54, h0, h1, h2, h3, h4) \
VMLOF f0, g0, h0 \
VMLOF f0, g3, h3 \
VMLOF f0, g1, h1 \
VMLOF f0, g4, h4 \
VMLOF f0, g2, h2 \
VMLOF f1, g54, T_0 \
VMLOF f1, g2, T_3 \
VMLOF f1, g0, T_1 \
VMLOF f1, g3, T_4 \
VMLOF f1, g1, T_2 \
VMALOF f2, g53, h0, h0 \
VMALOF f2, g1, h3, h3 \
VMALOF f2, g54, h1, h1 \
VMALOF f2, g2, h4, h4 \
VMALOF f2, g0, h2, h2 \
VMALOF f3, g52, T_0, T_0 \
VMALOF f3, g0, T_3, T_3 \
VMALOF f3, g53, T_1, T_1 \
VMALOF f3, g1, T_4, T_4 \
VMALOF f3, g54, T_2, T_2 \
VMALOF f4, g51, h0, h0 \
VMALOF f4, g54, h3, h3 \
VMALOF f4, g52, h1, h1 \
VMALOF f4, g0, h4, h4 \
VMALOF f4, g53, h2, h2 \
VAG T_0, h0, h0 \
VAG T_3, h3, h3 \
VAG T_1, h1, h1 \
VAG T_4, h4, h4 \
VAG T_2, h2, h2
// REDUCE performs the following carry operations in four
// stages, as specified in Bernstein & Schwabe:
//
// 1: h[0]->h[1] h[3]->h[4]
// 2: h[1]->h[2] h[4]->h[0]
// 3: h[0]->h[1] h[2]->h[3]
// 4: h[3]->h[4]
//
// The result is that all of the limbs are limited to 26-bits
// except for h[1] and h[4] which are limited to 27-bits.
//
// Note that although each limb is aligned at 26-bit intervals
// they may contain values that exceed 2² - 1, hence the need
// to carry the excess bits in each limb.
#define REDUCE(h0, h1, h2, h3, h4) \
VESRLG $26, h0, T_0 \
VESRLG $26, h3, T_1 \
VN MOD26, h0, h0 \
VN MOD26, h3, h3 \
VAG T_0, h1, h1 \
VAG T_1, h4, h4 \
VESRLG $26, h1, T_2 \
VESRLG $26, h4, T_3 \
VN MOD26, h1, h1 \
VN MOD26, h4, h4 \
VESLG $2, T_3, T_4 \
VAG T_3, T_4, T_4 \
VAG T_2, h2, h2 \
VAG T_4, h0, h0 \
VESRLG $26, h2, T_0 \
VESRLG $26, h0, T_1 \
VN MOD26, h2, h2 \
VN MOD26, h0, h0 \
VAG T_0, h3, h3 \
VAG T_1, h1, h1 \
VESRLG $26, h3, T_2 \
VN MOD26, h3, h3 \
VAG T_2, h4, h4
// EXPAND splits the 128-bit little-endian values in0 and in1
// into 26-bit big-endian limbs and places the results into
// the first and second lane of d[0:4] respectively.
//
// The EX0, EX1 and EX2 constants are arrays of byte indices
// for permutation. The permutation both reverses the bytes
// in the input and ensures the bytes are copied into the
// destination limb ready to be shifted into their final
// position.
#define EXPAND(in0, in1, d0, d1, d2, d3, d4) \
VPERM in0, in1, EX0, d0 \
VPERM in0, in1, EX1, d2 \
VPERM in0, in1, EX2, d4 \
VESRLG $26, d0, d1 \
VESRLG $30, d2, d3 \
VESRLG $4, d2, d2 \
VN MOD26, d0, d0 \ // [in0[0], in1[0]]
VN MOD26, d3, d3 \ // [in0[3], in1[3]]
VN MOD26, d1, d1 \ // [in0[1], in1[1]]
VN MOD24, d4, d4 \ // [in0[4], in1[4]]
VN MOD26, d2, d2 // [in0[2], in1[2]]
// func updateVX(state *macState, msg []byte)
TEXT ·updateVX(SB), NOSPLIT, $0
MOVD state+0(FP), R1
LMG msg+8(FP), R2, R3 // R2=msg_base, R3=msg_len
// load EX0, EX1 and EX2
MOVD $·constants<>(SB), R5
VLM (R5), EX0, EX2
// generate masks
VGMG $(64-24), $63, MOD24 // [0x00ffffff, 0x00ffffff]
VGMG $(64-26), $63, MOD26 // [0x03ffffff, 0x03ffffff]
// load h (accumulator) and r (key) from state
VZERO T_1 // [0, 0]
VL 0(R1), T_0 // [h[0], h[1]]
VLEG $0, 16(R1), T_1 // [h[2], 0]
VL 24(R1), T_2 // [r[0], r[1]]
VPDI $0, T_0, T_2, T_3 // [h[0], r[0]]
VPDI $5, T_0, T_2, T_4 // [h[1], r[1]]
// unpack h and r into 26-bit limbs
// note: h[2] may have the low 3 bits set, so h[4] is a 27-bit value
VN MOD26, T_3, H_0 // [h[0], r[0]]
VZERO H_1 // [0, 0]
VZERO H_3 // [0, 0]
VGMG $(64-12-14), $(63-12), T_0 // [0x03fff000, 0x03fff000] - 26-bit mask with low 12 bits masked out
VESLG $24, T_1, T_1 // [h[2]<<24, 0]
VERIMG $-26&63, T_3, MOD26, H_1 // [h[1], r[1]]
VESRLG $+52&63, T_3, H_2 // [h[2], r[2]] - low 12 bits only
VERIMG $-14&63, T_4, MOD26, H_3 // [h[1], r[1]]
VESRLG $40, T_4, H_4 // [h[4], r[4]] - low 24 bits only
VERIMG $+12&63, T_4, T_0, H_2 // [h[2], r[2]] - complete
VO T_1, H_4, H_4 // [h[4], r[4]] - complete
// replicate r across all 4 vector elements
VREPF $3, H_0, R_0 // [r[0], r[0], r[0], r[0]]
VREPF $3, H_1, R_1 // [r[1], r[1], r[1], r[1]]
VREPF $3, H_2, R_2 // [r[2], r[2], r[2], r[2]]
VREPF $3, H_3, R_3 // [r[3], r[3], r[3], r[3]]
VREPF $3, H_4, R_4 // [r[4], r[4], r[4], r[4]]
// zero out lane 1 of h
VLEIG $1, $0, H_0 // [h[0], 0]
VLEIG $1, $0, H_1 // [h[1], 0]
VLEIG $1, $0, H_2 // [h[2], 0]
VLEIG $1, $0, H_3 // [h[3], 0]
VLEIG $1, $0, H_4 // [h[4], 0]
// calculate 5r (ignore least significant limb)
VREPIF $5, T_0
VMLF T_0, R_1, R5_1 // [5r[1], 5r[1], 5r[1], 5r[1]]
VMLF T_0, R_2, R5_2 // [5r[2], 5r[2], 5r[2], 5r[2]]
VMLF T_0, R_3, R5_3 // [5r[3], 5r[3], 5r[3], 5r[3]]
VMLF T_0, R_4, R5_4 // [5r[4], 5r[4], 5r[4], 5r[4]]
// skip r² calculation if we are only calculating one block
CMPBLE R3, $16, skip
// calculate r²
MULTIPLY(R_0, R_1, R_2, R_3, R_4, R_0, R_1, R_2, R_3, R_4, R5_1, R5_2, R5_3, R5_4, M_0, M_1, M_2, M_3, M_4)
REDUCE(M_0, M_1, M_2, M_3, M_4)
VGBM $0x0f0f, T_0
VERIMG $0, M_0, T_0, R_0 // [r[0], r²[0], r[0], r²[0]]
VERIMG $0, M_1, T_0, R_1 // [r[1], r²[1], r[1], r²[1]]
VERIMG $0, M_2, T_0, R_2 // [r[2], r²[2], r[2], r²[2]]
VERIMG $0, M_3, T_0, R_3 // [r[3], r²[3], r[3], r²[3]]
VERIMG $0, M_4, T_0, R_4 // [r[4], r²[4], r[4], r²[4]]
// calculate 5r² (ignore least significant limb)
VREPIF $5, T_0
VMLF T_0, R_1, R5_1 // [5r[1], 5r²[1], 5r[1], 5r²[1]]
VMLF T_0, R_2, R5_2 // [5r[2], 5r²[2], 5r[2], 5r²[2]]
VMLF T_0, R_3, R5_3 // [5r[3], 5r²[3], 5r[3], 5r²[3]]
VMLF T_0, R_4, R5_4 // [5r[4], 5r²[4], 5r[4], 5r²[4]]
loop:
CMPBLE R3, $32, b2 // 2 or fewer blocks remaining, need to change key coefficients
// load next 2 blocks from message
VLM (R2), T_0, T_1
// update message slice
SUB $32, R3
MOVD $32(R2), R2
// unpack message blocks into 26-bit big-endian limbs
EXPAND(T_0, T_1, M_0, M_1, M_2, M_3, M_4)
// add 2¹² to each message block value
VLEIB $4, $1, M_4
VLEIB $12, $1, M_4
multiply:
// accumulate the incoming message
VAG H_0, M_0, M_0
VAG H_3, M_3, M_3
VAG H_1, M_1, M_1
VAG H_4, M_4, M_4
VAG H_2, M_2, M_2
// multiply the accumulator by the key coefficient
MULTIPLY(M_0, M_1, M_2, M_3, M_4, R_0, R_1, R_2, R_3, R_4, R5_1, R5_2, R5_3, R5_4, H_0, H_1, H_2, H_3, H_4)
// carry and partially reduce the partial products
REDUCE(H_0, H_1, H_2, H_3, H_4)
CMPBNE R3, $0, loop
finish:
// sum lane 0 and lane 1 and put the result in lane 1
VZERO T_0
VSUMQG H_0, T_0, H_0
VSUMQG H_3, T_0, H_3
VSUMQG H_1, T_0, H_1
VSUMQG H_4, T_0, H_4
VSUMQG H_2, T_0, H_2
// reduce again after summation
// TODO(mundaym): there might be a more efficient way to do this
// now that we only have 1 active lane. For example, we could
// simultaneously pack the values as we reduce them.
