mirror/erigon-pulse
1
0
Fork 0
mirror of https://gitlab.com/pulsechaincom/erigon-pulse.git synced 2024-12-25 04:57:17 +00:00
Code Issues Actions Packages Projects Releases Wiki Activity
2e17b8f515
erigon-pulse/crypto/blake2b/blake2b_generic.go

162 lines
3.5 KiB
Go
Raw Normal View History

core/vm, crypto/blake2b: add BLAKE2b compression func at 0x09 The precompile at 0x09 wraps the BLAKE2b F compression function: https://tools.ietf.org/html/rfc7693#section-3.2 The precompile requires 6 inputs tightly encoded, taking exactly 213 bytes, as explained below. - `rounds` - the number of rounds - 32-bit unsigned big-endian word - `h` - the state vector - 8 unsigned 64-bit little-endian words - `m` - the message block vector - 16 unsigned 64-bit little-endian words - `t_0, t_1` - offset counters - 2 unsigned 64-bit little-endian words - `f` - the final block indicator flag - 8-bit word [4 bytes for rounds][64 bytes for h][128 bytes for m][8 bytes for t_0] [8 bytes for t_1][1 byte for f] The boolean `f` parameter is considered as `true` if set to `1`. The boolean `f` parameter is considered as `false` if set to `0`. All other values yield an invalid encoding of `f` error. The precompile should compute the F function as specified in the RFC (https://tools.ietf.org/html/rfc7693#section-3.2) and return the updated state vector `h` with unchanged encoding (little-endian). See EIP-152 for details.
2019-06-17 17:19:47 +00:00
// Copyright 2016 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 blake2b
import (
"math/bits"
)
// the precomputed values for BLAKE2b
// there are 10 16-byte arrays - one for each round
// the entries are calculated from the sigma constants.
var precomputed = [10][16]byte{
{0, 2, 4, 6, 1, 3, 5, 7, 8, 10, 12, 14, 9, 11, 13, 15},
{14, 4, 9, 13, 10, 8, 15, 6, 1, 0, 11, 5, 12, 2, 7, 3},
{11, 12, 5, 15, 8, 0, 2, 13, 10, 3, 7, 9, 14, 6, 1, 4},
{7, 3, 13, 11, 9, 1, 12, 14, 2, 5, 4, 15, 6, 10, 0, 8},
{9, 5, 2, 10, 0, 7, 4, 15, 14, 11, 6, 3, 1, 12, 8, 13},
{2, 6, 0, 8, 12, 10, 11, 3, 4, 7, 15, 1, 13, 5, 14, 9},
{12, 1, 14, 4, 5, 15, 13, 10, 0, 6, 9, 8, 7, 3, 2, 11},
{13, 7, 12, 3, 11, 14, 1, 9, 5, 15, 8, 2, 0, 4, 6, 10},
{6, 14, 11, 0, 15, 9, 3, 8, 12, 13, 1, 10, 2, 7, 4, 5},
{10, 8, 7, 1, 2, 4, 6, 5, 15, 9, 3, 13, 11, 14, 12, 0},
}
core/vm, crypto/blake2b: add SSE, AVX and AVX2 code
2019-08-21 09:38:18 +00:00
func fGeneric(h *[8]uint64, m *[16]uint64, c0, c1 uint64, flag uint64, rounds uint64) {
core/vm, crypto/blake2b: add BLAKE2b compression func at 0x09 The precompile at 0x09 wraps the BLAKE2b F compression function: https://tools.ietf.org/html/rfc7693#section-3.2 The precompile requires 6 inputs tightly encoded, taking exactly 213 bytes, as explained below. - `rounds` - the number of rounds - 32-bit unsigned big-endian word - `h` - the state vector - 8 unsigned 64-bit little-endian words - `m` - the message block vector - 16 unsigned 64-bit little-endian words - `t_0, t_1` - offset counters - 2 unsigned 64-bit little-endian words - `f` - the final block indicator flag - 8-bit word [4 bytes for rounds][64 bytes for h][128 bytes for m][8 bytes for t_0] [8 bytes for t_1][1 byte for f] The boolean `f` parameter is considered as `true` if set to `1`. The boolean `f` parameter is considered as `false` if set to `0`. All other values yield an invalid encoding of `f` error. The precompile should compute the F function as specified in the RFC (https://tools.ietf.org/html/rfc7693#section-3.2) and return the updated state vector `h` with unchanged encoding (little-endian). See EIP-152 for details.
