diff options
Diffstat (limited to 'src/crypto/sha256')
-rw-r--r-- | src/crypto/sha256/sha256.go | 193 | ||||
-rw-r--r-- | src/crypto/sha256/sha256_test.go | 176 | ||||
-rw-r--r-- | src/crypto/sha256/sha256block.go | 128 | ||||
-rw-r--r-- | src/crypto/sha256/sha256block_386.s | 283 | ||||
-rw-r--r-- | src/crypto/sha256/sha256block_amd64.s | 256 | ||||
-rw-r--r-- | src/crypto/sha256/sha256block_decl.go | 11 |
6 files changed, 1047 insertions, 0 deletions
diff --git a/src/crypto/sha256/sha256.go b/src/crypto/sha256/sha256.go new file mode 100644 index 000000000..d84cebf2f --- /dev/null +++ b/src/crypto/sha256/sha256.go @@ -0,0 +1,193 @@ +// Copyright 2009 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 sha256 implements the SHA224 and SHA256 hash algorithms as defined +// in FIPS 180-4. +package sha256 + +import ( + "crypto" + "hash" +) + +func init() { + crypto.RegisterHash(crypto.SHA224, New224) + crypto.RegisterHash(crypto.SHA256, New) +} + +// The size of a SHA256 checksum in bytes. +const Size = 32 + +// The size of a SHA224 checksum in bytes. +const Size224 = 28 + +// The blocksize of SHA256 and SHA224 in bytes. +const BlockSize = 64 + +const ( + chunk = 64 + init0 = 0x6A09E667 + init1 = 0xBB67AE85 + init2 = 0x3C6EF372 + init3 = 0xA54FF53A + init4 = 0x510E527F + init5 = 0x9B05688C + init6 = 0x1F83D9AB + init7 = 0x5BE0CD19 + init0_224 = 0xC1059ED8 + init1_224 = 0x367CD507 + init2_224 = 0x3070DD17 + init3_224 = 0xF70E5939 + init4_224 = 0xFFC00B31 + init5_224 = 0x68581511 + init6_224 = 0x64F98FA7 + init7_224 = 0xBEFA4FA4 +) + +// digest represents the partial evaluation of a checksum. +type digest struct { + h [8]uint32 + x [chunk]byte + nx int + len uint64 + is224 bool // mark if this digest is SHA-224 +} + +func (d *digest) Reset() { + if !d.is224 { + d.h[0] = init0 + d.h[1] = init1 + d.h[2] = init2 + d.h[3] = init3 + d.h[4] = init4 + d.h[5] = init5 + d.h[6] = init6 + d.h[7] = init7 + } else { + d.h[0] = init0_224 + d.h[1] = init1_224 + d.h[2] = init2_224 + d.h[3] = init3_224 + d.h[4] = init4_224 + d.h[5] = init5_224 + d.h[6] = init6_224 + d.h[7] = init7_224 + } + d.nx = 0 + d.len = 0 +} + +// New returns a new hash.Hash computing the SHA256 checksum. +func New() hash.Hash { + d := new(digest) + d.Reset() + return d +} + +// New224 returns a new hash.Hash computing the SHA224 checksum. +func New224() hash.Hash { + d := new(digest) + d.is224 = true + d.Reset() + return d +} + +func (d *digest) Size() int { + if !d.is224 { + return Size + } + return Size224 +} + +func (d *digest) BlockSize() int { return BlockSize } + +func (d *digest) Write(p []byte) (nn int, err error) { + nn = len(p) + d.len += uint64(nn) + if d.nx > 0 { + n := copy(d.x[d.nx:], p) + d.nx += n + if d.nx == chunk { + block(d, d.x[:]) + d.nx = 0 + } + p = p[n:] + } + if len(p) >= chunk { + n := len(p) &^ (chunk - 1) + block(d, p[:n]) + p = p[n:] + } + if len(p) > 0 { + d.nx = copy(d.x[:], p) + } + return +} + +func (d0 *digest) Sum(in []byte) []byte { + // Make a copy of d0 so that caller can keep writing and summing. + d := *d0 + hash := d.checkSum() + if d.is224 { + return append(in, hash[:Size224]...) + } + return append(in, hash[:]...) +} + +func (d *digest) checkSum() [Size]byte { + len := d.len + // Padding. Add a 1 bit and 0 bits until 56 bytes mod 64. + var tmp [64]byte + tmp[0] = 0x80 + if len%64 < 56 { + d.Write(tmp[0 : 56-len%64]) + } else { + d.Write(tmp[0 : 64+56-len%64]) + } + + // Length in bits. + len <<= 3 + for i := uint(0); i < 8; i++ { + tmp[i] = byte(len >> (56 - 8*i)) + } + d.Write(tmp[0:8]) + + if d.nx != 0 { + panic("d.nx != 0") + } + + h := d.h[:] + if d.is224 { + h = d.h[:7] + } + + var digest [Size]byte + for i, s := range h { + digest[i*4] = byte(s >> 24) + digest[i*4+1] = byte(s >> 16) + digest[i*4+2] = byte(s >> 8) + digest[i*4+3] = byte(s) + } + + return digest +} + +// Sum256 returns the SHA256 checksum of the data. +func Sum256(data []byte) [Size]byte { + var d digest + d.Reset() + d.Write(data) + return d.checkSum() +} + +// Sum224 returns the SHA224 checksum of the data. +func Sum224(data []byte) (sum224 [Size224]byte) { + var d digest + d.is224 = true + d.Reset() + d.Write(data) + sum := d.checkSum() + copy(sum224[:], sum[:Size224]) + return +} diff --git a/src/crypto/sha256/sha256_test.go