REDUCE(H_0, H_1, H_2, H_3, H_4)
// carry h[1] through to h[4] so that only h[4] can exceed 2² - 1
// TODO(mundaym): in testing this final carry was unnecessary.
// Needs a proof before it can be removed though.
VESRLG $26, H_1, T_1
VN MOD26, H_1, H_1
VAQ T_1, H_2, H_2
VESRLG $26, H_2, T_2
VN MOD26, H_2, H_2
VAQ T_2, H_3, H_3
VESRLG $26, H_3, T_3
VN MOD26, H_3, H_3
VAQ T_3, H_4, H_4
// h is now < 2(2¹³ - 5)
// Pack each lane in h[0:4] into h[0:1].
VESLG $26, H_1, H_1
VESLG $26, H_3, H_3
VO H_0, H_1, H_0
VO H_2, H_3, H_2
VESLG $4, H_2, H_2
VLEIB $7, $48, H_1
VSLB H_1, H_2, H_2
VO H_0, H_2, H_0
VLEIB $7, $104, H_1
VSLB H_1, H_4, H_3
VO H_3, H_0, H_0
VLEIB $7, $24, H_1
VSRLB H_1, H_4, H_1
// update state
VSTEG $1, H_0, 0(R1)
VSTEG $0, H_0, 8(R1)
VSTEG $1, H_1, 16(R1)
RET
b2: // 2 or fewer blocks remaining
CMPBLE R3, $16, b1
// Load the 2 remaining blocks (17-32 bytes remaining).
MOVD $-17(R3), R0 // index of final byte to load modulo 16
VL (R2), T_0 // load full 16 byte block
VLL R0, 16(R2), T_1 // load final (possibly partial) block and pad with zeros to 16 bytes
// The Poly1305 algorithm requires that a 1 bit be appended to
// each message block. If the final block is less than 16 bytes
// long then it is easiest to insert the 1 before the message
// block is split into 26-bit limbs. If, on the other hand, the
// final message block is 16 bytes long then we append the 1 bit
// after expansion as normal.
MOVBZ $1, R0
MOVD $-16(R3), R3 // index of byte in last block to insert 1 at (could be 16)
CMPBEQ R3, $16, 2(PC) // skip the insertion if the final block is 16 bytes long
VLVGB R3, R0, T_1 // insert 1 into the byte at index R3
// Split both blocks into 26-bit limbs in the appropriate lanes.
EXPAND(T_0, T_1, M_0, M_1, M_2, M_3, M_4)
// Append a 1 byte to the end of the second to last block.
VLEIB $4, $1, M_4
// Append a 1 byte to the end of the last block only if it is a
// full 16 byte block.
CMPBNE R3, $16, 2(PC)
VLEIB $12, $1, M_4
// Finally, set up the coefficients for the final multiplication.
// We have previously saved r and 5r in the 32-bit even indexes
// of the R_[0-4] and R5_[1-4] coefficient registers.
//
// We want lane 0 to be multiplied by r² so that can be kept the
// same. We want lane 1 to be multiplied by r so we need to move
// the saved r value into the 32-bit odd index in lane 1 by
// rotating the 64-bit lane by 32.
VGBM $0x00ff, T_0 // [0, 0xffffffffffffffff] - mask lane 1 only
VERIMG $32, R_0, T_0, R_0 // [_, r²[0], _, r[0]]
VERIMG $32, R_1, T_0, R_1 // [_, r²[1], _, r[1]]
VERIMG $32, R_2, T_0, R_2 // [_, r²[2], _, r[2]]
VERIMG $32, R_3, T_0, R_3 // [_, r²[3], _, r[3]]
VERIMG $32, R_4, T_0, R_4 // [_, r²[4], _, r[4]]
VERIMG $32, R5_1, T_0, R5_1 // [_, 5r²[1], _, 5r[1]]
VERIMG $32, R5_2, T_0, R5_2 // [_, 5r²[2], _, 5r[2]]
VERIMG $32, R5_3, T_0, R5_3 // [_, 5r²[3], _, 5r[3]]
VERIMG $32, R5_4, T_0, R5_4 // [_, 5r²[4], _, 5r[4]]
MOVD $0, R3
BR multiply
skip:
CMPBEQ R3, $0, finish
b1: // 1 block remaining
// Load the final block (1-16 bytes). This will be placed into
// lane 0.
MOVD $-1(R3), R0
VLL R0, (R2), T_0 // pad to 16 bytes with zeros
// The Poly1305 algorithm requires that a 1 bit be appended to
// each message block. If the final block is less than 16 bytes
// long then it is easiest to insert the 1 before the message
// block is split into 26-bit limbs. If, on the other hand, the
// final message block is 16 bytes long then we append the 1 bit
// after expansion as normal.
MOVBZ $1, R0
CMPBEQ R3, $16, 2(PC)
VLVGB R3, R0, T_0
// Set the message block in lane 1 to the value 0 so that it
// can be accumulated without affecting the final result.
VZERO T_1
// Split the final message block into 26-bit limbs in lane 0.
// Lane 1 will be contain 0.
EXPAND(T_0, T_1, M_0, M_1, M_2, M_3, M_4)
// Append a 1 byte to the end of the last block only if it is a
// full 16 byte block.
CMPBNE R3, $16, 2(PC)
VLEIB $4, $1, M_4
// We have previously saved r and 5r in the 32-bit even indexes
// of the R_[0-4] and R5_[1-4] coefficient registers.
//
// We want lane 0 to be multiplied by r so we need to move the
// saved r value into the 32-bit odd index in lane 0. We want
// lane 1 to be set to the value 1. This makes multiplication
// a no-op. We do this by setting lane 1 in every register to 0
// and then just setting the 32-bit index 3 in R_0 to 1.
VZERO T_0
MOVD $0, R0
MOVD $0x10111213, R12
VLVGP R12, R0, T_1 // [_, 0x10111213, _, 0x00000000]
VPERM T_0, R_0, T_1, R_0 // [_, r[0], _, 0]
VPERM T_0, R_1, T_1, R_1 // [_, r[1], _, 0]
VPERM T_0, R_2, T_1, R_2 // [_, r[2], _, 0]
VPERM T_0, R_3, T_1, R_3 // [_, r[3], _, 0]
VPERM T_0, R_4, T_1, R_4 // [_, r[4], _, 0]
VPERM T_0, R5_1, T_1, R5_1 // [_, 5r[1], _, 0]
VPERM T_0, R5_2, T_1, R5_2 // [_, 5r[2], _, 0]
VPERM T_0, R5_3, T_1, R5_3 // [_, 5r[3], _, 0]
VPERM T_0, R5_4, T_1, R5_4 // [_, 5r[4], _, 0]
// Set the value of lane 1 to be 1.
