diff --git a/.github/workflows/test_test.yaml b/.github/workflows/test_test.yaml index 9e5dce1..466384c 100644 --- a/.github/workflows/test_test.yaml +++ b/.github/workflows/test_test.yaml @@ -12,7 +12,7 @@ jobs: - name: Set up Go uses: actions/setup-go@v2 with: - go-version: 1.21.3 + go-version: 1.21.5 - name: Install dependencies run: go mod download diff --git a/go.mod b/go.mod index b8d7771..3b90b89 100644 --- a/go.mod +++ b/go.mod @@ -1,14 +1,14 @@ module github.com/D3vl0per/crypt -go 1.21.3 +go 1.21.7 require ( filippo.io/age v1.1.1 - github.com/andybalholm/brotli v1.0.6 + github.com/andybalholm/brotli v1.1.0 github.com/cloudflare/circl v1.3.7 - github.com/klauspost/compress v1.17.4 + github.com/klauspost/compress v1.17.6 github.com/stretchr/testify v1.8.4 - golang.org/x/crypto v0.17.0 + golang.org/x/crypto v0.18.0 ) require ( diff --git a/go.sum b/go.sum index 813a421..d579e38 100644 --- a/go.sum +++ b/go.sum @@ -1,27 +1,19 @@ filippo.io/age v1.1.1 h1:pIpO7l151hCnQ4BdyBujnGP2YlUo0uj6sAVNHGBvXHg= filippo.io/age v1.1.1/go.mod h1:l03SrzDUrBkdBx8+IILdnn2KZysqQdbEBUQ4p3sqEQE= -github.com/andybalholm/brotli v1.0.6 h1:Yf9fFpf49Zrxb9NlQaluyE92/+X7UVHlhMNJN2sxfOI= -github.com/andybalholm/brotli v1.0.6/go.mod h1:fO7iG3H7G2nSZ7m0zPUDn85XEX2GTukHGRSepvi9Eig= -github.com/cloudflare/circl v1.3.6 h1:/xbKIqSHbZXHwkhbrhrt2YOHIwYJlXH94E3tI/gDlUg= -github.com/cloudflare/circl v1.3.6/go.mod h1:5XYMA4rFBvNIrhs50XuiBJ15vF2pZn4nnUKZrLbUZFA= +github.com/andybalholm/brotli v1.1.0 h1:eLKJA0d02Lf0mVpIDgYnqXcUn0GqVmEFny3VuID1U3M= +github.com/andybalholm/brotli v1.1.0/go.mod h1:sms7XGricyQI9K10gOSf56VKKWS4oLer58Q+mhRPtnY= github.com/cloudflare/circl v1.3.7 h1:qlCDlTPz2n9fu58M0Nh1J/JzcFpfgkFHHX3O35r5vcU= github.com/cloudflare/circl v1.3.7/go.mod h1:sRTcRWXGLrKw6yIGJ+l7amYJFfAXbZG0kBSc8r4zxgA= github.com/davecgh/go-spew v1.1.1 h1:vj9j/u1bqnvCEfJOwUhtlOARqs3+rkHYY13jYWTU97c= github.com/davecgh/go-spew v1.1.1/go.mod h1:J7Y8YcW2NihsgmVo/mv3lAwl/skON4iLHjSsI+c5H38= -github.com/klauspost/compress v1.17.3 h1:qkRjuerhUU1EmXLYGkSH6EZL+vPSxIrYjLNAK4slzwA= -github.com/klauspost/compress v1.17.3/go.mod h1:/dCuZOvVtNoHsyb+cuJD3itjs3NbnF6KH9zAO4BDxPM= -github.com/klauspost/compress v1.17.4 h1:Ej5ixsIri7BrIjBkRZLTo6ghwrEtHFk7ijlczPW4fZ4= -github.com/klauspost/compress v1.17.4/go.mod h1:/dCuZOvVtNoHsyb+cuJD3itjs3NbnF6KH9zAO4BDxPM= +github.com/klauspost/compress v1.17.6 h1:60eq2E/jlfwQXtvZEeBUYADs+BwKBWURIY+Gj2eRGjI= +github.com/klauspost/compress v1.17.6/go.mod h1:/dCuZOvVtNoHsyb+cuJD3itjs3NbnF6KH9zAO4BDxPM= github.com/pmezard/go-difflib v1.0.0 h1:4DBwDE0NGyQoBHbLQYPwSUPoCMWR5BEzIk/f1lZbAQM= github.com/pmezard/go-difflib v1.0.0/go.mod h1:iKH77koFhYxTK1pcRnkKkqfTogsbg7gZNVY4sRDYZ/4= github.com/stretchr/testify v1.8.4 h1:CcVxjf3Q8PM0mHUKJCdn+eZZtm5yQwehR5yeSVQQcUk= github.com/stretchr/testify v1.8.4/go.mod h1:sz/lmYIOXD/1dqDmKjjqLyZ2RngseejIcXlSw2iwfAo= -golang.org/x/crypto v0.15.0 h1:frVn1TEaCEaZcn3Tmd7Y2b5KKPaZ+I32Q2OA3kYp5TA= -golang.org/x/crypto v0.15.0/go.mod h1:4ChreQoLWfG3xLDer1WdlH5NdlQ3+mwnQq1YTKY+72g= -golang.org/x/crypto v0.17.0 h1:r8bRNjWL3GshPW3gkd+RpvzWrZAwPS49OmTGZ/uhM4k= -golang.org/x/crypto v0.17.0/go.mod h1:gCAAfMLgwOJRpTjQ2zCCt2OcSfYMTeZVSRtQlPC7Nq4= -golang.org/x/sys v0.14.0 h1:Vz7Qs629MkJkGyHxUlRHizWJRG2j8fbQKjELVSNhy7Q= -golang.org/x/sys v0.14.0/go.mod h1:/VUhepiaJMQUp4+oa/7Zr1D23ma6VTLIYjOOTFZPUcA= +golang.org/x/crypto v0.18.0 h1:PGVlW0xEltQnzFZ55hkuX5+KLyrMYhHld1YHO4AKcdc= +golang.org/x/crypto v0.18.0/go.mod h1:R0j02AL6hcrfOiy9T4ZYp/rcWeMxM3L6QYxlOuEG1mg= golang.org/x/sys v0.16.0 h1:xWw16ngr6ZMtmxDyKyIgsE93KNKz5HKmMa3b8ALHidU= golang.org/x/sys v0.16.0/go.mod h1:/VUhepiaJMQUp4+oa/7Zr1D23ma6VTLIYjOOTFZPUcA= gopkg.in/check.v1 v0.0.0-20161208181325-20d25e280405 h1:yhCVgyC4o1eVCa2tZl7eS0r+SDo693bJlVdllGtEeKM= diff --git a/vendor/github.com/andybalholm/brotli/README.md b/vendor/github.com/andybalholm/brotli/README.md index 1ea7fdb..0062521 100644 --- a/vendor/github.com/andybalholm/brotli/README.md +++ b/vendor/github.com/andybalholm/brotli/README.md @@ -2,6 +2,13 @@ This package is a brotli compressor and decompressor implemented in Go. It was translated from the reference implementation (https://github.com/google/brotli) with the `c2go` tool at https://github.com/andybalholm/c2go. +I have been working on new compression algorithms (not translated from C) +in the matchfinder package. +You can use them with the NewWriterV2 function. +Currently they give better results than the old implementation +(at least for compressing my test file, Newton’s *Opticks*) +on levels 2 to 6. + I am using it in production with https://github.com/andybalholm/redwood. API documentation is found at https://pkg.go.dev/github.com/andybalholm/brotli?tab=doc. diff --git a/vendor/github.com/andybalholm/brotli/bitwriter.go b/vendor/github.com/andybalholm/brotli/bitwriter.go new file mode 100644 index 0000000..dfc6036 --- /dev/null +++ b/vendor/github.com/andybalholm/brotli/bitwriter.go @@ -0,0 +1,56 @@ +package brotli + +/* Copyright 2010 Google Inc. All Rights Reserved. + + Distributed under MIT license. + See file LICENSE for detail or copy at https://opensource.org/licenses/MIT +*/ + +/* Write bits into a byte array. */ + +type bitWriter struct { + dst []byte + + // Data waiting to be written is the low nbits of bits. + bits uint64 + nbits uint +} + +func (w *bitWriter) writeBits(nb uint, b uint64) { + w.bits |= b << w.nbits + w.nbits += nb + if w.nbits >= 32 { + bits := w.bits + w.bits >>= 32 + w.nbits -= 32 + w.dst = append(w.dst, + byte(bits), + byte(bits>>8), + byte(bits>>16), + byte(bits>>24), + ) + } +} + +func (w *bitWriter) writeSingleBit(bit bool) { + if bit { + w.writeBits(1, 1) + } else { + w.writeBits(1, 0) + } +} + +func (w *bitWriter) jumpToByteBoundary() { + dst := w.dst + for w.nbits != 0 { + dst = append(dst, byte(w.bits)) + w.bits >>= 8 + if w.nbits > 8 { // Avoid underflow + w.nbits -= 8 + } else { + w.nbits = 0 + } + } + w.bits = 0 + w.dst = dst +} diff --git a/vendor/github.com/andybalholm/brotli/brotli_bit_stream.go b/vendor/github.com/andybalholm/brotli/brotli_bit_stream.go index 7acfb18..ee65529 100644 --- a/vendor/github.com/andybalholm/brotli/brotli_bit_stream.go +++ b/vendor/github.com/andybalholm/brotli/brotli_bit_stream.go @@ -7,12 +7,18 @@ import ( const maxHuffmanTreeSize = (2*numCommandSymbols + 1) -/* The maximum size of Huffman dictionary for distances assuming that - NPOSTFIX = 0 and NDIRECT = 0. */ +/* +The maximum size of Huffman dictionary for distances assuming that + + NPOSTFIX = 0 and NDIRECT = 0. +*/ const maxSimpleDistanceAlphabetSize = 140 -/* Represents the range of values belonging to a prefix code: - [offset, offset + 2^nbits) */ +/* +Represents the range of values belonging to a prefix code: + + [offset, offset + 2^nbits) +*/ type prefixCodeRange struct { offset uint32 nbits uint32 @@ -96,9 +102,12 @@ func nextBlockTypeCode(calculator *blockTypeCodeCalculator, type_ byte) uint { return type_code } -/* |nibblesbits| represents the 2 bits to encode MNIBBLES (0-3) - REQUIRES: length > 0 - REQUIRES: length <= (1 << 24) */ +/* +|nibblesbits| represents the 2 bits to encode MNIBBLES (0-3) + + REQUIRES: length > 0 + REQUIRES: length <= (1 << 24) +*/ func encodeMlen(length uint, bits *uint64, numbits *uint, nibblesbits *uint64) { var lg uint if length == 1 { @@ -132,8 +141,11 @@ func storeCommandExtra(cmd *command, storage_ix *uint, storage []byte) { writeBits(uint(insnumextra+getCopyExtra(copycode)), bits, storage_ix, storage) } -/* Data structure that stores almost everything that is needed to encode each - block switch command. */ +/* +Data structure that stores almost everything that is needed to encode each + + block switch command. +*/ type blockSplitCode struct { type_code_calculator blockTypeCodeCalculator type_depths [maxBlockTypeSymbols]byte @@ -154,9 +166,12 @@ func storeVarLenUint8(n uint, storage_ix *uint, storage []byte) { } } -/* Stores the compressed meta-block header. - REQUIRES: length > 0 - REQUIRES: length <= (1 << 24) */ +/* +Stores the compressed meta-block header. + + REQUIRES: length > 0 + REQUIRES: length <= (1 << 24) +*/ func storeCompressedMetaBlockHeader(is_final_block bool, length uint, storage_ix *uint, storage []byte) { var lenbits uint64 var nlenbits uint @@ -186,9 +201,12 @@ func storeCompressedMetaBlockHeader(is_final_block bool, length uint, storage_ix } } -/* Stores the uncompressed meta-block header. - REQUIRES: length > 0 - REQUIRES: length <= (1 << 24) */ +/* +Stores the uncompressed meta-block header. + + REQUIRES: length > 0 + REQUIRES: length <= (1 << 24) +*/ func storeUncompressedMetaBlockHeader(length uint, storage_ix *uint, storage []byte) { var lenbits uint64 var nlenbits uint @@ -312,8 +330,11 @@ func storeSimpleHuffmanTree(depths []byte, symbols []uint, num_symbols uint, max } } -/* num = alphabet size - depths = symbol depths */ +/* +num = alphabet size + + depths = symbol depths +*/ func storeHuffmanTree(depths []byte, num uint, tree []huffmanTree, storage_ix *uint, storage []byte) { var huffman_tree [numCommandSymbols]byte var huffman_tree_extra_bits [numCommandSymbols]byte @@ -367,8 +388,11 @@ func storeHuffmanTree(depths []byte, num uint, tree []huffmanTree, storage_ix *u storeHuffmanTreeToBitMask(huffman_tree_size, huffman_tree[:], huffman_tree_extra_bits[:], code_length_bitdepth[:], code_length_bitdepth_symbols[:], storage_ix, storage) } -/* Builds a Huffman tree from histogram[0:length] into depth[0:length] and - bits[0:length] and stores the encoded tree to the bit stream. */ +/* +Builds a Huffman tree from histogram[0:length] into depth[0:length] and + + bits[0:length] and stores the encoded tree to the bit stream. +*/ func buildAndStoreHuffmanTree(histogram []uint32, histogram_length uint, alphabet_size uint, tree []huffmanTree, depth []byte, bits []uint16, storage_ix *uint, storage []byte) { var count uint = 0 var s4 = [4]uint{0} @@ -623,6 +647,203 @@ func buildAndStoreHuffmanTreeFast(histogram []uint32, histogram_total uint, max_ } } +func buildAndStoreHuffmanTreeFastBW(histogram []uint32, histogram_total uint, max_bits uint, depth []byte, bits []uint16, bw *bitWriter) { + var count uint = 0 + var symbols = [4]uint{0} + var length uint = 0 + var total uint = histogram_total + for total != 0 { + if histogram[length] != 0 { + if count < 4 { + symbols[count] = length + } + + count++ + total -= uint(histogram[length]) + } + + length++ + } + + if count <= 1 { + bw.writeBits(4, 1) + bw.writeBits(max_bits, uint64(symbols[0])) + depth[symbols[0]] = 0 + bits[symbols[0]] = 0 + return + } + + for i := 0; i < int(length); i++ { + depth[i] = 0 + } + { + var max_tree_size uint = 2*length + 1 + tree, _ := huffmanTreePool.Get().(*[]huffmanTree) + if tree == nil || cap(*tree) < int(max_tree_size) { + tmp := make([]huffmanTree, max_tree_size) + tree = &tmp + } else { + *tree = (*tree)[:max_tree_size] + } + var count_limit uint32 + for count_limit = 1; ; count_limit *= 2 { + var node int = 0 + var l uint + for l = length; l != 0; { + l-- + if histogram[l] != 0 { + if histogram[l] >= count_limit { + initHuffmanTree(&(*tree)[node:][0], histogram[l], -1, int16(l)) + } else { + initHuffmanTree(&(*tree)[node:][0], count_limit, -1, int16(l)) + } + + node++ + } + } + { + var n int = node + /* Points to the next leaf node. */ /* Points to the next non-leaf node. */ + var sentinel huffmanTree + var i int = 0 + var j int = n + 1 + var k int + + sortHuffmanTreeItems(*tree, uint(n), huffmanTreeComparator(sortHuffmanTree1)) + + /* The nodes are: + [0, n): the sorted leaf nodes that we start with. + [n]: we add a sentinel here. + [n + 1, 2n): new parent nodes are added here, starting from + (n+1). These are naturally in ascending order. + [2n]: we add a sentinel at the end as well. + There will be (2n+1) elements at the end. */ + initHuffmanTree(&sentinel, math.MaxUint32, -1, -1) + + (*tree)[node] = sentinel + node++ + (*tree)[node] = sentinel + node++ + + for k = n - 1; k > 0; k-- { + var left int + var right int + if (*tree)[i].total_count_ <= (*tree)[j].total_count_ { + left = i + i++ + } else { + left = j + j++ + } + + if (*tree)[i].total_count_ <= (*tree)[j].total_count_ { + right = i + i++ + } else { + right = j + j++ + } + + /* The sentinel node becomes the parent node. */ + (*tree)[node-1].total_count_ = (*tree)[left].total_count_ + (*tree)[right].total_count_ + + (*tree)[node-1].index_left_ = int16(left) + (*tree)[node-1].index_right_or_value_ = int16(right) + + /* Add back the last sentinel node. */ + (*tree)[node] = sentinel + node++ + } + + if setDepth(2*n-1, *tree, depth, 14) { + /* We need to pack the Huffman tree in 14 bits. If this was not + successful, add fake entities to the lowest values and retry. */ + break + } + } + } + + huffmanTreePool.Put(tree) + } + + convertBitDepthsToSymbols(depth, length, bits) + if count <= 4 { + var i uint + + /* value of 1 indicates a simple Huffman code */ + bw.writeBits(2, 1) + + bw.writeBits(2, uint64(count)-1) /* NSYM - 1 */ + + /* Sort */ + for i = 0; i < count; i++ { + var j uint + for j = i + 1; j < count; j++ { + if depth[symbols[j]] < depth[symbols[i]] { + var tmp uint = symbols[j] + symbols[j] = symbols[i] + symbols[i] = tmp + } + } + } + + if count == 2 { + bw.writeBits(max_bits, uint64(symbols[0])) + bw.writeBits(max_bits, uint64(symbols[1])) + } else if count == 3 { + bw.writeBits(max_bits, uint64(symbols[0])) + bw.writeBits(max_bits, uint64(symbols[1])) + bw.writeBits(max_bits, uint64(symbols[2])) + } else { + bw.writeBits(max_bits, uint64(symbols[0])) + bw.writeBits(max_bits, uint64(symbols[1])) + bw.writeBits(max_bits, uint64(symbols[2])) + bw.writeBits(max_bits, uint64(symbols[3])) + + /* tree-select */ + bw.writeSingleBit(depth[symbols[0]] == 1) + } + } else { + var previous_value byte = 8 + var i uint + + /* Complex Huffman Tree */ + storeStaticCodeLengthCodeBW(bw) + + /* Actual RLE coding. */ + for i = 0; i < length; { + var value byte = depth[i] + var reps uint = 1 + var k uint + for k = i + 1; k < length && depth[k] == value; k++ { + reps++ + } + + i += reps + if value == 0 { + bw.writeBits(uint(kZeroRepsDepth[reps]), kZeroRepsBits[reps]) + } else { + if previous_value != value { + bw.writeBits(uint(kCodeLengthDepth[value]), uint64(kCodeLengthBits[value])) + reps-- + } + + if reps < 3 { + for reps != 0 { + reps-- + bw.writeBits(uint(kCodeLengthDepth[value]), uint64(kCodeLengthBits[value])) + } + } else { + reps -= 3 + bw.writeBits(uint(kNonZeroRepsDepth[reps]), kNonZeroRepsBits[reps]) + } + + previous_value = value + } + } + } +} + func indexOf(v []byte, v_size uint, value byte) uint { var i uint = 0 for ; i < v_size; i++ { @@ -674,12 +895,15 @@ func moveToFrontTransform(v_in []uint32, v_size uint, v_out []uint32) { } } -/* Finds runs of zeros in v[0..in_size) and replaces them with a prefix code of - the run length plus extra bits (lower 9 bits is the prefix code and the rest - are the extra bits). Non-zero values in v[] are shifted by - *max_length_prefix. Will not create prefix codes bigger than the initial - value of *max_run_length_prefix. The prefix code of run length L is simply - Log2Floor(L) and the number of extra bits is the same as the prefix code. */ +/* +Finds runs of zeros in v[0..in_size) and replaces them with a prefix code of + + the run length plus extra bits (lower 9 bits is the prefix code and the rest + are the extra bits). Non-zero values in v[] are shifted by + *max_length_prefix. Will not create prefix codes bigger than the initial + value of *max_run_length_prefix. The prefix code of run length L is simply + Log2Floor(L) and the number of extra bits is the same as the prefix code. +*/ func runLengthCodeZeros(in_size uint, v []uint32, out_size *uint, max_run_length_prefix *uint32) { var max_reps uint32 = 0 var i uint @@ -799,8 +1023,11 @@ func storeBlockSwitch(code *blockSplitCode, block_len uint32, block_type byte, i writeBits(uint(len_nextra), uint64(len_extra), storage_ix, storage) } -/* Builds a BlockSplitCode data structure from the block split given by the - vector of block types and block lengths and stores it to the bit stream. */ +/* +Builds a BlockSplitCode data structure from the block split given by the + + vector of block types and block lengths and stores it to the bit stream. +*/ func buildAndStoreBlockSplitCode(types []byte, lengths []uint32, num_blocks uint, num_types uint, tree []huffmanTree, code *blockSplitCode, storage_ix *uint, storage []byte) { var type_histo [maxBlockTypeSymbols]uint32 var length_histo [numBlockLenSymbols]uint32 @@ -919,14 +1146,20 @@ func cleanupBlockEncoder(self *blockEncoder) { blockEncoderPool.Put(self) } -/* Creates entropy codes of block lengths and block types and stores them - to the bit stream. */ +/* +Creates entropy codes of block lengths and block types and stores them + + to the bit stream. +*/ func buildAndStoreBlockSwitchEntropyCodes(self *blockEncoder, tree []huffmanTree, storage_ix *uint, storage []byte) { buildAndStoreBlockSplitCode(self.block_types_, self.block_lengths_, self.num_blocks_, self.num_block_types_, tree, &self.block_split_code_, storage_ix, storage) } -/* Stores the next symbol with the entropy code of the current block type. - Updates the block type and block length at block boundaries. */ +/* +Stores the next symbol with the entropy code of the current block type. + + Updates the block type and block length at block boundaries. +*/ func storeSymbol(self *blockEncoder, symbol uint, storage_ix *uint, storage []byte) { if self.block_len_ == 0 { self.block_ix_++ @@ -945,9 +1178,12 @@ func storeSymbol(self *blockEncoder, symbol uint, storage_ix *uint, storage []by } } -/* Stores the next symbol with the entropy code of the current block type and - context value. - Updates the block type and block length at block boundaries. */ +/* +Stores the next symbol with the entropy code of the current block type and + + context value. + Updates the block type and block length at block boundaries. +*/ func storeSymbolWithContext(self *blockEncoder, symbol uint, context uint, context_map []uint32, storage_ix *uint, storage []byte, context_bits uint) { if self.block_len_ == 0 { self.block_ix_++ @@ -1268,8 +1504,11 @@ func storeMetaBlockFast(input []byte, start_pos uint, length uint, mask uint, is } } -/* This is for storing uncompressed blocks (simple raw storage of - bytes-as-bytes). */ +/* +This is for storing uncompressed blocks (simple raw storage of + + bytes-as-bytes). +*/ func storeUncompressedMetaBlock(is_final_block bool, input []byte, position uint, mask uint, len uint, storage_ix *uint, storage []byte) { var masked_pos uint = position & mask storeUncompressedMetaBlockHeader(uint(len), storage_ix, storage) diff --git a/vendor/github.com/andybalholm/brotli/compress_fragment_two_pass.go b/vendor/github.com/andybalholm/brotli/compress_fragment_two_pass.go index 172dc7f..79f9c7f 100644 --- a/vendor/github.com/andybalholm/brotli/compress_fragment_two_pass.go +++ b/vendor/github.com/andybalholm/brotli/compress_fragment_two_pass.go @@ -39,8 +39,11 @@ func isMatch1(p1 []byte, p2 []byte, length uint) bool { return p1[4] == p2[4] && p1[5] == p2[5] } -/* Builds a command and distance prefix code (each 64 symbols) into "depth" and - "bits" based on "histogram" and stores it into the bit stream. */ +/* +Builds a command and distance prefix code (each 64 symbols) into "depth" and + + "bits" based on "histogram" and stores it into the bit stream. +*/ func buildAndStoreCommandPrefixCode(histogram []uint32, depth []byte, bits []uint16, storage_ix *uint, storage []byte) { var tree [129]huffmanTree var cmd_depth = [numCommandSymbols]byte{0} @@ -216,6 +219,25 @@ func storeMetaBlockHeader(len uint, is_uncompressed bool, storage_ix *uint, stor writeSingleBit(is_uncompressed, storage_ix, storage) } +func storeMetaBlockHeaderBW(len uint, is_uncompressed bool, bw *bitWriter) { + var nibbles uint = 6 + + /* ISLAST */ + bw.writeBits(1, 0) + + if len <= 1<<16 { + nibbles = 4 + } else if len <= 1<<20 { + nibbles = 5 + } + + bw.writeBits(2, uint64(nibbles)-4) + bw.writeBits(nibbles*4, uint64(len)-1) + + /* ISUNCOMPRESSED */ + bw.writeSingleBit(is_uncompressed) +} + func createCommands(input []byte, block_size uint, input_size uint, base_ip_ptr []byte, table []int, table_bits uint, min_match uint, literals *[]byte, commands *[]uint32) { var ip int = 0 var shift uint = 64 - table_bits @@ -710,19 +732,22 @@ func compressFragmentTwoPassImpl(input []byte, input_size uint, is_last bool, co } } -/* Compresses "input" string to the "*storage" buffer as one or more complete - meta-blocks, and updates the "*storage_ix" bit position. +/* +Compresses "input" string to the "*storage" buffer as one or more complete - If "is_last" is 1, emits an additional empty last meta-block. + meta-blocks, and updates the "*storage_ix" bit position. - REQUIRES: "input_size" is greater than zero, or "is_last" is 1. - REQUIRES: "input_size" is less or equal to maximal metablock size (1 << 24). - REQUIRES: "command_buf" and "literal_buf" point to at least - kCompressFragmentTwoPassBlockSize long arrays. - REQUIRES: All elements in "table[0..table_size-1]" are initialized to zero. - REQUIRES: "table_size" is a power of two - OUTPUT: maximal copy distance <= |input_size| - OUTPUT: maximal copy distance <= BROTLI_MAX_BACKWARD_LIMIT(18) */ + If "is_last" is 1, emits an additional empty last meta-block. + + REQUIRES: "input_size" is greater than zero, or "is_last" is 1. + REQUIRES: "input_size" is less or equal to maximal metablock size (1 << 24). + REQUIRES: "command_buf" and "literal_buf" point to at