2019-06-17 17:19:47 +00:00
v0, v1, v2, v3, v4, v5, v6, v7 := h[0], h[1], h[2], h[3], h[4], h[5], h[6], h[7]
core/vm, crypto/blake2b: add SSE, AVX and AVX2 code
2019-08-21 09:38:18 +00:00
v8, v9, v10, v11, v12, v13, v14, v15 := iv[0], iv[1], iv[2], iv[3], iv[4], iv[5], iv[6], iv[7]
core/vm, crypto/blake2b: add BLAKE2b compression func at 0x09 The precompile at 0x09 wraps the BLAKE2b F compression function: https://tools.ietf.org/html/rfc7693#section-3.2 The precompile requires 6 inputs tightly encoded, taking exactly 213 bytes, as explained below. - `rounds` - the number of rounds - 32-bit unsigned big-endian word - `h` - the state vector - 8 unsigned 64-bit little-endian words - `m` - the message block vector - 16 unsigned 64-bit little-endian words - `t_0, t_1` - offset counters - 2 unsigned 64-bit little-endian words - `f` - the final block indicator flag - 8-bit word [4 bytes for rounds][64 bytes for h][128 bytes for m][8 bytes for t_0] [8 bytes for t_1][1 byte for f] The boolean `f` parameter is considered as `true` if set to `1`. The boolean `f` parameter is considered as `false` if set to `0`. All other values yield an invalid encoding of `f` error. The precompile should compute the F function as specified in the RFC (https://tools.ietf.org/html/rfc7693#section-3.2) and return the updated state vector `h` with unchanged encoding (little-endian). See EIP-152 for details.
2019-06-17 17:19:47 +00:00
v12 ^= c0
v13 ^= c1
core/vm, crypto/blake2b: add SSE, AVX and AVX2 code
2019-08-21 09:38:18 +00:00
v14 ^= flag
core/vm, crypto/blake2b: add BLAKE2b compression func at 0x09 The precompile at 0x09 wraps the BLAKE2b F compression function: https://tools.ietf.org/html/rfc7693#section-3.2 The precompile requires 6 inputs tightly encoded, taking exactly 213 bytes, as explained below. - `rounds` - the number of rounds - 32-bit unsigned big-endian word - `h` - the state vector - 8 unsigned 64-bit little-endian words - `m` - the message block vector - 16 unsigned 64-bit little-endian words - `t_0, t_1` - offset counters - 2 unsigned 64-bit little-endian words - `f` - the final block indicator flag - 8-bit word [4 bytes for rounds][64 bytes for h][128 bytes for m][8 bytes for t_0] [8 bytes for t_1][1 byte for f] The boolean `f` parameter is considered as `true` if set to `1`. The boolean `f` parameter is considered as `false` if set to `0`. All other values yield an invalid encoding of `f` error. The precompile should compute the F function as specified in the RFC (https://tools.ietf.org/html/rfc7693#section-3.2) and return the updated state vector `h` with unchanged encoding (little-endian). See EIP-152 for details.
2019-06-17 17:19:47 +00:00
core/vm, crypto/blake2b: add SSE, AVX and AVX2 code
2019-08-21 09:38:18 +00:00
for i := 0; i < int(rounds); i++ {
s := &(precomputed[i%10])
core/vm, crypto/blake2b: add BLAKE2b compression func at 0x09 The precompile at 0x09 wraps the BLAKE2b F compression function: https://tools.ietf.org/html/rfc7693#section-3.2 The precompile requires 6 inputs tightly encoded, taking exactly 213 bytes, as explained below. - `rounds` - the number of rounds - 32-bit unsigned big-endian word - `h` - the state vector - 8 unsigned 64-bit little-endian words - `m` - the message block vector - 16 unsigned 64-bit little-endian words - `t_0, t_1` - offset counters - 2 unsigned 64-bit little-endian words - `f` - the final block indicator flag - 8-bit word [4 bytes for rounds][64 bytes for h][128 bytes for m][8 bytes for t_0] [8 bytes for t_1][1 byte for f] The boolean `f` parameter is considered as `true` if set to `1`. The boolean `f` parameter is considered as `false` if set to `0`. All other values yield an invalid encoding of `f` error. The precompile should compute the F function as specified in the RFC (https://tools.ietf.org/html/rfc7693#section-3.2) and return the updated state vector `h` with unchanged encoding (little-endian). See EIP-152 for details.