VLEIF $3, $1, R_0 // [_, r[0], _, 1]
MOVD $0, R3
BR multiply

144
vendor/golang.org/x/crypto/salsa20/salsa/hsalsa20.go generated vendored Normal file
View File

@@ -0,0 +1,144 @@
// Copyright 2012 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
// Package salsa provides low-level access to functions in the Salsa family.
package salsa // import "golang.org/x/crypto/salsa20/salsa"
// Sigma is the Salsa20 constant for 256-bit keys.
var Sigma = [16]byte{'e', 'x', 'p', 'a', 'n', 'd', ' ', '3', '2', '-', 'b', 'y', 't', 'e', ' ', 'k'}
// HSalsa20 applies the HSalsa20 core function to a 16-byte input in, 32-byte
// key k, and 16-byte constant c, and puts the result into the 32-byte array
// out.
func HSalsa20(out *[32]byte, in *[16]byte, k *[32]byte, c *[16]byte) {
x0 := uint32(c[0]) | uint32(c[1])<<8 | uint32(c[2])<<16 | uint32(c[3])<<24
x1 := uint32(k[0]) | uint32(k[1])<<8 | uint32(k[2])<<16 | uint32(k[3])<<24
x2 := uint32(k[4]) | uint32(k[5])<<8 | uint32(k[6])<<16 | uint32(k[7])<<24
x3 := uint32(k[8]) | uint32(k[9])<<8 | uint32(k[10])<<16 | uint32(k[11])<<24
x4 := uint32(k[12]) | uint32(k[13])<<8 | uint32(k[14])<<16 | uint32(k[15])<<24
x5 := uint32(c[4]) | uint32(c[5])<<8 | uint32(c[6])<<16 | uint32(c[7])<<24
x6 := uint32(in[0]) | uint32(in[1])<<8 | uint32(in[2])<<16 | uint32(in[3])<<24
x7 := uint32(in[4]) | uint32(in[5])<<8 | uint32(in[6])<<16 | uint32(in[7])<<24
x8 := uint32(in[8]) | uint32(in[9])<<8 | uint32(in[10])<<16 | uint32(in[11])<<24
x9 := uint32(in[12]) | uint32(in[13])<<8 | uint32(in[14])<<16 | uint32(in[15])<<24
x10 := uint32(c[8]) | uint32(c[9])<<8 | uint32(c[10])<<16 | uint32(c[11])<<24
x11 := uint32(k[16]) | uint32(k[17])<<8 | uint32(k[18])<<16 | uint32(k[19])<<24
x12 := uint32(k[20]) | uint32(k[21])<<8 | uint32(k[22])<<16 | uint32(k[23])<<24
x13 := uint32(k[24]) | uint32(k[25])<<8 | uint32(k[26])<<16 | uint32(k[27])<<24
x14 := uint32(k[28]) | uint32(k[29])<<8 | uint32(k[30])<<16 | uint32(k[31])<<24
x15 := uint32(c[12]) | uint32(c[13])<<8 | uint32(c[14])<<16 | uint32(c[15])<<24
for i := 0; i < 20; i += 2 {
u := x0 + x12
x4 ^= u<<7 | u>>(32-7)
u = x4 + x0
x8 ^= u<<9 | u>>(32-9)
u = x8 + x4
x12 ^= u<<13 | u>>(32-13)
u = x12 + x8
x0 ^= u<<18 | u>>(32-18)
u = x5 + x1
x9 ^= u<<7 | u>>(32-7)
u = x9 + x5
x13 ^= u<<9 | u>>(32-9)
u = x13 + x9
x1 ^= u<<13 | u>>(32-13)
u = x1 + x13
x5 ^= u<<18 | u>>(32-18)
u = x10 + x6
x14 ^= u<<7 | u>>(32-7)
u = x14 + x10
x2 ^= u<<9 | u>>(32-9)
u = x2 + x14
x6 ^= u<<13 | u>>(32-13)
u = x6 + x2
x10 ^= u<<18 | u>>(32-18)
u = x15 + x11
x3 ^= u<<7 | u>>(32-7)
u = x3 + x15
x7 ^= u<<9 | u>>(32-9)
u = x7 + x3
x11 ^= u<<13 | u>>(32-13)
u = x11 + x7
x15 ^= u<<18 | u>>(32-18)
u = x0 + x3
x1 ^= u<<7 | u>>(32-7)
u = x1 + x0
x2 ^= u<<9 | u>>(32-9)
u = x2 + x1
x3 ^= u<<13 | u>>(32-13)
u = x3 + x2
x0 ^= u<<18 | u>>(32-18)
u = x5 + x4
x6 ^= u<<7 | u>>(32-7)
u = x6 + x5
x7 ^= u<<9 | u>>(32-9)
u = x7 + x6
x4 ^= u<<13 | u>>(32-13)
u = x4 + x7
x5 ^= u<<18 | u>>(32-18)
u = x10 + x9
x11 ^= u<<7 | u>>(32-7)
u = x11 + x10
x8 ^= u<<9 | u>>(32-9)
u = x8 + x11
x9 ^= u<<13 | u>>(32-13)
u = x9 + x8
x10 ^= u<<18 | u>>(32-18)
u = x15 + x14
x12 ^= u<<7 | u>>(32-7)
u = x12 + x15
x13 ^= u<<9 | u>>(32-9)
u = x13 + x12
x14 ^= u<<13 | u>>(32-13)
u = x14 + x13
x15 ^= u<<18 | u>>(32-18)
}
out[0] = byte(x0)
out[1] = byte(x0 >> 8)
out[2] = byte(x0 >> 16)
out[3] = byte(x0 >> 24)
out[4] = byte(x5)
out[5] = byte(x5 >> 8)
out[6] = byte(x5 >> 16)
out[7] = byte(x5 >> 24)
out[8] = byte(x10)
out[9] = byte(x10 >> 8)
out[10] = byte(x10 >> 16)
out[11] = byte(x10 >> 24)
out[12] = byte(x15)
out[13] = byte(x15 >> 8)
out[14] = byte(x15 >> 16)
out[15] = byte(x15 >> 24)
out[16] = byte(x6)
out[17] = byte(x6 >> 8)
out[18] = byte(x6 >> 16)
out[19] = byte(x6 >> 24)
out[20] = byte(x7)
out[21] = byte(x7 >> 8)
out[22] = byte(x7 >> 16)
out[23] = byte(x7 >> 24)
out[24] = byte(x8)
out[25] = byte(x8 >> 8)
out[26] = byte(x8 >> 16)
out[27] = byte(x8 >> 24)
out[28] = byte(x9)
out[29] = byte(x9 >> 8)
out[30] = byte(x9 >> 16)
out[31] = byte(x9 >> 24)
}

199
vendor/golang.org/x/crypto/salsa20/salsa/salsa208.go generated vendored Normal file
View File

@@ -0,0 +1,199 @@
// Copyright 2012 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
package salsa
// Core208 applies the Salsa20/8 core function to the 64-byte array in and puts
// the result into the 64-byte array out. The input and output may be the same array.