least + kCompressFragmentTwoPassBlockSize long arrays. + REQUIRES: All elements in "table[0..table_size-1]" are initialized to zero. + REQUIRES: "table_size" is a power of two + OUTPUT: maximal copy distance <= |input_size| + OUTPUT: maximal copy distance <= BROTLI_MAX_BACKWARD_LIMIT(18) +*/ func compressFragmentTwoPass(input []byte, input_size uint, is_last bool, command_buf []uint32, literal_buf []byte, table []int, table_size uint, storage_ix *uint, storage []byte) { var initial_storage_ix uint = *storage_ix var table_bits uint = uint(log2FloorNonZero(table_size)) diff --git a/vendor/github.com/andybalholm/brotli/encoder.go b/vendor/github.com/andybalholm/brotli/encoder.go new file mode 100644 index 0000000..650d1e4 --- /dev/null +++ b/vendor/github.com/andybalholm/brotli/encoder.go @@ -0,0 +1,168 @@ +package brotli + +import "github.com/andybalholm/brotli/matchfinder" + +// An Encoder implements the matchfinder.Encoder interface, writing in Brotli format. +type Encoder struct { + wroteHeader bool + bw bitWriter + distCache []distanceCode +} + +func (e *Encoder) Reset() { + e.wroteHeader = false + e.bw = bitWriter{} +} + +func (e *Encoder) Encode(dst []byte, src []byte, matches []matchfinder.Match, lastBlock bool) []byte { + e.bw.dst = dst + if !e.wroteHeader { + e.bw.writeBits(4, 15) + e.wroteHeader = true + } + + var literalHisto [256]uint32 + var commandHisto [704]uint32 + var distanceHisto [64]uint32 + literalCount := 0 + commandCount := 0 + distanceCount := 0 + + if len(e.distCache) < len(matches) { + e.distCache = make([]distanceCode, len(matches)) + } + + // first pass: build the histograms + pos := 0 + + // d is the ring buffer of the last 4 distances. + d := [4]int{-10, -10, -10, -10} + for i, m := range matches { + if m.Unmatched > 0 { + for _, c := range src[pos : pos+m.Unmatched] { + literalHisto[c]++ + } + literalCount += m.Unmatched + } + + insertCode := getInsertLengthCode(uint(m.Unmatched)) + copyCode := getCopyLengthCode(uint(m.Length)) + if m.Length == 0 { + // If the stream ends with unmatched bytes, we need a dummy copy length. + copyCode = 2 + } + command := combineLengthCodes(insertCode, copyCode, false) + commandHisto[command]++ + commandCount++ + + if command >= 128 && m.Length != 0 { + var distCode distanceCode + switch m.Distance { + case d[3]: + distCode.code = 0 + case d[2]: + distCode.code = 1 + case d[1]: + distCode.code = 2 + case d[0]: + distCode.code = 3 + case d[3] - 1: + distCode.code = 4 + case d[3] + 1: + distCode.code = 5 + case d[3] - 2: + distCode.code = 6 + case d[3] + 2: + distCode.code = 7 + case d[3] - 3: + distCode.code = 8 + case d[3] + 3: + distCode.code = 9 + + // In my testing, codes 10–15 actually reduced the compression ratio. + + default: + distCode = getDistanceCode(m.Distance) + } + e.distCache[i] = distCode + distanceHisto[distCode.code]++ + distanceCount++ + if distCode.code != 0 { + d[0], d[1], d[2], d[3] = d[1], d[2], d[3], m.Distance + } + } + + pos += m.Unmatched + m.Length + } + + storeMetaBlockHeaderBW(uint(len(src)), false, &e.bw) + e.bw.writeBits(13, 0) + + var literalDepths [256]byte + var literalBits [256]uint16 + buildAndStoreHuffmanTreeFastBW(literalHisto[:], uint(literalCount), 8, literalDepths[:], literalBits[:], &e.bw) + + var commandDepths [704]byte + var commandBits [704]uint16 + buildAndStoreHuffmanTreeFastBW(commandHisto[:], uint(commandCount), 10, commandDepths[:], commandBits[:], &e.bw) + + var distanceDepths [64]byte + var distanceBits [64]uint16 + buildAndStoreHuffmanTreeFastBW(distanceHisto[:], uint(distanceCount), 6, distanceDepths[:], distanceBits[:], &e.bw) + + pos = 0 + for i, m := range matches { + insertCode := getInsertLengthCode(uint(m.Unmatched)) + copyCode := getCopyLengthCode(uint(m.Length)) + if m.Length == 0 { + // If the stream ends with unmatched bytes, we need a dummy copy length. + copyCode = 2 + } + command := combineLengthCodes(insertCode, copyCode, false) + e.bw.writeBits(uint(commandDepths[command]), uint64(commandBits[command])) + if kInsExtra[insertCode] > 0 { + e.bw.writeBits(uint(kInsExtra[insertCode]), uint64(m.Unmatched)-uint64(kInsBase[insertCode])) + } + if kCopyExtra[copyCode] > 0 { + e.bw.writeBits(uint(kCopyExtra[copyCode]), uint64(m.Length)-uint64(kCopyBase[copyCode])) + } + + if m.Unmatched > 0 { + for _, c := range src[pos : pos+m.Unmatched] { + e.bw.writeBits(uint(literalDepths[c]), uint64(literalBits[c])) + } + } + + if command >= 128 && m.Length != 0 { + distCode := e.distCache[i] + e.bw.writeBits(uint(distanceDepths[distCode.code]), uint64(distanceBits[distCode.code])) + if distCode.nExtra > 0 { + e.bw.writeBits(distCode.nExtra, distCode.extraBits) + } + } + + pos += m.Unmatched + m.Length + } + + if lastBlock { + e.bw.writeBits(2, 3) // islast + isempty + e.bw.jumpToByteBoundary() + } + return e.bw.dst +} + +type distanceCode struct { + code int + nExtra uint + extraBits uint64 +} + +func getDistanceCode(distance int) distanceCode { + d := distance + 3 + nbits := log2FloorNonZero(uint(d)) - 1 + prefix := (d >> nbits) & 1 + offset := (2 + prefix) << nbits + distcode := int(2*(nbits-1)) + prefix + 16 + extra := d - offset + return distanceCode{distcode, uint(nbits), uint64(extra)} +} diff --git a/vendor/github.com/andybalholm/brotli/entropy_encode_static.go b/vendor/github.com/andybalholm/brotli/entropy_encode_static.go index 5ddf3fc..294aff4 100644 --- a/vendor/github.com/andybalholm/brotli/entropy_encode_static.go +++ b/vendor/github.com/andybalholm/brotli/entropy_encode_static.go @@ -782,6 +782,11 @@ func storeStaticCodeLengthCode(storage_ix *uint, storage []byte) { writeBits(40, 0x0000FF55555554, storage_ix, storage) } +func storeStaticCodeLengthCodeBW(bw *bitWriter) { + bw.writeBits(32, 0x55555554) + bw.writeBits(8, 0xFF) +} + var kZeroRepsBits = [numCommandSymbols]uint64{ 0x00000000, 0x00000000, diff --git a/vendor/github.com/andybalholm/brotli/http.go b/vendor/github.com/andybalholm/brotli/http.go index b5b8904..3d3a8a0 100644 --- a/vendor/github.com/andybalholm/brotli/http.go +++ b/vendor/github.com/andybalholm/brotli/http.go @@ -20,7 +20,7 @@ func HTTPCompressor(w http.ResponseWriter, r *http.Request) io.WriteCloser { switch encoding { case "br": w.Header().Set("Content-Encoding", "br") - return NewWriter(w) + return NewWriterV2(w, DefaultCompression) case "gzip": w.Header().Set("Content-Encoding", "gzip") return gzip.NewWriter(w) diff --git a/vendor/github.com/andybalholm/brotli/matchfinder/emitter.go b/vendor/github.com/andybalholm/brotli/matchfinder/emitter.go new file mode 100644 index 0000000..37ed8e1 --- /dev/null +++ b/vendor/github.com/andybalholm/brotli/matchfinder/emitter.go @@ -0,0 +1,45 @@ +package matchfinder + +// An absoluteMatch is like a Match, but it stores indexes into the byte +// stream instead of lengths. +type absoluteMatch struct { + // Start is the index of the first byte. + Start int + + // End is the index of the byte after the last byte + // (so that End - Start = Length). + End int + + // Match is the index of the previous data that matches + // (Start - Match = Distance). + Match int +} + +// A matchEmitter manages the output of matches for a MatchFinder. +type matchEmitter struct { + // Dst is the destination slice that Matches are added to. + Dst []Match + + // NextEmit is the index of the next byte to emit. + NextEmit int +} + +func (e *matchEmitter) emit(m absoluteMatch) { + e.Dst = append(e.Dst, Match{ + Unmatched: m.Start - e.NextEmit, + Length: m.End - m.Start, + Distance: m.Start - m.Match, + }) + e.NextEmit = m.End +} + +// trim shortens m if it extends past maxEnd. Then if the length is at least +// minLength, the match is emitted. +func (e *matchEmitter) trim(m absoluteMatch, maxEnd int, minLength int) { + if m.End > maxEnd { + m.End = maxEnd + } + if m.End-m.Start >= minLength { + e.emit(m) + } +} diff --git a/vendor/github.com/andybalholm/brotli/matchfinder/m0.go b/vendor/github.com/andybalholm/brotli/matchfinder/m0.go new file mode 100644 index 0000000..773b7c4 --- /dev/null +++ b/vendor/github.com/andybalholm/brotli/matchfinder/m0.go @@ -0,0 +1,169 @@ +package matchfinder + +import ( + "encoding/binary" +) + +// M0 is an implementation of the MatchFinder interface based +// on the algorithm used by snappy, but modified to be more like the algorithm +// used by compression level 0 of the brotli reference implementation. +// +// It has a maximum block size of 65536 bytes. +type M0 struct { + // Lazy turns on "lazy matching," for higher compression but less speed. + Lazy bool + + MaxDistance int + MaxLength int +} + +func (M0) Reset() {} + +const ( + m0HashLen = 5 + + m0TableBits = 14 + m0TableSize = 1 << m0TableBits + m0Shift = 32 - m0TableBits + // m0TableMask is redundant, but helps the compiler eliminate bounds + // checks. + m0TableMask = m0TableSize - 1 +) + +func (m M0) hash(data uint64) uint64 { + hash := (data << (64 - 8*m0HashLen)) * hashMul64 + return hash >> (64 - m0TableBits) +} + +// FindMatches looks for matches in src, appends them to dst, and returns dst. +// src must not be longer than 65536 bytes. +func (m M0) FindMatches(dst []Match, src []byte) []Match { + const inputMargin = 16 - 1 + const minNonLiteralBlockSize = 1 + 1 + inputMargin + + if len(src) < minNonLiteralBlockSize { + dst = append(dst, Match{ + Unmatched: len(src), + }) + return dst + } + if len(src) > 65536 { + panic("block too long") + } + + var table [m0TableSize]uint16 + + // sLimit is when to stop looking for offset/length copies. The inputMargin + // lets us use a fast path for emitLiteral in the main loop, while we are + // looking for copies. + sLimit := len(src) - inputMargin + + // nextEmit is where in src the next emitLiteral should start from. + nextEmit := 0 + + // The encoded form must start with a literal, as there are no previous + // bytes to copy, so we start looking for hash matches at s == 1. + s := 1 + nextHash := m.hash(binary.LittleEndian.Uint64(src[s:])) + + for { + // Copied from the C++ snappy implementation: + // + // Heuristic match skipping: If 32 bytes are scanned with no matches + // found, start looking only at every other byte. If 32 more bytes are + // scanned (or skipped), look at every third byte, etc.. When a match + // is found, immediately go back to looking at every byte. This is a + // small loss (~5% performance, ~0.1% density) for compressible data + // due to more bookkeeping, but for non-compressible data (such as + // JPEG) it's a huge win since the compressor quickly "realizes" the + // data is incompressible and doesn't bother looking for matches + // everywhere. + // + // The "skip" variable keeps track of how many bytes there are since + // the last match; dividing it by 32 (ie. right-shifting by five) gives + // the number of bytes to move ahead for each iteration. + skip := 32 + + nextS := s + candidate := 0 + for { + s = nextS + bytesBetweenHashLookups := skip >> 5 + nextS = s + bytesBetweenHashLookups + skip += bytesBetweenHashLookups + if nextS > sLimit { + goto emitRemainder + } + candidate = int(table[nextHash&m0TableMask]) + table[nextHash&m0TableMask] = uint16(s) + nextHash = m.hash(binary.LittleEndian.Uint64(src[nextS:])) + if m.MaxDistance != 0 && s-candidate > m.MaxDistance { + continue + } + if binary.LittleEndian.Uint32(src[s:]) == binary.LittleEndian.Uint32(src[candidate:]) { + break + } + } + + // Invariant: we have a 4-byte match at s. + base := s + s = extendMatch(src, candidate+4, s+4) + + origBase := base + if m.Lazy && base+1 < sLimit { + newBase := base + 1 + h := m.hash(binary.LittleEndian.Uint64(src[newBase:])) + newCandidate := int(table[h&m0TableMask]) + table[h&m0TableMask] = uint16(newBase) + okDistance := true + if m.MaxDistance != 0 && newBase-newCandidate > m.MaxDistance { + okDistance = false + } + if okDistance && binary.LittleEndian.Uint32(src[newBase:]) == binary.LittleEndian.Uint32(src[newCandidate:]) { + newS := extendMatch(src, newCandidate+4, newBase+4) + if newS-newBase > s-base+1 { + s = newS + base = newBase + candidate = newCandidate + } + } + } + + if m.MaxLength != 0 && s-base > m.MaxLength { + s = base + m.MaxLength + } + dst = append(dst, Match{ + Unmatched: base - nextEmit, + Length: s - base, + Distance: base - candidate, + }) + nextEmit = s + if s >= sLimit { + goto emitRemainder + } + + if m.Lazy { + // If lazy matching is enabled, we update the hash table for + // every byte in the match. + for i := origBase + 2; i < s-1; i++ { + x := binary.LittleEndian.Uint64(src[i:]) + table[m.hash(x)&m0TableMask] = uint16(i) + } + } + + // We could immediately start working at s now, but to improve + // compression we first update the hash table at s-1 and at s. + x := binary.LittleEndian.Uint64(src[s-1:]) + prevHash := m.hash(x >> 0) + table[prevHash&m0TableMask] = uint16(s - 1) + nextHash = m.hash(x >> 8) + } + +emitRemainder: + if nextEmit < len(src) { + dst = append(dst, Match{ + Unmatched: len(src) - nextEmit, + }) + } + return dst +} diff --git a/vendor/github.com/andybalholm/brotli/matchfinder/m4.go b/vendor/github.com/andybalholm/brotli/matchfinder/m4.go new file mode 100644 index 0000000..5b2acba --- /dev/null +++ b/vendor/github.com/andybalholm/brotli/matchfinder/m4.go @@ -0,0 +1,297 @@ +package matchfinder + +import ( + "encoding/binary" + "math/bits" + "runtime" +) + +// M4 is an implementation of the MatchFinder +// interface that uses a hash table to find matches, +// optional match chains, +// and the advanced parsing technique from +// https://fastcompression.blogspot.com/2011/12/advanced-parsing-strategies.html. +type M4 struct { + // MaxDistance is the maximum distance (in bytes) to look back for + // a match. The default is 65535. + MaxDistance int + + // MinLength is the length of the shortest match to return. + // The default is 4. + MinLength int + + // HashLen is the number of bytes to use to calculate the hashes. + // The maximum is 8 and the default is 6. + HashLen int + + // TableBits is the number of bits in the hash table indexes. + // The default is 17 (128K entries). + TableBits int + + // ChainLength is how many entries to search on the "match chain" of older + // locations with the same hash as the current location. + ChainLength int + + // DistanceBitCost is used when comparing two matches to see + // which is better. The comparison is primarily based on the length + // of the matches, but it can also take the