b/src/crypto/sha256/sha256_test.go new file mode 100644 index 000000000..1d883d390 --- /dev/null +++ b/src/crypto/sha256/sha256_test.go @@ -0,0 +1,176 @@ +// Copyright 2009 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. + +// SHA256 hash algorithm. See FIPS 180-2. + +package sha256 + +import ( + "fmt" + "io" + "testing" +) + +type sha256Test struct { + out string + in string +} + +var golden = []sha256Test{ + {"e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", ""}, + {"ca978112ca1bbdcafac231b39a23dc4da786eff8147c4e72b9807785afee48bb", "a"}, + {"fb8e20fc2e4c3f248c60c39bd652f3c1347298bb977b8b4d5903b85055620603", "ab"}, + {"ba7816bf8f01cfea414140de5dae2223b00361a396177a9cb410ff61f20015ad", "abc"}, + {"88d4266fd4e6338d13b845fcf289579d209c897823b9217da3e161936f031589", "abcd"}, + {"36bbe50ed96841d10443bcb670d6554f0a34b761be67ec9c4a8ad2c0c44ca42c", "abcde"}, + {"bef57ec7f53a6d40beb640a780a639c83bc29ac8a9816f1fc6c5c6dcd93c4721", "abcdef"}, + {"7d1a54127b222502f5b79b5fb0803061152a44f92b37e23c6527baf665d4da9a", "abcdefg"}, + {"9c56cc51b374c3ba189210d5b6d4bf57790d351c96c47c02190ecf1e430635ab", "abcdefgh"}, + {"19cc02f26df43cc571bc9ed7b0c4d29224a3ec229529221725ef76d021c8326f", "abcdefghi"}, + {"72399361da6a7754fec986dca5b7cbaf1c810a28ded4abaf56b2106d06cb78b0", "abcdefghij"}, + {"a144061c271f152da4d151034508fed1c138b8c976339de229c3bb6d4bbb4fce", "Discard medicine more than two years old."}, + {"6dae5caa713a10ad04b46028bf6dad68837c581616a1589a265a11288d4bb5c4", "He who has a shady past knows that nice guys finish last."}, + {"ae7a702a9509039ddbf29f0765e70d0001177914b86459284dab8b348c2dce3f", "I wouldn't marry him with a ten foot pole."}, + {"6748450b01c568586715291dfa3ee018da07d36bb7ea6f180c1af6270215c64f", "Free! Free!/A trip/to Mars/for 900/empty jars/Burma Shave"}, + {"14b82014ad2b11f661b5ae6a99b75105c2ffac278cd071cd6c05832793635774", "The days of the digital watch are numbered. -Tom Stoppard"}, + {"7102cfd76e2e324889eece5d6c41921b1e142a4ac5a2692be78803097f6a48d8", "Nepal premier won't resign."}, + {"23b1018cd81db1d67983c5f7417c44da9deb582459e378d7a068552ea649dc9f", "For every action there is an equal and opposite government program."}, + {"8001f190dfb527261c4cfcab70c98e8097a7a1922129bc4096950e57c7999a5a", "His money is twice tainted: 'taint yours and 'taint mine."}, + {"8c87deb65505c3993eb24b7a150c4155e82eee6960cf0c3a8114ff736d69cad5", "There is no reason for any individual to have a computer in their home. -Ken Olsen, 1977"}, + {"bfb0a67a19cdec3646498b2e0f751bddc41bba4b7f30081b0b932aad214d16d7", "It's a tiny change to the code and not completely disgusting. - Bob Manchek"}, + {"7f9a0b9bf56332e19f5a0ec1ad9c1425a153da1c624868fda44561d6b74daf36", "size: a.out: bad magic"}, + {"b13f81b8aad9e3666879af19886140904f7f429ef083286195982a7588858cfc", "The major problem is with sendmail. -Mark Horton"}, + {"b26c38d61519e894480c70c8374ea35aa0ad05b2ae3d6674eec5f52a69305ed4", "Give me a rock, paper and scissors and I will move the world. CCFestoon"}, + {"049d5e26d4f10222cd841a119e38bd8d2e0d1129728688449575d4ff42b842c1", "If the enemy is within range, then so are you."}, + {"0e116838e3cc1c1a14cd045397e29b4d087aa11b0853fc69ec82e90330d60949", "It's well we cannot hear the screams/That we create in others' dreams."}, + {"4f7d8eb5bcf11de2a56b971021a444aa4eafd6ecd0f307b5109e4e776cd0fe46", "You remind me of a TV show, but that's all right: I watch it anyway."}, + {"61c0cc4c4bd8406d5120b3fb4ebc31ce87667c162f29468b3c779675a85aebce", "C is as portable as Stonehedge!!"}, + {"1fb2eb3688093c4a3f80cd87a5547e2ce940a4f923243a79a2a1e242220693ac", "Even if I could be Shakespeare, I think I should still choose to be Faraday. - A. Huxley"}, + {"395585ce30617b62c80b93e8208ce866d4edc811a177fdb4b82d3911d8696423", "The fugacity of a constituent in a mixture of gases at a given temperature is proportional to its mole fraction. Lewis-Randall Rule"}, + {"4f9b189a13d030838269dce846b16a1ce9ce81fe63e65de2f636863336a98fe6", "How can you write a big system without C++? -Paul Glick"}, +} + +var golden224 = []sha256Test{ + {"d14a028c2a3a2bc9476102bb288234c415a2b01f828ea62ac5b3e42f", ""}, + {"abd37534c7d9a2efb9465de931cd7055ffdb8879563ae98078d6d6d5", "a"}, + {"db3cda86d4429a1d39c148989566b38f7bda0156296bd364ba2f878b", "ab"}, + {"23097d223405d8228642a477bda255b32aadbce4bda0b3f7e36c9da7", "abc"}, + {"a76654d8e3550e9a2d67a0eeb6c67b220e5885eddd3fde135806e601", "abcd"}, + {"bdd03d560993e675516ba5a50638b6531ac2ac3d5847c61916cfced6", "abcde"}, + {"7043631cb415556a275a4ebecb802c74ee9f6153908e1792a90b6a98", "abcdef"}, + {"d1884e711701ad81abe0c77a3b0ea12e19ba9af64077286c72fc602d", "abcdefg"}, + {"17eb7d40f0356f8598e89eafad5f6c759b1f822975d9c9b737c8a517", "abcdefgh"}, + {"aeb35915346c584db820d2de7af3929ffafef9222a9bcb26516c7334", "abcdefghi"}, + {"d35e1e5af29ddb0d7e154357df4ad9842afee527c689ee547f753188", "abcdefghij"}, + {"19297f1cef7ddc8a7e947f5c5a341e10f7245045e425db67043988d7", "Discard medicine more than two years old."}, + {"0f10c2eb436251f777fbbd125e260d36aecf180411726c7c885f599a", "He who has a shady past knows that nice guys finish last."}, + {"4d1842104919f314cad8a3cd20b3cba7e8ed3e7abed62b57441358f6", "I wouldn't marry him with a ten foot pole."}, + {"a8ba85c6fe0c48fbffc72bbb2f03fcdbc87ae2dc7a56804d1590fb3b", "Free! Free!/A trip/to Mars/for 900/empty jars/Burma Shave"}, + {"5543fbab26e67e8885b1a852d567d1cb8b9bfe42e0899584c50449a9", "The days of the digital watch are numbered. -Tom Stoppard"}, + {"65ca107390f5da9efa05d28e57b221657edc7e43a9a18fb15b053ddb", "Nepal premier won't resign."}, + {"84953962be366305a9cc9b5cd16ed019edc37ac96c0deb3e12cca116", "For every action there is an equal and opposite government program."}, + {"35a189ce987151dfd00b3577583cc6a74b9869eecf894459cb52038d", "His money is twice tainted: 'taint yours and 'taint mine."}, + {"2fc333713983edfd4ef2c0da6fb6d6415afb94987c91e4069eb063e6", "There is no reason for any individual to have a computer in their home. -Ken Olsen, 1977"}, + {"cbe32d38d577a1b355960a4bc3c659c2dc4670859a19777a875842c4", "It's a tiny change to the code and not completely disgusting. - Bob Manchek"}, + {"a2dc118ce959e027576413a7b440c875cdc8d40df9141d6ef78a57e1", "size: a.out: bad magic"}, + {"d10787e24052bcff26dc484787a54ed819e4e4511c54890ee977bf81", "The major problem is with sendmail. -Mark Horton"}, + {"62efcf16ab8a893acdf2f348aaf06b63039ff1bf55508c830532c9fb", "Give me a rock, paper and scissors and I will move the world. CCFestoon"}, + {"3e9b7e4613c59f58665104c5fa86c272db5d3a2ff30df5bb194a5c99", "If the enemy is within range, then so are you."}, + {"5999c208b8bdf6d471bb7c359ac5b829e73a8211dff686143a4e7f18", "It's well we cannot hear the screams/That we create in others' dreams."}, + {"3b2d67ff54eabc4ef737b14edf87c64280ef582bcdf2a6d56908b405", "You remind me of a TV show, but that's all right: I watch it anyway."}, + {"d0733595d20e4d3d6b5c565a445814d1bbb2fd08b9a3b8ffb97930c6", "C is as portable as Stonehedge!!"}, + {"43fb8aeed8a833175c9295c1165415f98c866ef08a4922959d673507", "Even if I could be Shakespeare, I think I should still choose to be Faraday. - A. Huxley"}, + {"ec18e66e93afc4fb1604bc2baedbfd20b44c43d76e65c0996d7851c6", "The fugacity of a constituent in a mixture of gases at a given temperature is proportional to its mole fraction. Lewis-Randall Rule"}, + {"86ed2eaa9c75ba98396e5c9fb2f679ecf0ea2ed1e0ee9ceecb4a9332", "How can you write a big system without C++? -Paul Glick"}, +} + +func TestGolden(t *testing.T) { + for i := 0; i < len(golden); i++ { + g := golden[i] + s := fmt.Sprintf("%x", Sum256([]byte(g.in))) + if s != g.out { + t.Fatalf("Sum256 function: sha256(%s) = %s want %s", g.in, s, g.out) + } + c := New() + for j := 0; j < 3; j++ { + if j < 2 { + io.WriteString(c, g.in) + } else { + io.WriteString(c, g.in[0:len(g.in)/2]) + c.Sum(nil) + io.WriteString(c, g.in[len(g.in)/2:]) + } + s := fmt.Sprintf("%x", c.Sum(nil)) + if s != g.out { + t.Fatalf("sha256[%d](%s) = %s want %s", j, g.in, s, g.out) + } + c.Reset() + } + } + for i := 0; i < len(golden224); i++ { + g := golden224[i] + s := fmt.Sprintf("%x", Sum224([]byte(g.in))) + if s != g.out { + t.Fatalf("Sum224 function: sha224(%s) = %s want %s", g.in, s, g.out) + } + c := New224() + for j := 0; j < 3; j++ { + if j < 2 { + io.WriteString(c, g.in) + } else { + io.WriteString(c, g.in[0:len(g.in)/2]) + c.Sum(nil) + io.WriteString(c, g.in[len(g.in)/2:]) + } + s := fmt.Sprintf("%x", c.Sum(nil)) + if s != g.out { + t.Fatalf("sha224[%d](%s) = %s want %s", j, g.in, s, g.out) + } + c.Reset() + } + } +} + +func TestSize(t *testing.T) { + c := New() + if got := c.Size(); got != Size { + t.Errorf("Size = %d; want %d", got, Size) + } + c = New224() + if got := c.Size(); got != Size224 { + t.Errorf("New224.Size = %d; want %d", got, Size224) + } +} + +func TestBlockSize(t *testing.T) { + c := New() + if got := c.BlockSize(); got != BlockSize { + t.Errorf("BlockSize = %d want %d", got, BlockSize) + } +} + +var bench = New() +var buf = make([]byte, 8192) + +func benchmarkSize(b *testing.B, size int) { + b.SetBytes(int64(size)) + sum := make([]byte, bench.Size()) + for i := 0; i < b.N; i++ { + bench.Reset() + bench.Write(buf[:size]) + bench.Sum(sum[:0]) + } +} + +func BenchmarkHash8Bytes(b *testing.B) { + benchmarkSize(b, 8) +} + +func BenchmarkHash1K(b *testing.B) { + benchmarkSize(b, 1024) +} + +func BenchmarkHash8K(b *testing.B) { + benchmarkSize(b, 8192) +} diff --git a/src/crypto/sha256/sha256block.go b/src/crypto/sha256/sha256block.go new file mode 100644 index 000000000..ca5efd156 --- /dev/null +++ b/src/crypto/sha256/sha256block.go @@ -0,0 +1,128 @@ +// Copyright 2009 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 !386,!amd64 + +// SHA256 block step. +// In its own file so that a faster assembly or C version +// can be substituted easily. + +package sha256 + +var _K = []uint32{ + 0x428a2f98, + 0x71374491, + 0xb5c0fbcf, + 0xe9b5dba5, + 0x3956c25b, + 0x59f111f1, + 0x923f82a4, + 0xab1c5ed5, + 0xd807aa98, + 0x12835b01, + 0x243185be, + 0x550c7dc3, + 0x72be5d74, + 0x80deb1fe, + 0x9bdc06a7, + 0xc19bf174, + 0xe49b69c1, + 0xefbe4786, + 0x0fc19dc6, + 0x240ca1cc, + 0x2de92c6f, + 0x4a7484aa, + 0x5cb0a9dc, + 0x76f988da, + 0x983e5152, + 0xa831c66d, + 0xb00327c8, + 0xbf597fc7, + 0xc6e00bf3, + 0xd5a79147, + 0x06ca6351, + 0x14292967, + 0x27b70a85, + 0x2e1b2138, + 0x4d2c6dfc, + 0x53380d13, + 0x650a7354, + 0x766a0abb, + 0x81c2c92e, + 0x92722c85, + 0xa2bfe8a1, + 0xa81a664b, + 0xc24b8b70, + 0xc76c51a3, + 0xd192e819, + 0xd6990624, + 0xf40e3585, + 0x106aa070, + 0x19a4c116, + 0x1e376c08, + 0x2748774c, + 0x34b0bcb5, + 0x391c0cb3, + 0x4ed8aa4a, + 0x5b9cca4f, + 0x682e6ff3, + 0x748f82ee, + 0x78a5636f, + 0x84c87814, + 0x8cc70208, + 0x90befffa, + 0xa4506ceb, + 0xbef9a3f7, + 0xc67178f2, +} + +func block(dig *digest, p []byte) { + var w [64]uint32 + h0, h1, h2, h3, h4, h5, h6, h7 := dig.h[0], dig.h[1], dig.h[2], dig.h[3], dig.h[4], dig.h[5], dig.h[6], dig.h[7] + for len(p) >= chunk { + // Can interlace the computation of w with the + // rounds below if needed for speed. + for i := 0; i < 16; i++ { + j := i * 4 + w[i] = uint32(p[j])<<24 | uint32(p[j+1])<<16 | uint32(p[j+2])<<8 | uint32(p[j+3]) + } + for i := 16; i < 64; i++ { + v1 := w[i-2] + t1 := (v1>>17 | v1<<(32-17)) ^ (v1>>19 | v1<<(32-19)) ^ (v1 >> 10) + v2 := w[i-15] + t2 := (v2>>7 | v2<<(32-7)) ^ (v2>>18 | v2<<(32-18)) ^ (v2 >> 3) + w[i] = t1 + w[i-7] + t2 + w[i-16] + } + + a, b, c, d, e, f, g, h := h0, h1, h2, h3, h4, h5, h6, h7 + + for i := 0; i < 64; i++ { + t1 := h + ((e>>6 | e<<(32-6)) ^ (e>>11 | e<<(32-11)) ^ (e>>25 | e<<(32-25))) + ((e & f) ^ (^e & g)) + _K[i] + w[i] + + t2 := ((a>>2 | a<<(32-2)) ^ (a>>13 | a<<(32-13)) ^ (a>>22 | a<<(32-22))) + ((a & b) ^ (a & c) ^ (b & c)) + + h = g + g = f + f = e + e = d + t1 + d = c + c = b + b = a + a = t1 + t2 + } + + h0 += a + h1 += b + h2 += c + h3 += d + h4 += e + h5 += f + h6 += g + h7 += h + + p = p[chunk:] + } + + dig.h[0], dig.h[1], dig.h[2], dig.h[3], dig.h[4], dig.h[5], dig.h[6], dig.h[7] = h0, h1, h2, h3, h4, h5, h6, h7 +} diff --git a/src/crypto/sha256/sha256block_386.s b/src/crypto/sha256/sha256block_386.s new file mode 100644 index 000000000..73ae2bf30 --- /dev/null +++ b/src/crypto/sha256/sha256block_386.s @@ -0,0 +1,283 @@ +// Copyright 2013 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. + +// SHA256 block routine. See sha256block.go for Go equivalent. +// +// The algorithm is detailed in FIPS 180-4: +// +// http://csrc.nist.gov/publications/fips/fips180-4/fips-180-4.pdf +// +// Wt = Mt; for 0 <= t <= 15 +// Wt = SIGMA1(Wt-2) + SIGMA0(Wt-15) + Wt-16; for 16 <= t <= 63 +// +// a = H0 +// b = H1 +// c = H2 +// d = H3 +// e = H4 +// f = H5 +// g = H6 +// h = H7 +// +// for t = 0 to 63 { +// T1 = h + BIGSIGMA1(e) + Ch(e,f,g) + Kt + Wt +// T2 = BIGSIGMA0(a) + Maj(a,b,c) +// h = g +// g = f +// f = e +// e = d + T1 +// d = c +// c = b +// b = a +// a = T1 + T2 +// } +// +// H0 = a + H0 +// H1 = b + H1 +// H2 = c + H2 +// H3 = d + H3 +// H4 = e + H4 +// H5 = f + H5 +// H6 = g + H6 +// H7 = h + H7 + +// Wt = Mt; for 0 <= t <= 15 +#define MSGSCHEDULE0(index) \ + MOVL (index*4)(SI), AX; \ + BSWAPL AX; \ + MOVL AX, (index*4)(BP) + +// Wt = SIGMA1(Wt-2) + Wt-7 + SIGMA0(Wt-15) + Wt-16; for 16 <= t <= 63 +// SIGMA0(x) = ROTR(7,x) XOR ROTR(18,x) XOR SHR(3,x) +// SIGMA1(x) = ROTR(17,x) XOR ROTR(19,x) XOR SHR(10,x) +#define MSGSCHEDULE1(index) \ + MOVL ((index-2)*4)(BP), AX; \ + MOVL AX, CX; \ + RORL $17, AX; \ + MOVL CX, DX; \ + RORL $19, CX; \ + SHRL $10, DX; \ + MOVL ((index-15)*4)(BP), BX; \ + XORL CX, AX; \ + MOVL BX, CX; \ + XORL DX, AX; \ + RORL $7, BX; \ + MOVL CX, DX; \ + SHRL $3, DX; \ + RORL $18, CX; \ + ADDL ((index-7)*4)(BP), AX; \ + XORL CX, BX; \ + XORL DX, BX; \ + ADDL ((index-16)*4)(BP), BX; \ + ADDL BX, AX; \ + MOVL AX, ((index)*4)(BP) + +// Calculate T1 in AX - uses AX, BX, CX and DX registers. +// Wt is passed in AX. +// T1 = h + BIGSIGMA1(e) + Ch(e, f, g) + Kt + Wt +// BIGSIGMA1(x) = ROTR(6,x) XOR ROTR(11,x) XOR ROTR(25,x) +// Ch(x, y, z) = (x AND y) XOR (NOT x AND z) +#define SHA256T1(const, e, f, g, h) \ + MOVL (h*4)(DI), BX; \ + ADDL AX, BX; \ + MOVL (e*4)(DI), AX; \ + ADDL $const, BX; \ + MOVL (e*4)(DI), CX; \ + RORL $6, AX; \ + MOVL (e*4)(DI), DX; \ + RORL $11, CX; \ + XORL CX, AX; \ + MOVL (e*4)(DI), CX; \ + RORL $25, DX; \ + ANDL (f*4)(DI), CX; \ + XORL AX, DX; \ + MOVL (e*4)(DI), AX; \ + NOTL AX; \ + ADDL DX, BX; \ + ANDL (g*4)(DI), AX; \ + XORL CX, AX; \ + ADDL BX, AX + +// Calculate T2 in BX - uses AX, BX, CX and DX registers. +// T2 = BIGSIGMA0(a) + Maj(a, b, c) +// BIGSIGMA0(x) = ROTR(2,x) XOR ROTR(13,x) XOR ROTR(22,x) +// Maj(x, y, z) = (x AND y) XOR (x AND z) XOR (y AND z) +#define SHA256T2(a, b, c) \ + MOVL (a*4)(DI), AX; \ + MOVL (c*4)(DI), BX; \ + RORL $2, AX; \ + MOVL (a*4)(DI), DX; \ + ANDL (b*4)(DI), BX; \ + RORL $13, DX; \ + MOVL (a*4)(DI), CX; \ + ANDL (c*4)(DI), CX; \ + XORL DX, AX; \ + XORL CX, BX; \ + MOVL (a*4)(DI), DX; \ + MOVL (b*4)(DI), CX; \ + RORL $22, DX; \ + ANDL (a*4)(DI), CX; \ + XORL CX, BX; \ + XORL DX, AX; \ + ADDL AX, BX + +// Calculate T1 and T2, then e = d + T1 and a = T1 + T2. +// The values for e and a are stored in d and h, ready for rotation. +#define SHA256ROUND(index, const, a, b, c, d, e, f, g, h) \ + SHA256T1(const, e, f, g, h); \ + MOVL AX, 292(SP); \ + SHA256T2(a, b, c); \ + MOVL 292(SP), AX; \ + ADDL AX, BX; \ + ADDL AX, (d*4)(DI); \ + MOVL BX, (h*4)(DI) + +#define SHA256ROUND0(index, const, a, b, c, d, e, f, g, h) \ + MSGSCHEDULE0(index); \ + SHA256ROUND(index, const, a, b, c, d, e, f, g, h) + +#define SHA256ROUND1(index, const, a, b, c, d, e, f, g, h) \ + MSGSCHEDULE1(index); \ + SHA256ROUND(index, const, a, b, c, d, e, f, g, h) + +TEXT ·block(SB),0,$296-12 + MOVL p_base+4(FP), SI + MOVL p_len+8(FP), DX + SHRL $6, DX + SHLL $6, DX + + LEAL (SI)(DX*1), DI + MOVL DI, 288(SP) + CMPL SI, DI + JEQ end + + LEAL 256(SP), DI // variables + + MOVL dig+0(FP), BP + MOVL (0*4)(BP), AX // a = H0 + MOVL AX, (0*4)(DI) + MOVL (1*4)(BP), BX // b = H1 + MOVL BX, (1*4)(DI) + MOVL (2*4)(BP), CX // c = H2 + MOVL CX, (2*4)(DI) + MOVL (3*4)(BP), DX // d = H3 + MOVL DX, (3*4)(DI) + MOVL (4*4)(BP), AX // e = H4 + MOVL AX, (4*4)(DI) + MOVL (5*4)(BP), BX // f = H5 + MOVL BX, (5*4)(DI) + MOVL (6*4)(BP), CX // g = H6 + MOVL CX, (6*4)(DI) + MOVL (7*4)(BP), DX // h = H7 + MOVL DX, (7*4)(DI) + +loop: + MOVL SP, BP // message schedule + + SHA256ROUND0(0, 0x428a2f98, 0, 1, 2, 3, 4, 5, 6, 7) + SHA256ROUND0(1, 0x71374491, 7, 0, 1, 2, 3, 4, 5, 6) + SHA256ROUND0(2, 0xb5c0fbcf, 6, 7, 0, 1, 2, 3, 4, 5) + SHA256ROUND0(3, 0xe9b5dba5, 5, 6, 7, 0, 1, 2, 3, 4) + SHA256ROUND0(4, 0x3956c25b, 4, 5, 6, 7, 0, 1, 2, 3) + SHA256ROUND0(5, 0x59f111f1, 3, 4, 5, 6, 7, 0, 1, 2) + SHA256ROUND0(6, 0x923f82a4, 2, 3, 4, 5, 6, 7, 0, 1) + SHA256ROUND0(7, 0xab1c5ed5, 1, 2, 3, 4, 5, 6, 7, 0) + SHA256ROUND0(8, 0xd807aa98, 0, 1, 2, 3, 4, 5, 6, 7) + SHA256ROUND0(9, 0x12835b01, 7, 0, 1, 2, 3, 4, 5, 6) + SHA256ROUND0(10, 0x243185be, 6, 7, 0, 1, 2, 3, 4, 5) + SHA256ROUND0(11, 0x550c7dc3, 5, 6, 7, 0, 1, 2, 3, 4) + SHA256ROUND0(12, 0x72be5d74, 4, 5, 6, 7, 0, 1, 2, 3) + SHA256ROUND0(13, 0x80deb1fe, 3, 4, 5, 6, 7, 0, 1, 2) + SHA256ROUND0(14, 0x9bdc06a7, 2, 3, 4, 5, 6, 7, 0, 1) + SHA256ROUND0(15, 