func Core208(out *[64]byte, in *[64]byte) {
j0 := uint32(in[0]) | uint32(in[1])<<8 | uint32(in[2])<<16 | uint32(in[3])<<24
j1 := uint32(in[4]) | uint32(in[5])<<8 | uint32(in[6])<<16 | uint32(in[7])<<24
j2 := uint32(in[8]) | uint32(in[9])<<8 | uint32(in[10])<<16 | uint32(in[11])<<24
j3 := uint32(in[12]) | uint32(in[13])<<8 | uint32(in[14])<<16 | uint32(in[15])<<24
j4 := uint32(in[16]) | uint32(in[17])<<8 | uint32(in[18])<<16 | uint32(in[19])<<24
j5 := uint32(in[20]) | uint32(in[21])<<8 | uint32(in[22])<<16 | uint32(in[23])<<24
j6 := uint32(in[24]) | uint32(in[25])<<8 | uint32(in[26])<<16 | uint32(in[27])<<24
j7 := uint32(in[28]) | uint32(in[29])<<8 | uint32(in[30])<<16 | uint32(in[31])<<24
j8 := uint32(in[32]) | uint32(in[33])<<8 | uint32(in[34])<<16 | uint32(in[35])<<24
j9 := uint32(in[36]) | uint32(in[37])<<8 | uint32(in[38])<<16 | uint32(in[39])<<24
j10 := uint32(in[40]) | uint32(in[41])<<8 | uint32(in[42])<<16 | uint32(in[43])<<24
j11 := uint32(in[44]) | uint32(in[45])<<8 | uint32(in[46])<<16 | uint32(in[47])<<24
j12 := uint32(in[48]) | uint32(in[49])<<8 | uint32(in[50])<<16 | uint32(in[51])<<24
j13 := uint32(in[52]) | uint32(in[53])<<8 | uint32(in[54])<<16 | uint32(in[55])<<24
j14 := uint32(in[56]) | uint32(in[57])<<8 | uint32(in[58])<<16 | uint32(in[59])<<24
j15 := uint32(in[60]) | uint32(in[61])<<8 | uint32(in[62])<<16 | uint32(in[63])<<24
x0, x1, x2, x3, x4, x5, x6, x7, x8 := j0, j1, j2, j3, j4, j5, j6, j7, j8
x9, x10, x11, x12, x13, x14, x15 := j9, j10, j11, j12, j13, j14, j15
for i := 0; i < 8; i += 2 {
u := x0 + x12
x4 ^= u<<7 | u>>(32-7)
u = x4 + x0
x8 ^= u<<9 | u>>(32-9)
u = x8 + x4
x12 ^= u<<13 | u>>(32-13)
u = x12 + x8
x0 ^= u<<18 | u>>(32-18)
u = x5 + x1
x9 ^= u<<7 | u>>(32-7)
u = x9 + x5
x13 ^= u<<9 | u>>(32-9)
u = x13 + x9
x1 ^= u<<13 | u>>(32-13)
u = x1 + x13
x5 ^= u<<18 | u>>(32-18)
u = x10 + x6
x14 ^= u<<7 | u>>(32-7)
u = x14 + x10
x2 ^= u<<9 | u>>(32-9)
u = x2 + x14
x6 ^= u<<13 | u>>(32-13)
u = x6 + x2
x10 ^= u<<18 | u>>(32-18)
u = x15 + x11
x3 ^= u<<7 | u>>(32-7)
u = x3 + x15
x7 ^= u<<9 | u>>(32-9)
u = x7 + x3
x11 ^= u<<13 | u>>(32-13)
u = x11 + x7
x15 ^= u<<18 | u>>(32-18)
u = x0 + x3
x1 ^= u<<7 | u>>(32-7)
u = x1 + x0
x2 ^= u<<9 | u>>(32-9)
u = x2 + x1
x3 ^= u<<13 | u>>(32-13)
u = x3 + x2
x0 ^= u<<18 | u>>(32-18)
u = x5 + x4
x6 ^= u<<7 | u>>(32-7)
u = x6 + x5
x7 ^= u<<9 | u>>(32-9)
u = x7 + x6
x4 ^= u<<13 | u>>(32-13)
u = x4 + x7
x5 ^= u<<18 | u>>(32-18)
u = x10 + x9
x11 ^= u<<7 | u>>(32-7)
u = x11 + x10
x8 ^= u<<9 | u>>(32-9)
u = x8 + x11
x9 ^= u<<13 | u>>(32-13)
u = x9 + x8
x10 ^= u<<18 | u>>(32-18)
u = x15 + x14
x12 ^= u<<7 | u>>(32-7)
u = x12 + x15
x13 ^= u<<9 | u>>(32-9)
u = x13 + x12
x14 ^= u<<13 | u>>(32-13)
u = x14 + x13
x15 ^= u<<18 | u>>(32-18)
}
x0 += j0
x1 += j1
x2 += j2
x3 += j3
x4 += j4
x5 += j5
x6 += j6
x7 += j7
x8 += j8
x9 += j9
x10 += j10
x11 += j11
x12 += j12
x13 += j13
x14 += j14
x15 += j15
out[0] = byte(x0)
out[1] = byte(x0 >> 8)
out[2] = byte(x0 >> 16)
out[3] = byte(x0 >> 24)
out[4] = byte(x1)
out[5] = byte(x1 >> 8)
out[6] = byte(x1 >> 16)
out[7] = byte(x1 >> 24)
out[8] = byte(x2)
out[9] = byte(x2 >> 8)
out[10] = byte(x2 >> 16)
out[11] = byte(x2 >> 24)
out[12] = byte(x3)
out[13] = byte(x3 >> 8)
out[14] = byte(x3 >> 16)
out[15] = byte(x3 >> 24)
out[16] = byte(x4)
out[17] = byte(x4 >> 8)
out[18] = byte(x4 >> 16)
out[19] = byte(x4 >> 24)
out[20] = byte(x5)
out[21] = byte(x5 >> 8)
out[22] = byte(x5 >> 16)
out[23] = byte(x5 >> 24)
out[24] = byte(x6)
out[25] = byte(x6 >> 8)
out[26] = byte(x6 >> 16)
out[27] = byte(x6 >> 24)
out[28] = byte(x7)
out[29] = byte(x7 >> 8)
out[30] = byte(x7 >> 16)
out[31] = byte(x7 >> 24)
out[32] = byte(x8)
out[33] = byte(x8 >> 8)
out[34] = byte(x8 >> 16)
out[35] = byte(x8 >> 24)
out[36] = byte(x9)
out[37] = byte(x9 >> 8)
out[38] = byte(x9 >> 16)
out[39] = byte(x9 >> 24)
out[40] = byte(x10)
out[41] = byte(x10 >> 8)
out[42] = byte(x10 >> 16)
out[43] = byte(x10 >> 24)
out[44] = byte(x11)
out[45] = byte(x11 >> 8)
out[46] = byte(x11 >> 16)
out[47] = byte(x11 >> 24)
out[48] = byte(x12)
out[49] = byte(x12 >> 8)
out[50] = byte(x12 >> 16)
out[51] = byte(x12 >> 24)
out[52] = byte(x13)
out[53] = byte(x13 >> 8)
out[54] = byte(x13 >> 16)
out[55] = byte(x13 >> 24)
out[56] = byte(x14)
out[57] = byte(x14 >> 8)
out[58] = byte(x14 >> 16)
out[59] = byte(x14 >> 24)
out[60] = byte(x15)
out[61] = byte(x15 >> 8)
out[62] = byte(x15 >> 16)
out[63] = byte(x15 >> 24)
}

View File

@@ -0,0 +1,23 @@
// Copyright 2012 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
// +build amd64,!appengine,!gccgo
package salsa
//go:noescape
// salsa2020XORKeyStream is implemented in salsa20_amd64.s.
func salsa2020XORKeyStream(out, in *byte, n uint64, nonce, key *byte)
// XORKeyStream crypts bytes from in to out using the given key and counters.
// In and out must overlap entirely or not at all. Counter
// contains the raw salsa20 counter bytes (both nonce and block counter).
func XORKeyStream(out, in []byte, counter *[16]byte, key *[32]byte) {
if len(in) == 0 {
return
}
_ = out[len(in)-1]
salsa2020XORKeyStream(&out[0], &in[0], uint64(len(in)), &counter[0], &key[0])
}

View File

@@ -0,0 +1,883 @@
// Copyright 2012 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
// +build amd64,!appengine,!gccgo
// This code was translated into a form compatible with 6a from the public
// domain sources in SUPERCOP: https://bench.cr.yp.to/supercop.html
// func salsa2020XORKeyStream(out, in *byte, n uint64, nonce, key *byte)
// This needs up to 64 bytes at 360(SP); hence the non-obvious frame size.