distance into account, + // in terms of the number of bits needed to represent the distance. + // One byte of length is given a score of 256, so 32 (256/8) would + // be a reasonable first guess for the value of one bit. + // (The default is 0, which bases the comparison solely on length.) + DistanceBitCost int + + table []uint32 + chain []uint16 + + history []byte +} + +func (q *M4) Reset() { + for i := range q.table { + q.table[i] = 0 + } + q.history = q.history[:0] + q.chain = q.chain[:0] +} + +func (q *M4) score(m absoluteMatch) int { + return (m.End-m.Start)*256 + bits.LeadingZeros32(uint32(m.Start-m.Match))*q.DistanceBitCost +} + +func (q *M4) FindMatches(dst []Match, src []byte) []Match { + if q.MaxDistance == 0 { + q.MaxDistance = 65535 + } + if q.MinLength == 0 { + q.MinLength = 4 + } + if q.HashLen == 0 { + q.HashLen = 6 + } + if q.TableBits == 0 { + q.TableBits = 17 + } + if len(q.table) < 1< q.MaxDistance*2 { + // Trim down the history buffer. + delta := len(q.history) - q.MaxDistance + copy(q.history, q.history[delta:]) + q.history = q.history[:q.MaxDistance] + if q.ChainLength > 0 { + q.chain = q.chain[:q.MaxDistance] + } + + for i, v := range q.table { + newV := int(v) - delta + if newV < 0 { + newV = 0 + } + q.table[i] = uint32(newV) + } + } + + // Append src to the history buffer. + e.NextEmit = len(q.history) + q.history = append(q.history, src...) + if q.ChainLength > 0 { + q.chain = append(q.chain, make([]uint16, len(src))...) + } + src = q.history + + // matches stores the matches that have been found but not emitted, + // in reverse order. (matches[0] is the most recent one.) + var matches [3]absoluteMatch + for i := e.NextEmit; i < len(src)-7; i++ { + if matches[0] != (absoluteMatch{}) && i >= matches[0].End { + // We have found some matches, and we're far enough along that we probably + // won't find overlapping matches, so we might as well emit them. + if matches[1] != (absoluteMatch{}) { + e.trim(matches[1], matches[0].Start, q.MinLength) + } + e.emit(matches[0]) + matches = [3]absoluteMatch{} + } + + // Calculate and store the hash. + h := ((binary.LittleEndian.Uint64(src[i:]) & (1<<(8*q.HashLen) - 1)) * hashMul64) >> (64 - q.TableBits) + candidate := int(q.table[h]) + q.table[h] = uint32(i) + if q.ChainLength > 0 && candidate != 0 { + delta := i - candidate + if delta < 1<<16 { + q.chain[i] = uint16(delta) + } + } + + if i < matches[0].End && i != matches[0].End+2-q.HashLen { + continue + } + if candidate == 0 || i-candidate > q.MaxDistance { + continue + } + + // Look for a match. + var currentMatch absoluteMatch + + if i-candidate != matches[0].Start-matches[0].Match { + if binary.LittleEndian.Uint32(src[candidate:]) == binary.LittleEndian.Uint32(src[i:]) { + m := extendMatch2(src, i, candidate, e.NextEmit) + if m.End-m.Start > q.MinLength { + currentMatch = m + } + } + } + + for j := 0; j < q.ChainLength; j++ { + delta := q.chain[candidate] + if delta == 0 { + break + } + candidate -= int(delta) + if candidate <= 0 || i-candidate > q.MaxDistance { + break + } + if i-candidate != matches[0].Start-matches[0].Match { + if binary.LittleEndian.Uint32(src[candidate:]) == binary.LittleEndian.Uint32(src[i:]) { + m := extendMatch2(src, i, candidate, e.NextEmit) + if m.End-m.Start > q.MinLength && q.score(m) > q.score(currentMatch) { + currentMatch = m + } + } + } + } + + if currentMatch.End-currentMatch.Start < q.MinLength { + continue + } + + overlapPenalty := 0 + if matches[0] != (absoluteMatch{}) { + overlapPenalty = 275 + if currentMatch.Start <= matches[1].End { + // This match would completely replace the previous match, + // so there is no penalty for overlap. + overlapPenalty = 0 + } + } + + if q.score(currentMatch) <= q.score(matches[0])+overlapPenalty { + continue + } + + matches = [3]absoluteMatch{ + currentMatch, + matches[0], + matches[1], + } + + if matches[2] == (absoluteMatch{}) { + continue + } + + // We have three matches, so it's time to emit one and/or eliminate one. + switch { + case matches[0].Start < matches[2].End: + // The first and third matches overlap; discard the one in between. + matches = [3]absoluteMatch{ + matches[0], + matches[2], + absoluteMatch{}, + } + + case matches[0].Start < matches[2].End+q.MinLength: + // The first and third matches don't overlap, but there's no room for + // another match between them. Emit the first match and discard the second. + e.emit(matches[2]) + matches = [3]absoluteMatch{ + matches[0], + absoluteMatch{}, + absoluteMatch{}, + } + + default: + // Emit the first match, shortening it if necessary to avoid overlap with the second. + e.trim(matches[2], matches[1].Start, q.MinLength) + matches[2] = absoluteMatch{} + } + } + + // We've found all the matches now; emit the remaining ones. + if matches[1] != (absoluteMatch{}) { + e.trim(matches[1], matches[0].Start, q.MinLength) + } + if matches[0] != (absoluteMatch{}) { + e.emit(matches[0]) + } + + dst = e.Dst + if e.NextEmit < len(src) { + dst = append(dst, Match{ + Unmatched: len(src) - e.NextEmit, + }) + } + + return dst +} + +const hashMul64 = 0x1E35A7BD1E35A7BD + +// extendMatch returns the largest k such that k <= len(src) and that +// src[i:i+k-j] and src[j:k] have the same contents. +// +// It assumes that: +// +// 0 <= i && i < j && j <= len(src) +func extendMatch(src []byte, i, j int) int { + switch runtime.GOARCH { + case "amd64": + // As long as we are 8 or more bytes before the end of src, we can load and + // compare 8 bytes at a time. If those 8 bytes are equal, repeat. + for j+8 < len(src) { + iBytes := binary.LittleEndian.Uint64(src[i:]) + jBytes := binary.LittleEndian.Uint64(src[j:]) + if iBytes != jBytes { + // If those 8 bytes were not equal, XOR the two 8 byte values, and return + // the index of the first byte that differs. The BSF instruction finds the + // least significant 1 bit, the amd64 architecture is little-endian, and + // the shift by 3 converts a bit index to a byte index. + return j + bits.TrailingZeros64(iBytes^jBytes)>>3 + } + i, j = i+8, j+8 + } + case "386": + // On a 32-bit CPU, we do it 4 bytes at a time. + for j+4 < len(src) { + iBytes := binary.LittleEndian.Uint32(src[i:]) + jBytes := binary.LittleEndian.Uint32(src[j:]) + if iBytes != jBytes { + return j + bits.TrailingZeros32(iBytes^jBytes)>>3 + } + i, j = i+4, j+4 + } + } + for ; j < len(src) && src[i] == src[j]; i, j = i+1, j+1 { + } + return j +} + +// Given a 4-byte match at src[start] and src[candidate], extendMatch2 extends it +// upward as far as possible, and downward no farther than to min. +func extendMatch2(src []byte, start, candidate, min int) absoluteMatch { + end := extendMatch(src, candidate+4, start+4) + for start > min && candidate > 0 && src[start-1] == src[candidate-1] { + start-- + candidate-- + } + return absoluteMatch{ + Start: start, + End: end, + Match: candidate, + } +} diff --git a/vendor/github.com/andybalholm/brotli/matchfinder/matchfinder.go b/vendor/github.com/andybalholm/brotli/matchfinder/matchfinder.go new file mode 100644 index 0000000..f6bcfdb --- /dev/null +++ b/vendor/github.com/andybalholm/brotli/matchfinder/matchfinder.go @@ -0,0 +1,103 @@ +// The matchfinder package defines reusable