2019-06-17 17:19:47 +00:00
v0 += m[s[0]]
v0 += v4
v12 ^= v0
v12 = bits.RotateLeft64(v12, -32)
v8 += v12
v4 ^= v8
v4 = bits.RotateLeft64(v4, -24)
v1 += m[s[1]]
v1 += v5
v13 ^= v1
v13 = bits.RotateLeft64(v13, -32)
v9 += v13
v5 ^= v9
v5 = bits.RotateLeft64(v5, -24)
v2 += m[s[2]]
v2 += v6
v14 ^= v2
v14 = bits.RotateLeft64(v14, -32)
v10 += v14
v6 ^= v10
v6 = bits.RotateLeft64(v6, -24)
v3 += m[s[3]]
v3 += v7
v15 ^= v3
v15 = bits.RotateLeft64(v15, -32)
v11 += v15
v7 ^= v11
v7 = bits.RotateLeft64(v7, -24)
v0 += m[s[4]]
v0 += v4
v12 ^= v0
v12 = bits.RotateLeft64(v12, -16)
v8 += v12
v4 ^= v8
v4 = bits.RotateLeft64(v4, -63)
v1 += m[s[5]]
v1 += v5
v13 ^= v1
v13 = bits.RotateLeft64(v13, -16)
v9 += v13
v5 ^= v9
v5 = bits.RotateLeft64(v5, -63)
v2 += m[s[6]]
v2 += v6
v14 ^= v2
v14 = bits.RotateLeft64(v14, -16)
v10 += v14
v6 ^= v10
v6 = bits.RotateLeft64(v6, -63)
v3 += m[s[7]]
v3 += v7
v15 ^= v3
v15 = bits.RotateLeft64(v15, -16)
v11 += v15
v7 ^= v11
v7 = bits.RotateLeft64(v7, -63)
v0 += m[s[8]]
v0 += v5
v15 ^= v0
v15 = bits.RotateLeft64(v15, -32)
v10 += v15
v5 ^= v10
v5 = bits.RotateLeft64(v5, -24)
v1 += m[s[9]]
v1 += v6
v12 ^= v1
v12 = bits.RotateLeft64(v12, -32)
v11 += v12
v6 ^= v11
v6 = bits.RotateLeft64(v6, -24)
v2 += m[s[10]]
v2 += v7
v13 ^= v2
v13 = bits.RotateLeft64(v13, -32)
v8 += v13
v7 ^= v8
v7 = bits.RotateLeft64(v7, -24)
v3 += m[s[11]]
v3 += v4
v14 ^= v3
v14 = bits.RotateLeft64(v14, -32)
v9 += v14
v4 ^= v9
v4 = bits.RotateLeft64(v4, -24)
v0 += m[s[12]]
v0 += v5
v15 ^= v0
v15 = bits.RotateLeft64(v15, -16)
v10 += v15
v5 ^= v10
v5 = bits.RotateLeft64(v5, -63)
v1 += m[s[13]]
v1 += v6
v12 ^= v1
v12 = bits.RotateLeft64(v12, -16)
v11 += v12
v6 ^= v11
v6 = bits.RotateLeft64(v6, -63)
v2 += m[s[14]]
v2 += v7
v13 ^= v2
v13 = bits.RotateLeft64(v13, -16)
v8 += v13
v7 ^= v8
v7 = bits.RotateLeft64(v7, -63)
v3 += m[s[15]]
v3 += v4
v14 ^= v3
v14 = bits.RotateLeft64(v14, -16)
v9 += v14
v4 ^= v9
v4 = bits.RotateLeft64(v4, -63)
}
h[0] ^= v0 ^ v8
h[1] ^= v1 ^ v9
h[2] ^= v2 ^ v10
h[3] ^= v3 ^ v11
h[4] ^= v4 ^ v12
h[5] ^= v5 ^ v13
h[6] ^= v6 ^ v14
h[7] ^= v7 ^ v15
}
Reference in New Issue Copy Permalink