0xc19bf174, 1, 2, 3, 4, 5, 6, 7, 0) + + SHA256ROUND1(16, 0xe49b69c1, 0, 1, 2, 3, 4, 5, 6, 7) + SHA256ROUND1(17, 0xefbe4786, 7, 0, 1, 2, 3, 4, 5, 6) + SHA256ROUND1(18, 0x0fc19dc6, 6, 7, 0, 1, 2, 3, 4, 5) + SHA256ROUND1(19, 0x240ca1cc, 5, 6, 7, 0, 1, 2, 3, 4) + SHA256ROUND1(20, 0x2de92c6f, 4, 5, 6, 7, 0, 1, 2, 3) + SHA256ROUND1(21, 0x4a7484aa, 3, 4, 5, 6, 7, 0, 1, 2) + SHA256ROUND1(22, 0x5cb0a9dc, 2, 3, 4, 5, 6, 7, 0, 1) + SHA256ROUND1(23, 0x76f988da, 1, 2, 3, 4, 5, 6, 7, 0) + SHA256ROUND1(24, 0x983e5152, 0, 1, 2, 3, 4, 5, 6, 7) + SHA256ROUND1(25, 0xa831c66d, 7, 0, 1, 2, 3, 4, 5, 6) + SHA256ROUND1(26, 0xb00327c8, 6, 7, 0, 1, 2, 3, 4, 5) + SHA256ROUND1(27, 0xbf597fc7, 5, 6, 7, 0, 1, 2, 3, 4) + SHA256ROUND1(28, 0xc6e00bf3, 4, 5, 6, 7, 0, 1, 2, 3) + SHA256ROUND1(29, 0xd5a79147, 3, 4, 5, 6, 7, 0, 1, 2) + SHA256ROUND1(30, 0x06ca6351, 2, 3, 4, 5, 6, 7, 0, 1) + SHA256ROUND1(31, 0x14292967, 1, 2, 3, 4, 5, 6, 7, 0) + SHA256ROUND1(32, 0x27b70a85, 0, 1, 2, 3, 4, 5, 6, 7) + SHA256ROUND1(33, 0x2e1b2138, 7, 0, 1, 2, 3, 4, 5, 6) + SHA256ROUND1(34, 0x4d2c6dfc, 6, 7, 0, 1, 2, 3, 4, 5) + SHA256ROUND1(35, 0x53380d13, 5, 6, 7, 0, 1, 2, 3, 4) + SHA256ROUND1(36, 0x650a7354, 4, 5, 6, 7, 0, 1, 2, 3) + SHA256ROUND1(37, 0x766a0abb, 3, 4, 5, 6, 7, 0, 1, 2) + SHA256ROUND1(38, 0x81c2c92e, 2, 3, 4, 5, 6, 7, 0, 1) + SHA256ROUND1(39, 0x92722c85, 1, 2, 3, 4, 5, 6, 7, 0) + SHA256ROUND1(40, 0xa2bfe8a1, 0, 1, 2, 3, 4, 5, 6, 7) + SHA256ROUND1(41, 0xa81a664b, 7, 0, 1, 2, 3, 4, 5, 6) + SHA256ROUND1(42, 0xc24b8b70, 6, 7, 0, 1, 2, 3, 4, 5) + SHA256ROUND1(43, 0xc76c51a3, 5, 6, 7, 0, 1, 2, 3, 4) + SHA256ROUND1(44, 0xd192e819, 4, 5, 6, 7, 0, 1, 2, 3) + SHA256ROUND1(45, 0xd6990624, 3, 4, 5, 6, 7, 0, 1, 2) + SHA256ROUND1(46, 0xf40e3585, 2, 3, 4, 5, 6, 7, 0, 1) + SHA256ROUND1(47, 0x106aa070, 1, 2, 3, 4, 5, 6, 7, 0) + SHA256ROUND1(48, 0x19a4c116, 0, 1, 2, 3, 4, 5, 6, 7) + SHA256ROUND1(49, 0x1e376c08, 7, 0, 1, 2, 3, 4, 5, 6) + SHA256ROUND1(50, 0x2748774c, 6, 7, 0, 1, 2, 3, 4, 5) + SHA256ROUND1(51, 0x34b0bcb5, 5, 6, 7, 0, 1, 2, 3, 4) + SHA256ROUND1(52, 0x391c0cb3, 4, 5, 6, 7, 0, 1, 2, 3) + SHA256ROUND1(53, 0x4ed8aa4a, 3, 4, 5, 6, 7, 0, 1, 2) + SHA256ROUND1(54, 0x5b9cca4f, 2, 3, 4, 5, 6, 7, 0, 1) + SHA256ROUND1(55, 0x682e6ff3, 1, 2, 3, 4, 5, 6, 7, 0) + SHA256ROUND1(56, 0x748f82ee, 0, 1, 2, 3, 4, 5, 6, 7) + SHA256ROUND1(57, 0x78a5636f, 7, 0, 1, 2, 3, 4, 5, 6) + SHA256ROUND1(58, 0x84c87814, 6, 7, 0, 1, 2, 3, 4, 5) + SHA256ROUND1(59, 0x8cc70208, 5, 6, 7, 0, 1, 2, 3, 4) + SHA256ROUND1(60, 0x90befffa, 4, 5, 6, 7, 0, 1, 2, 3) + SHA256ROUND1(61, 0xa4506ceb, 3, 4, 5, 6, 7, 0, 1, 2) + SHA256ROUND1(62, 0xbef9a3f7, 2, 3, 4, 5, 6, 7, 0, 1) + SHA256ROUND1(63, 0xc67178f2, 1, 2, 3, 4, 5, 6, 7, 0) + + MOVL dig+0(FP), BP + MOVL (0*4)(BP), AX // H0 = a + H0 + ADDL (0*4)(DI), AX + MOVL AX, (0*4)(DI) + MOVL AX, (0*4)(BP) + MOVL (1*4)(BP), BX // H1 = b + H1 + ADDL (1*4)(DI), BX + MOVL BX, (1*4)(DI) + MOVL BX, (1*4)(BP) + MOVL (2*4)(BP), CX // H2 = c + H2 + ADDL (2*4)(DI), CX + MOVL CX, (2*4)(DI) + MOVL CX, (2*4)(BP) + MOVL (3*4)(BP), DX // H3 = d + H3 + ADDL (3*4)(DI), DX + MOVL DX, (3*4)(DI) + MOVL DX, (3*4)(BP) + MOVL (4*4)(BP), AX // H4 = e + H4 + ADDL (4*4)(DI), AX + MOVL AX, (4*4)(DI) + MOVL AX, (4*4)(BP) + MOVL (5*4)(BP), BX // H5 = f + H5 + ADDL (5*4)(DI), BX + MOVL BX, (5*4)(DI) + MOVL BX, (5*4)(BP) + MOVL (6*4)(BP), CX // H6 = g + H6 + ADDL (6*4)(DI), CX + MOVL CX, (6*4)(DI) + MOVL CX, (6*4)(BP) + MOVL (7*4)(BP), DX // H7 = h + H7 + ADDL (7*4)(DI), DX + MOVL DX, (7*4)(DI) + MOVL DX, (7*4)(BP) + + ADDL $64, SI + CMPL SI, 288(SP) + JB loop + +end: + RET diff --git a/src/crypto/sha256/sha256block_amd64.s b/src/crypto/sha256/sha256block_amd64.s new file mode 100644 index 000000000..868eaed48 --- /dev/null +++ b/src/crypto/sha256/sha256block_amd64.s @@ -0,0 +1,256 @@ +// Copyright 2013 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. + +#include "textflag.h" + +// SHA256 block routine. See sha256block.go for Go equivalent. +// +// The algorithm is detailed in FIPS 180-4: +// +// http://csrc.nist.gov/publications/fips/fips180-4/fips-180-4.pdf +// +// Wt = Mt; for 0 <= t <= 15 +// Wt = SIGMA1(Wt-2) + SIGMA0(Wt-15) + Wt-16; for 16 <= t <= 63 +// +// a = H0 +// b = H1 +// c = H2 +// d = H3 +// e = H4 +// f = H5 +// g = H6 +// h = H7 +// +// for t = 0 to 63 { +// T1 = h + BIGSIGMA1(e) + Ch(e,f,g) + Kt + Wt +// T2 = BIGSIGMA0(a) + Maj(a,b,c) +// h = g +// g = f +// f = e +// e = d + T1 +// d = c +// c = b +// b = a +// a = T1 + T2 +// } +// +// H0 = a + H0 +// H1 = b + H1 +// H2 = c + H2 +// H3 = d + H3 +// H4 = e + H4 +// H5 = f + H5 +// H6 = g + H6 +// H7 = h + H7 + +// Wt = Mt; for 0 <= t <= 15 +#define MSGSCHEDULE0(index) \ + MOVL (index*4)(SI), AX; \ + BSWAPL AX; \ + MOVL AX, (index*4)(BP) + +// Wt = SIGMA1(Wt-2) + Wt-7 + SIGMA0(Wt-15) + Wt-16; for 16 <= t <= 63 +// SIGMA0(x) = ROTR(7,x) XOR ROTR(18,x) XOR SHR(3,x) +// SIGMA1(x) = ROTR(17,x) XOR ROTR(19,x) XOR SHR(10,x) +#define MSGSCHEDULE1(index) \ + MOVL ((index-2)*4)(BP), AX; \ + MOVL AX, CX; \ + RORL $17, AX; \ + MOVL CX, DX; \ + RORL $19, CX; \ + SHRL $10, DX; \ + MOVL ((index-15)*4)(BP), BX; \ + XORL CX, AX; \ + MOVL BX, CX; \ + XORL DX, AX; \ + RORL $7, BX; \ + MOVL CX, DX; \ + SHRL $3, DX; \ + RORL $18, CX; \ + ADDL ((index-7)*4)(BP), AX; \ + XORL CX, BX; \ + XORL DX, BX; \ + ADDL ((index-16)*4)(BP), BX; \ + ADDL BX, AX; \ + MOVL AX, ((index)*4)(BP) + +// Calculate T1 in AX - uses AX, CX and DX registers. +// h is also used as an accumulator. Wt is passed in AX. +// T1 = h + BIGSIGMA1(e) + Ch(e, f, g) + Kt + Wt +// BIGSIGMA1(x) = ROTR(6,x) XOR ROTR(11,x) XOR ROTR(25,x) +// Ch(x, y, z) = (x AND y) XOR (NOT x AND z) +#define SHA256T1(const, e, f, g, h) \ + ADDL AX, h; \ + MOVL e, AX; \ + ADDL $const, h; \ + MOVL e, CX; \ + RORL $6, AX; \ + MOVL e, DX; \ + RORL $11, CX; \ + XORL CX, AX; \ + MOVL e, CX; \ + RORL $25, DX; \ + ANDL f, CX; \ + XORL AX, DX; \ + MOVL e, AX; \ + NOTL AX; \ + ADDL DX, h; \ + ANDL g, AX; \ + XORL CX, AX; \ + ADDL h, AX + +// Calculate T2 in BX - uses BX, CX, DX and DI registers. +// T2 = BIGSIGMA0(a) + Maj(a, b, c) +// BIGSIGMA0(x) = ROTR(2,x) XOR ROTR(13,x) XOR ROTR(22,x) +// Maj(x, y, z) = (x AND y) XOR (x AND z) XOR (y AND z) +#define SHA256T2(a, b, c) \ + MOVL a, DI; \ + MOVL c, BX; \ + RORL $2, DI; \ + MOVL a, DX; \ + ANDL b, BX; \ + RORL $13, DX; \ + MOVL a, CX; \ + ANDL c, CX; \ + XORL DX, DI; \ + XORL CX, BX; \ + MOVL a, DX; \ + MOVL b, CX; \ + RORL $22, DX; \ + ANDL a, CX; \ + XORL CX, BX; \ + XORL DX, DI; \ + ADDL DI, BX + +// Calculate T1 and T2, then e = d + T1 and a = T1 + T2. +// The values for e and a are stored in d and h, ready for rotation. +#define SHA256ROUND(index, const, a, b, c, d, e, f, g, h) \ + SHA256T1(const, e, f, g, h); \ + SHA256T2(a, b, c); \ + MOVL BX, h; \ + ADDL AX, d; \ + ADDL AX, h + +#define SHA256ROUND0(index, const, a, b, c, d, e, f, g, h) \ + MSGSCHEDULE0(index); \ + SHA256ROUND(index, const, a, b, c, d, e, f, g, h) + +#define SHA256ROUND1(index, const, a, b, c, d, e, f, g, h) \ + MSGSCHEDULE1(index); \ + SHA256ROUND(index, const, a, b, c, d, e, f, g, h) + +TEXT ·block(SB),0,$264-32 + MOVQ p_base+8(FP), SI + MOVQ p_len+16(FP), DX + SHRQ $6, DX + SHLQ $6, DX + + LEAQ (SI)(DX*1), DI + MOVQ DI, 256(SP) + CMPQ SI, DI + JEQ end + + MOVQ dig+0(FP), BP + MOVL (0*4)(BP), R8 // a = H0 + MOVL (1*4)(BP), R9 // b = H1 + MOVL (2*4)(BP), R10 // c = H2 + MOVL (3*4)(BP), R11 // d = H3 + MOVL (4*4)(BP), R12 // e = H4 + MOVL (5*4)(BP), R13 // f = H5 + MOVL (6*4)(BP), R14 // g = H6 + MOVL (7*4)(BP), R15 // h = H7 + +loop: + MOVQ SP, BP // message schedule + + SHA256ROUND0(0, 0x428a2f98, R8, R9, R10, R11, R12, R13, R14, R15) + SHA256ROUND0(1, 0x71374491, R15, R8, R9, R10, R11, R12, R13, R14) + SHA256ROUND0(2, 0xb5c0fbcf, R14, R15, R8, R9, R10, R11, R12, R13) + SHA256ROUND0(3, 0xe9b5dba5, R13, R14, R15, R8, R9, R10, R11, R12) + SHA256ROUND0(4, 0x3956c25b, R12, R13, R14, R15, R8, R9, R10, R11) + SHA256ROUND0(5, 0x59f111f1, R11, R12, R13, R14, R15, R8, R9, R10) + SHA256ROUND0(6, 0x923f82a4, R10, R11, R12, R13, R14, R15, R8, R9) + SHA256ROUND0(7, 0xab1c5ed5, R9, R10, R11, R12, R13, R14, R15, R8) + SHA256ROUND0(8, 0xd807aa98, R8, R9, R10, R11, R12, R13, R14, R15) + SHA256ROUND0(9, 0x12835b01, R15, R8, R9, R10, R11, R12, R13, R14) + SHA256ROUND0(10, 0x243185be, R14, R15, R8, R9, R10, R11, R12, R13) + SHA256ROUND0(11, 0x550c7dc3, R13, R14, R15, R8, R9, R10, R11, R12) + SHA256ROUND0(12, 0x72be5d74, R12, R13, R14, R15, R8, R9, R10, R11) + SHA256ROUND0(13, 0x80deb1fe, R11, R12, R13, R14, R15, R8, R9, R10) + SHA256ROUND0(14, 0x9bdc06a7, R10, R11, R12, R13, R14, R15, R8, R9) + SHA256ROUND0(15, 