TEXT ·salsa2020XORKeyStream(SB),0,$456-40 // frame = 424 + 32 byte alignment
MOVQ out+0(FP),DI
MOVQ in+8(FP),SI
MOVQ n+16(FP),DX
MOVQ nonce+24(FP),CX
MOVQ key+32(FP),R8
MOVQ SP,R12
MOVQ SP,R9
ADDQ $31, R9
ANDQ $~31, R9
MOVQ R9, SP
MOVQ DX,R9
MOVQ CX,DX
MOVQ R8,R10
CMPQ R9,$0
JBE DONE
START:
MOVL 20(R10),CX
MOVL 0(R10),R8
MOVL 0(DX),AX
MOVL 16(R10),R11
MOVL CX,0(SP)
MOVL R8, 4 (SP)
MOVL AX, 8 (SP)
MOVL R11, 12 (SP)
MOVL 8(DX),CX
MOVL 24(R10),R8
MOVL 4(R10),AX
MOVL 4(DX),R11
MOVL CX,16(SP)
MOVL R8, 20 (SP)
MOVL AX, 24 (SP)
MOVL R11, 28 (SP)
MOVL 12(DX),CX
MOVL 12(R10),DX
MOVL 28(R10),R8
MOVL 8(R10),AX
MOVL DX,32(SP)
MOVL CX, 36 (SP)
MOVL R8, 40 (SP)
MOVL AX, 44 (SP)
MOVQ $1634760805,DX
MOVQ $857760878,CX
MOVQ $2036477234,R8
MOVQ $1797285236,AX
MOVL DX,48(SP)
MOVL CX, 52 (SP)
MOVL R8, 56 (SP)
MOVL AX, 60 (SP)
CMPQ R9,$256
JB BYTESBETWEEN1AND255
MOVOA 48(SP),X0
PSHUFL $0X55,X0,X1
PSHUFL $0XAA,X0,X2
PSHUFL $0XFF,X0,X3
PSHUFL $0X00,X0,X0
MOVOA X1,64(SP)
MOVOA X2,80(SP)
MOVOA X3,96(SP)
MOVOA X0,112(SP)
MOVOA 0(SP),X0
PSHUFL $0XAA,X0,X1
PSHUFL $0XFF,X0,X2
PSHUFL $0X00,X0,X3
PSHUFL $0X55,X0,X0
MOVOA X1,128(SP)
MOVOA X2,144(SP)
MOVOA X3,160(SP)
MOVOA X0,176(SP)
MOVOA 16(SP),X0
PSHUFL $0XFF,X0,X1
PSHUFL $0X55,X0,X2
PSHUFL $0XAA,X0,X0
MOVOA X1,192(SP)
MOVOA X2,208(SP)
MOVOA X0,224(SP)
MOVOA 32(SP),X0
PSHUFL $0X00,X0,X1
PSHUFL $0XAA,X0,X2
PSHUFL $0XFF,X0,X0
MOVOA X1,240(SP)
MOVOA X2,256(SP)
MOVOA X0,272(SP)
BYTESATLEAST256:
MOVL 16(SP),DX
MOVL 36 (SP),CX
MOVL DX,288(SP)
MOVL CX,304(SP)
SHLQ $32,CX
ADDQ CX,DX
ADDQ $1,DX
MOVQ DX,CX
SHRQ $32,CX
MOVL DX, 292 (SP)
MOVL CX, 308 (SP)
ADDQ $1,DX
MOVQ DX,CX
SHRQ $32,CX
MOVL DX, 296 (SP)
MOVL CX, 312 (SP)
ADDQ $1,DX
MOVQ DX,CX
SHRQ $32,CX
MOVL DX, 300 (SP)
MOVL CX, 316 (SP)
ADDQ $1,DX
MOVQ DX,CX
SHRQ $32,CX
MOVL DX,16(SP)
MOVL CX, 36 (SP)
MOVQ R9,352(SP)
MOVQ $20,DX
MOVOA 64(SP),X0
MOVOA 80(SP),X1
MOVOA 96(SP),X2
MOVOA 256(SP),X3
MOVOA 272(SP),X4
MOVOA 128(SP),X5
MOVOA 144(SP),X6
MOVOA 176(SP),X7
MOVOA 192(SP),X8
MOVOA 208(SP),X9
MOVOA 224(SP),X10
MOVOA 304(SP),X11
MOVOA 112(SP),X12
MOVOA 160(SP),X13
MOVOA 240(SP),X14
MOVOA 288(SP),X15
MAINLOOP1:
MOVOA X1,320(SP)
MOVOA X2,336(SP)
MOVOA X13,X1
PADDL X12,X1
MOVOA X1,X2
PSLLL $7,X1
PXOR X1,X14
PSRLL $25,X2
PXOR X2,X14
MOVOA X7,X1
PADDL X0,X1
MOVOA X1,X2
PSLLL $7,X1
PXOR X1,X11
PSRLL $25,X2
PXOR X2,X11
MOVOA X12,X1
PADDL X14,X1
MOVOA X1,X2
PSLLL $9,X1
PXOR X1,X15
PSRLL $23,X2
PXOR X2,X15
MOVOA X0,X1
PADDL X11,X1
MOVOA X1,X2
PSLLL $9,X1
PXOR X1,X9
PSRLL $23,X2
PXOR X2,X9
MOVOA X14,X1
PADDL X15,X1
MOVOA X1,X2
PSLLL $13,X1
PXOR X1,X13
PSRLL $19,X2
PXOR X2,X13
MOVOA X11,X1
PADDL X9,X1
MOVOA X1,X2
PSLLL $13,X1
PXOR X1,X7
PSRLL $19,X2
PXOR X2,X7
MOVOA X15,X1
PADDL X13,X1
MOVOA X1,X2
PSLLL $18,X1
PXOR X1,X12
PSRLL $14,X2
PXOR X2,X12
MOVOA 320(SP),X1
MOVOA X12,320(SP)
MOVOA X9,X2
PADDL X7,X2
MOVOA X2,X12
PSLLL $18,X2
PXOR X2,X0
PSRLL $14,X12
PXOR X12,X0
MOVOA X5,X2
PADDL X1,X2
MOVOA X2,X12
PSLLL $7,X2
PXOR X2,X3
PSRLL $25,X12
PXOR X12,X3
MOVOA 336(SP),X2
MOVOA X0,336(SP)
MOVOA X6,X0
PADDL X2,X0
MOVOA X0,X12
PSLLL $7,X0
PXOR X0,X4
PSRLL $25,X12
PXOR X12,X4
MOVOA X1,X0
PADDL X3,X0
MOVOA X0,X12
PSLLL $9,X0
PXOR X0,X10
PSRLL $23,X12
PXOR X12,X10
MOVOA X2,X0
PADDL X4,X0
MOVOA X0,X12
PSLLL $9,X0
PXOR X0,X8
PSRLL $23,X12
PXOR X12,X8
MOVOA X3,X0
PADDL X10,X0
MOVOA X0,X12
PSLLL $13,X0
PXOR X0,X5
PSRLL $19,X12
PXOR X12,X5
MOVOA X4,X0
PADDL X8,X0
MOVOA X0,X12
PSLLL $13,X0
PXOR X0,X6
PSRLL $19,X12
PXOR X12,X6
MOVOA X10,X0
PADDL X5,X0
MOVOA X0,X12
PSLLL $18,X0
PXOR X0,X1
PSRLL $14,X12
PXOR X12,X1
MOVOA 320(SP),X0
MOVOA X1,320(SP)
MOVOA X4,X1
PADDL X0,X1
MOVOA X1,X12
PSLLL $7,X1
PXOR X1,X7
PSRLL $25,X12
PXOR X12,X7
MOVOA X8,X1
PADDL X6,X1
MOVOA X1,X12
PSLLL $18,X1
PXOR X1,X2
PSRLL $14,X12
PXOR X12,X2
MOVOA 336(SP),X12
MOVOA X2,336(SP)
MOVOA X14,X1
PADDL X12,X1
MOVOA X1,X2
PSLLL $7,X1
PXOR X1,X5
PSRLL $25,X2
PXOR X2,X5
MOVOA X0,X1
PADDL X7,X1
MOVOA X1,X2
PSLLL $9,X1
PXOR X1,X10
PSRLL $23,X2
PXOR X2,X10
MOVOA X12,X1
PADDL X5,X1
MOVOA X1,X2
PSLLL $9,X1
PXOR X1,X8
PSRLL $23,X2
PXOR X2,X8
MOVOA X7,X1
PADDL X10,X1
MOVOA X1,X2
PSLLL $13,X1
PXOR X1,X4
PSRLL $19,X2
PXOR X2,X4
MOVOA X5,X1
PADDL X8,X1
MOVOA X1,X2
PSLLL $13,X1
PXOR X1,X14
PSRLL $19,X2
PXOR X2,X14
MOVOA X10,X1