components for data compression. +// +// Many compression libraries have two main parts: +// - Something that looks for repeated sequences of bytes +// - An encoder for the compressed data format (often an entropy coder) +// +// Although these are logically two separate steps, the implementations are +// usually closely tied together. You can't use flate's matcher with snappy's +// encoder, for example. This package defines interfaces and an intermediate +// representation to allow mixing and matching compression components. +package matchfinder + +import "io" + +// A Match is the basic unit of LZ77 compression. +type Match struct { + Unmatched int // the number of unmatched bytes since the previous match + Length int // the number of bytes in the matched string; it may be 0 at the end of the input + Distance int // how far back in the stream to copy from +} + +// A MatchFinder performs the LZ77 stage of compression, looking for matches. +type MatchFinder interface { + // FindMatches looks for matches in src, appends them to dst, and returns dst. + FindMatches(dst []Match, src []byte) []Match + + // Reset clears any internal state, preparing the MatchFinder to be used with + // a new stream. + Reset() +} + +// An Encoder encodes the data in its final format. +type Encoder interface { + // Encode appends the encoded format of src to dst, using the match + // information from matches. + Encode(dst []byte, src []byte, matches []Match, lastBlock bool) []byte + + // Reset clears any internal state, preparing the Encoder to be used with + // a new stream. + Reset() +} + +// A Writer uses MatchFinder and Encoder to write compressed data to Dest. +type Writer struct { + Dest io.Writer + MatchFinder MatchFinder + Encoder Encoder + + // BlockSize is the number of bytes to compress at a time. If it is zero, + // each Write operation will be treated as one block. + BlockSize int + + err error + inBuf []byte + outBuf []byte + matches []Match +} + +func (w *Writer) Write(p []byte) (n int, err error) { + if w.err != nil { + return 0, w.err + } + + if w.BlockSize == 0 { + return w.writeBlock(p, false) + } + + w.inBuf = append(w.inBuf, p...) + var pos int + for pos = 0; pos+w.BlockSize <= len(w.inBuf) && w.err == nil; pos += w.BlockSize { + w.writeBlock(w.inBuf[pos:pos+w.BlockSize], false) + } + if pos > 0 { + n := copy(w.inBuf, w.inBuf[pos:]) + w.inBuf = w.inBuf[:n] + } + + return len(p), w.err +} + +func (w *Writer) writeBlock(p []byte, lastBlock bool) (n int, err error) { + w.outBuf = w.outBuf[:0] + w.matches = w.MatchFinder.FindMatches(w.matches[:0], p) + w.outBuf = w.Encoder.Encode(w.outBuf, p, w.matches, lastBlock) + _, w.err = w.Dest.Write(w.outBuf) + return len(p), w.err +} + +func (w *Writer) Close() error { + w.writeBlock(w.inBuf, true) + w.inBuf = w.inBuf[:0] + return w.err +} + +func (w *Writer) Reset(newDest io.Writer) { + w.MatchFinder.Reset() + w.Encoder.Reset() + w.err = nil + w.inBuf = w.inBuf[:0] + w.outBuf = w.outBuf[:0] + w.matches = w.matches[:0] + w.Dest = newDest +} diff --git a/vendor/github.com/andybalholm/brotli/matchfinder/textencoder.go b/vendor/github.com/andybalholm/brotli/matchfinder/textencoder.go new file mode 100644 index 0000000..75ecc59 --- /dev/null +++ b/vendor/github.com/andybalholm/brotli/matchfinder/textencoder.go @@ -0,0 +1,53 @@ +package matchfinder + +import "fmt" + +// A TextEncoder is an Encoder that produces a human-readable representation of +// the LZ77 compression. Matches are replaced with symbols. +type TextEncoder struct{} + +func (t TextEncoder) Reset() {} + +func (t TextEncoder) Encode(dst []byte, src []byte, matches []Match, lastBlock bool) []byte { + pos := 0 + for _, m := range matches { + if m.Unmatched > 0 { + dst = append(dst, src[pos:pos+m.Unmatched]...) + pos += m.Unmatched + } + if m.Length > 0 { + dst = append(dst, []byte(fmt.Sprintf("<%d,%d>", m.Length, m.Distance))...) + pos += m.Length + } + } + if pos < len(src) { + dst = append(dst, src[pos:]...) + } + return dst +} + +// A NoMatchFinder implements MatchFinder, but doesn't find any matches. +// It can be used to implement the equivalent of the standard library flate package's +// HuffmanOnly setting. +type NoMatchFinder struct{} + +func (n NoMatchFinder) Reset() {} + +func (n NoMatchFinder) FindMatches(dst []Match, src []byte) []Match { + return append(dst, Match{ + Unmatched: len(src), + }) +} + +// AutoReset wraps a MatchFinder that can return references to data in previous +// blocks, and calls Reset before each block. It is useful for (e.g.) using a +// snappy Encoder with a MatchFinder designed for flate. (Snappy doesn't +// support references between blocks.) +type AutoReset struct { + MatchFinder +} + +func (a AutoReset) FindMatches(dst []Match, src []byte) []Match { + a.Reset() + return a.MatchFinder.FindMatches(dst, src) +} diff --git a/vendor/github.com/andybalholm/brotli/writer.go b/vendor/github.com/andybalholm/brotli/writer.go index 39feaef..8a68811 100644 --- a/vendor/github.com/andybalholm/brotli/writer.go +++ b/vendor/github.com/andybalholm/brotli/writer.go @@ -3,6 +3,8 @@ package brotli import ( "errors" "io" + + "github.com/andybalholm/brotli/matchfinder" ) const ( @@ -117,3 +119,44 @@ type nopCloser struct { } func (nopCloser) Close() error { return nil } + +// NewWriterV2 is like NewWriterLevel, but it uses the new implementation +// based on the matchfinder package. It currently supports up to level 7; +// if a higher level is specified, level 7 will be used. +func NewWriterV2(dst io.Writer, level int) *matchfinder.Writer { + var mf matchfinder.MatchFinder + if level < 2 { + mf = matchfinder.M0{Lazy: level == 1} + } else { + hashLen := 6 + if level >= 6 { + hashLen = 5 + } + chainLen := 64 + switch level { + case 2: + chainLen = 0 + case 3: + chainLen = 1 + case 4: + chainLen = 2 + case 5: + chainLen = 4 + case 6: + chainLen = 8 + } + mf = &matchfinder.M4{ + MaxDistance: 1 << 20, + ChainLength: chainLen, + HashLen: hashLen, + DistanceBitCost: 57, + } + } + + return &matchfinder.Writer{ + Dest: dst, + MatchFinder: mf, + Encoder: &Encoder{}, + BlockSize: 1 << 16, + } +} diff --git a/vendor/github.com/klauspost/compress/.goreleaser.yml b/vendor/github.com/klauspost/compress/.goreleaser.yml index 4c28dff..a229538 100644 --- a/vendor/github.com/klauspost/compress/.goreleaser.yml +++ b/vendor/github.com/klauspost/compress/.goreleaser.yml @@ -3,7 +3,6 @@ before: hooks: - ./gen.sh - - go install mvdan.cc/garble@v0.10.1 builds: - @@ -32,7 +31,6 @@ builds: - mips64le goarm: - 7 - gobinary: garble - id: "s2d" binary: s2d @@ -59,7 +57,6 @@ builds: - mips64le goarm: - 7 - gobinary: garble - id: "s2sx" binary: s2sx @@ -87,7 +84,6 @@ builds: - mips64le goarm: - 7 - gobinary: garble archives: - diff --git a/vendor/github.com/klauspost/compress/README.md b/vendor/github.com/klauspost/compress/README.md index 7e83f58..ac281fb 100644 --- a/vendor/github.com/klauspost/compress/README.md +++ b/vendor/github.com/klauspost/compress/README.md @@ -16,6 +16,26 @@ This package provides various compression algorithms. # changelog +* Jan 26th, 2024 - [v1.17.5](https://github.com/klauspost/compress/releases/tag/v1.17.5) + * flate: Fix reset with