0xc19bf174, R9, R10, R11, R12, R13, R14, R15, R8) + + SHA256ROUND1(16, 0xe49b69c1, R8, R9, R10, R11, R12, R13, R14, R15) + SHA256ROUND1(17, 0xefbe4786, R15, R8, R9, R10, R11, R12, R13, R14) + SHA256ROUND1(18, 0x0fc19dc6, R14, R15, R8, R9, R10, R11, R12, R13) + SHA256ROUND1(19, 0x240ca1cc, R13, R14, R15, R8, R9, R10, R11, R12) + SHA256ROUND1(20, 0x2de92c6f, R12, R13, R14, R15, R8, R9, R10, R11) + SHA256ROUND1(21, 0x4a7484aa, R11, R12, R13, R14, R15, R8, R9, R10) + SHA256ROUND1(22, 0x5cb0a9dc, R10, R11, R12, R13, R14, R15, R8, R9) + SHA256ROUND1(23, 0x76f988da, R9, R10, R11, R12, R13, R14, R15, R8) + SHA256ROUND1(24, 0x983e5152, R8, R9, R10, R11, R12, R13, R14, R15) + SHA256ROUND1(25, 0xa831c66d, R15, R8, R9, R10, R11, R12, R13, R14) + SHA256ROUND1(26, 0xb00327c8, R14, R15, R8, R9, R10, R11, R12, R13) + SHA256ROUND1(27, 0xbf597fc7, R13, R14, R15, R8, R9, R10, R11, R12) + SHA256ROUND1(28, 0xc6e00bf3, R12, R13, R14, R15, R8, R9, R10, R11) + SHA256ROUND1(29, 0xd5a79147, R11, R12, R13, R14, R15, R8, R9, R10) + SHA256ROUND1(30, 0x06ca6351, R10, R11, R12, R13, R14, R15, R8, R9) + SHA256ROUND1(31, 0x14292967, R9, R10, R11, R12, R13, R14, R15, R8) + SHA256ROUND1(32, 0x27b70a85, R8, R9, R10, R11, R12, R13, R14, R15) + SHA256ROUND1(33, 0x2e1b2138, R15, R8, R9, R10, R11, R12, R13, R14) + SHA256ROUND1(34, 0x4d2c6dfc, R14, R15, R8, R9, R10, R11, R12, R13) + SHA256ROUND1(35, 0x53380d13, R13, R14, R15, R8, R9, R10, R11, R12) + SHA256ROUND1(36, 0x650a7354, R12, R13, R14, R15, R8, R9, R10, R11) + SHA256ROUND1(37, 0x766a0abb, R11, R12, R13, R14, R15, R8, R9, R10) + SHA256ROUND1(38, 0x81c2c92e, R10, R11, R12, R13, R14, R15, R8, R9) + SHA256ROUND1(39, 0x92722c85, R9, R10, R11, R12, R13, R14, R15, R8) + SHA256ROUND1(40, 0xa2bfe8a1, R8, R9, R10, R11, R12, R13, R14, R15) + SHA256ROUND1(41, 0xa81a664b, R15, R8, R9, R10, R11, R12, R13, R14) + SHA256ROUND1(42, 0xc24b8b70, R14, R15, R8, R9, R10, R11, R12, R13) + SHA256ROUND1(43, 0xc76c51a3, R13, R14, R15, R8, R9, R10, R11, R12) + SHA256ROUND1(44, 0xd192e819, R12, R13, R14, R15, R8, R9, R10, R11) + SHA256ROUND1(45, 0xd6990624, R11, R12, R13, R14, R15, R8, R9, R10) + SHA256ROUND1(46, 0xf40e3585, R10, R11, R12, R13, R14, R15, R8, R9) + SHA256ROUND1(47, 0x106aa070, R9, R10, R11, R12, R13, R14, R15, R8) + SHA256ROUND1(48, 0x19a4c116, R8, R9, R10, R11, R12, R13, R14, R15) + SHA256ROUND1(49, 0x1e376c08, R15, R8, R9, R10, R11, R12, R13, R14) + SHA256ROUND1(50, 0x2748774c, R14, R15, R8, R9, R10, R11, R12, R13) + SHA256ROUND1(51, 0x34b0bcb5, R13, R14, R15, R8, R9, R10, R11, R12) + SHA256ROUND1(52, 0x391c0cb3, R12, R13, R14, R15, R8, R9, R10, R11) + SHA256ROUND1(53, 0x4ed8aa4a, R11, R12, R13, R14, R15, R8, R9, R10) + SHA256ROUND1(54, 0x5b9cca4f, R10, R11, R12, R13, R14, R15, R8, R9) + SHA256ROUND1(55, 0x682e6ff3, R9, R10, R11, R12, R13, R14, R15, R8) + SHA256ROUND1(56, 0x748f82ee, R8, R9, R10, R11, R12, R13, R14, R15) + SHA256ROUND1(57, 0x78a5636f, R15, R8, R9, R10, R11, R12, R13, R14) + SHA256ROUND1(58, 0x84c87814, R14, R15, R8, R9, R10, R11, R12, R13) + SHA256ROUND1(59, 0x8cc70208, R13, R14, R15, R8, R9, R10, R11, R12) + SHA256ROUND1(60, 0x90befffa, R12, R13, R14, R15, R8, R9, R10, R11) + SHA256ROUND1(61, 0xa4506ceb, R11, R12, R13, R14, R15, R8, R9, R10) + SHA256ROUND1(62, 0xbef9a3f7, R10, R11, R12, R13, R14, R15, R8, R9) + SHA256ROUND1(63, 0xc67178f2, R9, R10, R11, R12, R13, R14, R15, R8) + + MOVQ dig+0(FP), BP + ADDL (0*4)(BP), R8 // H0 = a + H0 + MOVL R8, (0*4)(BP) + ADDL (1*4)(BP), R9 // H1 = b + H1 + MOVL R9, (1*4)(BP) + ADDL (2*4)(BP), R10 // H2 = c + H2 + MOVL R10, (2*4)(BP) + ADDL (3*4)(BP), R11 // H3 = d + H3 + MOVL R11, (3*4)(BP) + ADDL (4*4)(BP), R12 // H4 = e + H4 + MOVL R12, (4*4)(BP) + ADDL (5*4)(BP), R13 // H5 = f + H5 + MOVL R13, (5*4)(BP) + ADDL (6*4)(BP), R14 // H6 = g + H6 + MOVL R14, (6*4)(BP) + ADDL (7*4)(BP), R15 // H7 = h + H7 + MOVL R15, (7*4)(BP) + + ADDQ $64, SI + CMPQ SI, 256(SP) + JB loop + +end: + RET diff --git a/src/crypto/sha256/sha256block_decl.go b/src/crypto/sha256/sha256block_decl.go new file mode 100644 index 000000000..a50c97871 --- /dev/null +++ b/src/crypto/sha256/sha256block_decl.go @@ -0,0 +1,11 @@ +// Copyright 2013 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 386 amd64 + +package sha256 + +//go:noescape + +func block(dig *digest, p []byte) |