PADDL X4,X1
MOVOA X1,X2
PSLLL $18,X1
PXOR X1,X0
PSRLL $14,X2
PXOR X2,X0
MOVOA 320(SP),X1
MOVOA X0,320(SP)
MOVOA X8,X0
PADDL X14,X0
MOVOA X0,X2
PSLLL $18,X0
PXOR X0,X12
PSRLL $14,X2
PXOR X2,X12
MOVOA X11,X0
PADDL X1,X0
MOVOA X0,X2
PSLLL $7,X0
PXOR X0,X6
PSRLL $25,X2
PXOR X2,X6
MOVOA 336(SP),X2
MOVOA X12,336(SP)
MOVOA X3,X0
PADDL X2,X0
MOVOA X0,X12
PSLLL $7,X0
PXOR X0,X13
PSRLL $25,X12
PXOR X12,X13
MOVOA X1,X0
PADDL X6,X0
MOVOA X0,X12
PSLLL $9,X0
PXOR X0,X15
PSRLL $23,X12
PXOR X12,X15
MOVOA X2,X0
PADDL X13,X0
MOVOA X0,X12
PSLLL $9,X0
PXOR X0,X9
PSRLL $23,X12
PXOR X12,X9
MOVOA X6,X0
PADDL X15,X0
MOVOA X0,X12
PSLLL $13,X0
PXOR X0,X11
PSRLL $19,X12
PXOR X12,X11
MOVOA X13,X0
PADDL X9,X0
MOVOA X0,X12
PSLLL $13,X0
PXOR X0,X3
PSRLL $19,X12
PXOR X12,X3
MOVOA X15,X0
PADDL X11,X0
MOVOA X0,X12
PSLLL $18,X0
PXOR X0,X1
PSRLL $14,X12
PXOR X12,X1
MOVOA X9,X0
PADDL X3,X0
MOVOA X0,X12
PSLLL $18,X0
PXOR X0,X2
PSRLL $14,X12
PXOR X12,X2
MOVOA 320(SP),X12
MOVOA 336(SP),X0
SUBQ $2,DX
JA MAINLOOP1
PADDL 112(SP),X12
PADDL 176(SP),X7
PADDL 224(SP),X10
PADDL 272(SP),X4
MOVD X12,DX
MOVD X7,CX
MOVD X10,R8
MOVD X4,R9
PSHUFL $0X39,X12,X12
PSHUFL $0X39,X7,X7
PSHUFL $0X39,X10,X10
PSHUFL $0X39,X4,X4
XORL 0(SI),DX
XORL 4(SI),CX
XORL 8(SI),R8
XORL 12(SI),R9
MOVL DX,0(DI)
MOVL CX,4(DI)
MOVL R8,8(DI)
MOVL R9,12(DI)
MOVD X12,DX
MOVD X7,CX
MOVD X10,R8
MOVD X4,R9
PSHUFL $0X39,X12,X12
PSHUFL $0X39,X7,X7
PSHUFL $0X39,X10,X10
PSHUFL $0X39,X4,X4
XORL 64(SI),DX
XORL 68(SI),CX
XORL 72(SI),R8
XORL 76(SI),R9
MOVL DX,64(DI)
MOVL CX,68(DI)
MOVL R8,72(DI)
MOVL R9,76(DI)
MOVD X12,DX
MOVD X7,CX
MOVD X10,R8
MOVD X4,R9
PSHUFL $0X39,X12,X12
PSHUFL $0X39,X7,X7
PSHUFL $0X39,X10,X10
PSHUFL $0X39,X4,X4
XORL 128(SI),DX
XORL 132(SI),CX
XORL 136(SI),R8
XORL 140(SI),R9
MOVL DX,128(DI)
MOVL CX,132(DI)
MOVL R8,136(DI)
MOVL R9,140(DI)
MOVD X12,DX
MOVD X7,CX
MOVD X10,R8
MOVD X4,R9
XORL 192(SI),DX
XORL 196(SI),CX
XORL 200(SI),R8
XORL 204(SI),R9
MOVL DX,192(DI)
MOVL CX,196(DI)
MOVL R8,200(DI)
MOVL R9,204(DI)
PADDL 240(SP),X14
PADDL 64(SP),X0
PADDL 128(SP),X5
PADDL 192(SP),X8
MOVD X14,DX
MOVD X0,CX
MOVD X5,R8
MOVD X8,R9
PSHUFL $0X39,X14,X14
PSHUFL $0X39,X0,X0
PSHUFL $0X39,X5,X5
PSHUFL $0X39,X8,X8
XORL 16(SI),DX
XORL 20(SI),CX
XORL 24(SI),R8
XORL 28(SI),R9
MOVL DX,16(DI)
MOVL CX,20(DI)
MOVL R8,24(DI)
MOVL R9,28(DI)
MOVD X14,DX
MOVD X0,CX
MOVD X5,R8
MOVD X8,R9
PSHUFL $0X39,X14,X14
PSHUFL $0X39,X0,X0
PSHUFL $0X39,X5,X5
PSHUFL $0X39,X8,X8
XORL 80(SI),DX
XORL 84(SI),CX
XORL 88(SI),R8
XORL 92(SI),R9
MOVL DX,80(DI)
MOVL CX,84(DI)
MOVL R8,88(DI)
MOVL R9,92(DI)
MOVD X14,DX
MOVD X0,CX
MOVD X5,R8
MOVD X8,R9
PSHUFL $0X39,X14,X14
PSHUFL $0X39,X0,X0
PSHUFL $0X39,X5,X5
PSHUFL $0X39,X8,X8
XORL 144(SI),DX
XORL 148(SI),CX
XORL 152(SI),R8
XORL 156(SI),R9
MOVL DX,144(DI)
MOVL CX,148(DI)
MOVL R8,152(DI)
MOVL R9,156(DI)
MOVD X14,DX
MOVD X0,CX
MOVD X5,R8
MOVD X8,R9
XORL 208(SI),DX
XORL 212(SI),CX
XORL 216(SI),R8
XORL 220(SI),R9
MOVL DX,208(DI)
MOVL CX,212(DI)
MOVL R8,216(DI)
MOVL R9,220(DI)
PADDL 288(SP),X15
PADDL 304(SP),X11
PADDL 80(SP),X1
PADDL 144(SP),X6
MOVD X15,DX
MOVD X11,CX
MOVD X1,R8
MOVD X6,R9
PSHUFL $0X39,X15,X15
PSHUFL $0X39,X11,X11
PSHUFL $0X39,X1,X1
PSHUFL $0X39,X6,X6
XORL 32(SI),DX
XORL 36(SI),CX
XORL 40(SI),R8
XORL 44(SI),R9
MOVL DX,32(DI)
MOVL CX,36(DI)
MOVL R8,40(DI)
MOVL R9,44(DI)
MOVD X15,DX
MOVD X11,CX
MOVD X1,R8
MOVD X6,R9
PSHUFL $0X39,X15,X15
PSHUFL $0X39,X11,X11
PSHUFL $0X39,X1,X1
PSHUFL $0X39,X6,X6
XORL 96(SI),DX
XORL 100(SI),CX
XORL 104(SI),R8
XORL 108(SI),R9
MOVL DX,96(DI)
MOVL CX,100(DI)
MOVL R8,104(DI)
MOVL R9,108(DI)
MOVD X15,DX
MOVD X11,CX
MOVD X1,R8
MOVD X6,R9
PSHUFL $0X39,X15,X15
PSHUFL $0X39,X11,X11
PSHUFL $0X39,X1,X1
PSHUFL $0X39,X6,X6
XORL 160(SI),DX
XORL 164(SI),CX
XORL 168(SI),R8
XORL 172(SI),R9
MOVL DX,160(DI)
MOVL CX,164(DI)
MOVL R8,168(DI)
MOVL R9,172(DI)
MOVD X15,DX
MOVD X11,CX
MOVD X1,R8
MOVD X6,R9
XORL 224(SI),DX
XORL 228(SI),CX
XORL 232(SI),R8
XORL 236(SI),R9
MOVL DX,224(DI)
MOVL CX,228(DI)
MOVL R8,232(DI)
MOVL R9,236(DI)
PADDL 160(SP),X13
PADDL 208(SP),X9
PADDL 256(SP),X3
PADDL 96(SP),X2
MOVD X13,DX
MOVD X9,CX
MOVD X3,R8
MOVD X2,R9
PSHUFL $0X39,X13,X13
PSHUFL $0X39,X9,X9
PSHUFL $0X39,X3,X3
PSHUFL $0X39,X2,X2
XORL 48(SI),DX
XORL 52(SI),CX
XORL 56(SI),R8
XORL 60(SI),R9
MOVL DX,48(DI)
MOVL CX,52(DI)
MOVL R8,56(DI)
MOVL R9,60(DI)
MOVD X13,DX
MOVD X9,CX
MOVD X3,R8
MOVD X2,R9
PSHUFL $0X39,X13,X13
PSHUFL $0X39,X9,X9
PSHUFL $0X39,X3,X3
PSHUFL $0X39,X2,X2
XORL 112(SI),DX
XORL 116(SI),CX
XORL 120(SI),R8
XORL 124(SI),R9
MOVL DX,112(DI)
MOVL CX,116(DI)
MOVL R8,120(DI)
MOVL R9,124(DI)
MOVD X13,DX
MOVD X9,CX
MOVD X3,R8
MOVD X2,R9
PSHUFL $0X39,X13,X13
PSHUFL $0X39,X9,X9
PSHUFL $0X39,X3,X3
PSHUFL $0X39,X2,X2
XORL 176(SI),DX
XORL 180(SI),CX
XORL 184(SI),R8
XORL 188(SI),R9
MOVL DX,176(DI)
MOVL CX,180(DI)
MOVL R8,184(DI)
MOVL R9,188(DI)
MOVD X13,DX
MOVD X9,CX
MOVD X3,R8
MOVD X2,R9
XORL 240(SI),DX
XORL 244(SI),CX
XORL 248(SI),R8
XORL 252(SI),R9
MOVL DX,240(DI)
MOVL CX,244(DI)
MOVL R8,248(DI)
MOVL R9,252(DI)
MOVQ 352(SP),R9
SUBQ $256,R9
ADDQ $256,SI
ADDQ $256,DI
CMPQ R9,$256
JAE BYTESATLEAST256
CMPQ R9,$0
JBE DONE
BYTESBETWEEN1AND255:
CMPQ R9,$64
JAE NOCOPY
MOVQ DI,DX
LEAQ 360(SP),DI
MOVQ R9,CX
REP; MOVSB
LEAQ 360(SP),DI
LEAQ 360(SP),SI
NOCOPY:
MOVQ R9,352(SP)
MOVOA 48(SP),X0
MOVOA 0(SP),X1
MOVOA 16(SP),X2
MOVOA 32(SP),X3
MOVOA X1,X4
MOVQ $20,CX
MAINLOOP2:
PADDL X0,X4
MOVOA X0,X5
MOVOA X4,X6
PSLLL $7,X4
PSRLL $25,X6
PXOR X4,X3
PXOR X6,X3
PADDL X3,X5
MOVOA X3,X4
MOVOA X5,X6
PSLLL $9,X5
PSRLL $23,X6
PXOR X5,X2
PSHUFL $0X93,X3,X3
PXOR X6,X2
PADDL X2,X4
MOVOA X2,X5
MOVOA X4,X6
PSLLL $13,X4
PSRLL $19,X6
PXOR X4,X1
PSHUFL $0X4E,X2,X2
PXOR X6,X1
PADDL X1,X5
MOVOA X3,X4
MOVOA X5,X6
PSLLL $18,X5
PSRLL $14,X6
PXOR X5,X0
PSHUFL $0X39,X1,X1
PXOR X6,X0
PADDL X0,X4
MOVOA X0,X5
MOVOA X4,X6
PSLLL $7,X4
PSRLL $25,X6
PXOR X4,X1
PXOR X6,X1
PADDL X1,X5
MOVOA X1,X4
MOVOA X5,X6
PSLLL $9,X5
PSRLL $23,X6
PXOR X5,X2
PSHUFL $0X93,X1,X1
PXOR X6,X2
PADDL X2,X4
MOVOA X2,X5
MOVOA X4,X6
PSLLL $13,X4
PSRLL $19,X6
PXOR X4,X3
PSHUFL $0X4E,X2,X2
PXOR X6,X3
PADDL X3,X5
MOVOA X1,X4
MOVOA X5,X6
PSLLL $18,X5
PSRLL $14,X6
PXOR X5,X0
PSHUFL $0X39,X3,X3
PXOR X6,X0
PADDL X0,X4
MOVOA X0,X5
MOVOA X4,X6
PSLLL $7,X4
PSRLL $25,X6
PXOR X4,X3
PXOR X6,X3
PADDL X3,X5
MOVOA X3,X4
MOVOA X5,X6
PSLLL $9,X5
PSRLL $23,X6
PXOR X5,X2
PSHUFL $0X93,X3,X3
PXOR X6,X2
PADDL X2,X4
MOVOA X2,X5
MOVOA X4,X6
PSLLL $13,X4
PSRLL $19,X6
PXOR X4,X1
PSHUFL $0X4E,X2,X2
PXOR X6,X1
PADDL X1,X5
MOVOA X3,X4
MOVOA X5,X6
PSLLL $18,X5
PSRLL $14,X6
PXOR X5,X0
PSHUFL $0X39,X1,X1
PXOR X6,X0
PADDL X0,X4
MOVOA X0,X5
MOVOA X4,X6
PSLLL $7,X4
PSRLL $25,X6
PXOR X4,X1
PXOR X6,X1
PADDL X1,X5
MOVOA X1,X4
MOVOA X5,X6
PSLLL $9,X5
PSRLL $23,X6
PXOR X5,X2
PSHUFL $0X93,X1,X1
PXOR X6,X2
PADDL X2,X4
MOVOA X2,X5
MOVOA X4,X6
PSLLL $13,X4
PSRLL $19,X6
PXOR X4,X3
PSHUFL $0X4E,X2,X2
PXOR X6,X3
SUBQ $4,CX
PADDL X3,X5
MOVOA X1,X4
MOVOA X5,X6
PSLLL $18,X5
PXOR X7,X7
PSRLL $14,X6
PXOR X5,X0
PSHUFL $0X39,X3,X3
PXOR X6,X0
JA MAINLOOP2
PADDL 48(SP),X0
PADDL 0(SP),X1
PADDL 16(SP),X2
PADDL 32(SP),X3
MOVD X0,CX
MOVD X1,R8
MOVD X2,R9
MOVD X3,AX
PSHUFL $0X39,X0,X0
PSHUFL $0X39,X1,X1
PSHUFL $0X39,X2,X2
PSHUFL $0X39,X3,X3
XORL 0(SI),CX
XORL 48(SI),R8
XORL 32(SI),R9
XORL 16(SI),AX
MOVL CX,0(DI)
MOVL R8,48(DI)
MOVL R9,32(DI)
MOVL AX,16(DI)
MOVD X0,CX
MOVD X1,R8
MOVD X2,R9
MOVD X3,AX
PSHUFL $0X39,X0,X0
PSHUFL $0X39,X1,X1
PSHUFL $0X39,X2,X2
PSHUFL $0X39,X3,X3
XORL 20(SI),CX
XORL 4(SI),R8
XORL 52(SI),R9
XORL 36(SI),AX
MOVL CX,20(DI)
MOVL R8,4(DI)
MOVL R9,52(DI)
MOVL AX,36(DI)
MOVD X0,CX
MOVD X1,R8
MOVD X2,R9
MOVD X3,AX
PSHUFL $0X39,X0,X0
PSHUFL $0X39,X1,X1
PSHUFL $0X39,X2,X2
PSHUFL $0X39,X3,X3
XORL 40(SI),CX
XORL 24(SI),R8
XORL 8(SI),R9
XORL 56(SI),AX
MOVL CX,40(DI)
MOVL R8,24(DI)
MOVL R9,8(DI)
MOVL AX,56(DI)
MOVD X0,CX
MOVD X1,R8
MOVD X2,R9
MOVD X3,AX
XORL 60(SI),CX
XORL 44(SI),R8
XORL 28(SI),R9
XORL 12(SI),AX
MOVL CX,60(DI)
MOVL R8,44(DI)
MOVL R9,28(DI)
MOVL AX,12(DI)
MOVQ 352(SP),R9
MOVL 16(SP),CX
MOVL 36 (SP),R8
ADDQ $1,CX
SHLQ $32,R8
ADDQ R8,CX
MOVQ CX,R8
SHRQ $32,R8
MOVL CX,16(SP)
MOVL R8, 36 (SP)
CMPQ R9,$64
JA BYTESATLEAST65
JAE BYTESATLEAST64
MOVQ DI,SI
MOVQ DX,DI
MOVQ R9,CX
REP; MOVSB
BYTESATLEAST64:
DONE:
MOVQ R12,SP
RET
BYTESATLEAST65:
SUBQ $64,R9
ADDQ $64,DI
ADDQ $64,SI
JMP BYTESBETWEEN1AND255

View File

@@ -0,0 +1,14 @@
// Copyright 2019 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
// +build !amd64 appengine gccgo
package salsa
// XORKeyStream crypts bytes from in to out using the given key and counters.
// In and out must overlap entirely or not at all. Counter
// contains the raw salsa20 counter bytes (both nonce and block counter).
func XORKeyStream(out, in []byte, counter *[16]byte, key *[32]byte) {
genericXORKeyStream(out, in, counter, key)
}

231
vendor/golang.org/x/crypto/salsa20/salsa/salsa20_ref.go generated vendored Normal file
View File

@@ -0,0 +1,231 @@
// Copyright 2012 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
package salsa
const rounds = 20
// core applies the Salsa20 core function to 16-byte input in, 32-byte key k,
// and 16-byte constant c, and puts the result into 64-byte array out.
func core(out *[64]byte, in *[16]byte, k *[32]byte, c *[16]byte) {
j0 := uint32(c[0]) | uint32(c[1])<<8 | uint32(c[2])<<16 | uint32(c[3])<<24
j1 := uint32(k[0]) | uint32(k[1])<<8 | uint32(k[2])<<16 | uint32(k[3])<<24
j2 := uint32(k[4]) | uint32(k[5])<<8 | uint32(k[6])<<16 | uint32(k[7])<<24
j3 := uint32(k[8]) | uint32(k[9])<<8 | uint32(k[10])<<16 | uint32(k[11])<<24
j4 := uint32(k[12]) | uint32(k[13])<<8 | uint32(k[14])<<16 | uint32(k[15])<<24
j5 := uint32(c[4]) | uint32(c[5])<<8 | uint32(c[6])<<16 | uint32(c[7])<<24
j6 := uint32(in[0]) | uint32(in[1])<<8 | uint32(in[2])<<16 | uint32(in[3])<<24
j7 := uint32(in[4]) | uint32(in[5])<<8 | uint32(in[6])<<16 | uint32(in[7])<<24
j8 := uint32(in[8]) | uint32(in[9])<<8 | uint32(in[10])<<16 | uint32(in[11])<<24
j9 := uint32(in[12]) | uint32(in[13])<<8 | uint32(in[14])<<16 | uint32(in[15])<<24
j10 := uint32(c[8]) | uint32(c[9])<<8 | uint32(c[10])<<16 | uint32(c[11])<<24
j11 := uint32(k[16]) | uint32(k[17])<<8 | uint32(k[18])<<16 | uint32(k[19])<<24
j12 := uint32(k[20]) | uint32(k[21])<<8 | uint32(k[22])<<16 | uint32(k[23])<<24
j13 := uint32(k[24]) | uint32(k[25])<<8 | uint32(k[26])<<16 | uint32(k[27])<<24
j14 := uint32(k[28]) | uint32(k[29])<<8 | uint32(k[30])<<16 | uint32(k[31])<<24
j15 := uint32(c[12]) | uint32(c[13])<<8 | uint32(c[14])<<16 | uint32(c[15])<<24
x0, x1, x2, x3, x4, x5, x6, x7, x8 := j0, j1, j2, j3, j4, j5, j6, j7, j8
x9, x10, x11, x12, x13, x14, x15 := j9, j10, j11, j12, j13, j14, j15
for i := 0; i < rounds; i += 2 {
u := x0 + x12
x4 ^= u<<7 | u>>(32-7)
u = x4 + x0
x8 ^= u<<9 | u>>(32-9)
u = x8 + x4
x12 ^= u<<13 | u>>(32-13)
u = x12 + x8
x0 ^= u<<18 | u>>(32-18)
u = x5 + x1
x9 ^= u<<7 | u>>(32-7)
u = x9 + x5
x13 ^= u<<9 | u>>(32-9)
u = x13 + x9
x1 ^= u<<13 | u>>(32-13)
u = x1 + x13
x5 ^= u<<18 | u>>(32-18)
u = x10 + x6
x14 ^= u<<7 | u>>(32-7)
u = x14 + x10
x2 ^= u<<9 | u>>(32-9)
u = x2 + x14
x6 ^= u<<13 | u>>(32-13)
u = x6 + x2
x10 ^= u<<18 | u>>(32-18)
u = x15 + x11
x3 ^= u<<7 | u>>(32-7)
u = x3 + x15
x7 ^= u<<9 | u>>(32-9)
u = x7 + x3
x11 ^= u<<13 | u>>(32-13)
u = x11 + x7
x15 ^= u<<18 | u>>(32-18)
u = x0 + x3
x1 ^= u<<7 | u>>(32-7)
u = x1 + x0
x2 ^= u<<9 | u>>(32-9)
u = x2 + x1
x3 ^= u<<13 | u>>(32-13)
u = x3 + x2
x0 ^= u<<18 | u>>(32-18)
u = x5 + x4
x6 ^= u<<7 | u>>(32-7)
u = x6 + x5
x7 ^= u<<9 | u>>(32-9)
u = x7 + x6
x4 ^= u<<13 | u>>(32-13)
u = x4 + x7
x5 ^= u<<18 | u>>(32-18)
u = x10 + x9
x11 ^= u<<7 | u>>(32-7)
u = x11 + x10
x8 ^= u<<9 | u>>(32-9)
u = x8 + x11
x9 ^= u<<13 | u>>(32-13)
u = x9 + x8
x10 ^= u<<18 | u>>(32-18)
u = x15 + x14
x12 ^= u<<7 | u>>(32-7)
u = x12 + x15
x13 ^= u<<9 | u>>(32-9)
u = x13 + x12
x14 ^= u<<13 | u>>(32-13)
u = x14 + x13
x15 ^= u<<18 | u>>(32-18)
}
x0 += j0
x1 += j1
x2 += j2
x3 += j3
x4 += j4
x5 += j5
x6 += j6
x7 += j7
x8 += j8
x9 += j9
x10 += j10
x11 += j11
x12 += j12
x13 += j13
x14 += j14
x15 += j15
out[0] = byte(x0)
out[1] = byte(x0 >> 8)
out[2] = byte(x0 >> 16)
out[3] = byte(x0 >> 24)
out[4] = byte(x1)
out[5] = byte(x1 >> 8)
out[6] = byte(x1 >> 16)
out[7] = byte(x1 >> 24)
out[8] = byte(x2)
out[9] = byte(x2 >> 8)
out[10] = byte(x2 >> 16)
out[11] = byte(x2 >> 24)
out[12] = byte(x3)
out[13] = byte(x3 >> 8)
out[14] = byte(x3 >> 16)
out[15] = byte(x3 >> 24)
out[16] = byte(x4)
out[17] = byte(x4 >> 8)
out[18] = byte(x4 >> 16)
out[19] = byte(x4 >> 24)
out[20] = byte(x5)
out[21] = byte(x5 >> 8)
out[22] = byte(x5 >> 16)
out[23] = byte(x5 >> 24)
out[24] = byte(x6)
out[25] = byte(x6 >> 8)
out[26] = byte(x6 >> 16)
out[27] = byte(x6 >> 24)
out[28] = byte(x7)
out[29] = byte(x7 >> 8)
out[30] = byte(x7 >> 16)
out[31] = byte(x7 >> 24)
out[32] = byte(x8)
out[33] = byte(x8 >> 8)
out[34] = byte(x8 >> 16)
out[35] = byte(x8 >> 24)
out[36] = byte(x9)
out[37] = byte(x9 >> 8)
out[38] = byte(x9 >> 16)
out[39] = byte(x9 >> 24)
out[40] = byte(x10)
out[41] = byte(x10 >> 8)
out[42] = byte(x10 >> 16)
out[43] = byte(x10 >> 24)
out[44] = byte(x11)
out[45] = byte(x11 >> 8)
out[46] = byte(x11 >> 16)
out[47] = byte(x11 >> 24)
out[48] = byte(x12)
out[49] = byte(x12 >> 8)
out[50] = byte(x12 >> 16)
out[51] = byte(x12 >> 24)
out[52] = byte(x13)
out[53] = byte(x13 >> 8)
out[54] = byte(x13 >> 16)
out[55] = byte(x13 >> 24)
out[56] = byte(x14)
out[57] = byte(x14 >> 8)
out[58] = byte(x14 >> 16)
out[59] = byte(x14 >> 24)
out[60] = byte(x15)
out[61] = byte(x15 >> 8)
out[62] = byte(x15 >> 16)
out[63] = byte(x15 >> 24)
}
// genericXORKeyStream is the generic implementation of XORKeyStream to be used
// when no assembly implementation is available.
func genericXORKeyStream(out, in []byte, counter *[16]byte, key *[32]byte) {
var block [64]byte
var counterCopy [16]byte
copy(counterCopy[:], counter[:])
for len(in) >= 64 {
core(&block, &counterCopy, key, &Sigma)
for i, x := range block {
out[i] = in[i] ^ x
}
u := uint32(1)
for i := 8; i < 16; i++ {
u += uint32(counterCopy[i])
counterCopy[i] = byte(u)
u >>= 8
}
in = in[64:]
out = out[64:]
}
if len(in) > 0 {
core(&block, &counterCopy, key, &Sigma)
for i, v := range in {
out[i] = v ^ block[i]
}
}
}