dictionary on custom window encodes by @klauspost in https://github.com/klauspost/compress/pull/912 + * zstd: Add Frame header encoding and stripping by @klauspost in https://github.com/klauspost/compress/pull/908 + * zstd: Limit better/best default window to 8MB by @klauspost in https://github.com/klauspost/compress/pull/913 + * zstd: Speed improvements by @greatroar in https://github.com/klauspost/compress/pull/896 https://github.com/klauspost/compress/pull/910 + * s2: Fix callbacks for skippable blocks and disallow 0xfe (Padding) by @Jille in https://github.com/klauspost/compress/pull/916 https://github.com/klauspost/compress/pull/917 +https://github.com/klauspost/compress/pull/919 https://github.com/klauspost/compress/pull/918 + +* Dec 1st, 2023 - [v1.17.4](https://github.com/klauspost/compress/releases/tag/v1.17.4) + * huff0: Speed up symbol counting by @greatroar in https://github.com/klauspost/compress/pull/887 + * huff0: Remove byteReader by @greatroar in https://github.com/klauspost/compress/pull/886 + * gzhttp: Allow overriding decompression on transport https://github.com/klauspost/compress/pull/892 + * gzhttp: Clamp compression level https://github.com/klauspost/compress/pull/890 + * gzip: Error out if reserved bits are set https://github.com/klauspost/compress/pull/891 + +* Nov 15th, 2023 - [v1.17.3](https://github.com/klauspost/compress/releases/tag/v1.17.3) + * fse: Fix max header size https://github.com/klauspost/compress/pull/881 + * zstd: Improve better/best compression https://github.com/klauspost/compress/pull/877 + * gzhttp: Fix missing content type on Close https://github.com/klauspost/compress/pull/883 + * Oct 22nd, 2023 - [v1.17.2](https://github.com/klauspost/compress/releases/tag/v1.17.2) * zstd: Fix rare *CORRUPTION* output in "best" mode. See https://github.com/klauspost/compress/pull/876 @@ -554,7 +574,7 @@ For direct deflate use, NewStatelessWriter and StatelessDeflate are available. S A `bufio.Writer` can of course be used to control write sizes. For example, to use a 4KB buffer: -``` +```go // replace 'ioutil.Discard' with your output. gzw, err := gzip.NewWriterLevel(ioutil.Discard, gzip.StatelessCompression) if err != nil { diff --git a/vendor/github.com/klauspost/compress/flate/deflate.go b/vendor/github.com/klauspost/compress/flate/deflate.go index de912e1..66d1657 100644 --- a/vendor/github.com/klauspost/compress/flate/deflate.go +++ b/vendor/github.com/klauspost/compress/flate/deflate.go @@ -212,7 +212,7 @@ func (d *compressor) writeBlockSkip(tok *tokens, index int, eof bool) error { // Should only be used after a start/reset. func (d *compressor) fillWindow(b []byte) { // Do not fill window if we are in store-only or huffman mode. - if d.level <= 0 { + if d.level <= 0 && d.level > -MinCustomWindowSize { return } if d.fast != nil { diff --git a/vendor/github.com/klauspost/compress/s2sx.mod b/vendor/github.com/klauspost/compress/s2sx.mod index 2263853..5a4412f 100644 --- a/vendor/github.com/klauspost/compress/s2sx.mod +++ b/vendor/github.com/klauspost/compress/s2sx.mod @@ -1,4 +1,4 @@ module github.com/klauspost/compress -go 1.16 +go 1.19 diff --git a/vendor/github.com/klauspost/compress/zstd/decodeheader.go b/vendor/github.com/klauspost/compress/zstd/decodeheader.go index f6a2409..6a5a298 100644 --- a/vendor/github.com/klauspost/compress/zstd/decodeheader.go +++ b/vendor/github.com/klauspost/compress/zstd/decodeheader.go @@ -95,42 +95,54 @@ type Header struct { // If there isn't enough input, io.ErrUnexpectedEOF is returned. // The FirstBlock.OK will indicate if enough information was available to decode the first block header. func (h *Header) Decode(in []byte) error { + _, err := h.DecodeAndStrip(in) + return err +} + +// DecodeAndStrip will decode the header from the beginning of the stream +// and on success return the remaining bytes. +// This will decode the frame header and the first block header if enough bytes are provided. +// It is recommended to provide at least HeaderMaxSize bytes. +// If the frame header cannot be read an error will be returned. +// If there isn't enough input, io.ErrUnexpectedEOF is returned. +// The FirstBlock.OK will indicate if enough information was available to decode the first block header. +func (h *Header) DecodeAndStrip(in []byte) (remain []byte, err error) { *h = Header{} if len(in) < 4 { - return io.ErrUnexpectedEOF + return nil, io.ErrUnexpectedEOF } h.HeaderSize += 4 b, in := in[:4], in[4:] if string(b) != frameMagic { if string(b[1:4]) != skippableFrameMagic || b[0]&0xf0 != 0x50 { - return ErrMagicMismatch + return nil, ErrMagicMismatch } if len(in) < 4 { - return io.ErrUnexpectedEOF + return nil, io.ErrUnexpectedEOF } h.HeaderSize += 4 h.Skippable = true h.SkippableID = int(b[0] & 0xf) h.SkippableSize = binary.LittleEndian.Uint32(in) - return nil + return in[4:], nil } // Read Window_Descriptor // https://github.com/facebook/zstd/blob/dev/doc/zstd_compression_format.md#window_descriptor if len(in) < 1 { - return io.ErrUnexpectedEOF + return nil, io.ErrUnexpectedEOF } fhd, in := in[0], in[1:] h.HeaderSize++ h.SingleSegment = fhd&(1<<5) != 0 h.HasCheckSum = fhd&(1<<2) != 0 if fhd&(1<<3) != 0 { - return errors.New("reserved bit set on frame header") + return nil, errors.New("reserved bit set on frame header") } if !h.SingleSegment { if len(in) < 1 { - return io.ErrUnexpectedEOF + return nil, io.ErrUnexpectedEOF } var wd byte wd, in = in[0], in[1:] @@ -148,7 +160,7 @@ func (h *Header) Decode(in []byte) error { size = 4 } if len(in) < int(size) { - return io.ErrUnexpectedEOF + return nil, io.ErrUnexpectedEOF } b, in = in[:size], in[size:] h.HeaderSize += int(size) @@ -178,7 +190,7 @@ func (h *Header) Decode(in []byte) error { if fcsSize > 0 { h.HasFCS = true if len(in) < fcsSize { - return io.ErrUnexpectedEOF + return nil, io.ErrUnexpectedEOF } b, in = in[:fcsSize], in[fcsSize:] h.HeaderSize += int(fcsSize) @@ -199,7 +211,7 @@ func (h *Header) Decode(in []byte) error { // Frame Header done, we will not fail from now on. if len(in) < 3 { - return nil + return in, nil } tmp := in[:3] bh := uint32(tmp[0]) | (uint32(tmp[1]) << 8) | (uint32(tmp[2]) << 16) @@ -209,7 +221,7 @@ func (h *Header) Decode(in []byte) error { cSize := int(bh >> 3) switch blockType { case blockTypeReserved: - return nil + return in, nil case blockTypeRLE: h.FirstBlock.Compressed = true h.FirstBlock.DecompressedSize = cSize @@ -225,5 +237,25 @@ func (h *Header) Decode(in []byte) error { } h.FirstBlock.OK = true - return nil + return in, nil +} + +// AppendTo will append the encoded header to the dst slice. +// There is no error checking performed on the header values. +func (h *Header) AppendTo(dst []byte) ([]byte, error) { + if h.Skippable { + magic := [4]byte{0x50, 0x2a, 0x4d, 0x18} + magic[0] |= byte(h.SkippableID & 0xf) + dst = append(dst, magic[:]...) + f := h.SkippableSize + return append(dst, uint8(f), uint8(f>>8), uint8(f>>16), uint8(f>>24)), nil + } + f := frameHeader{ + ContentSize: h.FrameContentSize, + WindowSize: uint32(h.WindowSize), + SingleSegment: h.SingleSegment, + Checksum: h.HasCheckSum, + DictID: h.DictionaryID, + } + return f.appendTo(dst), nil } diff --git a/vendor/github.com/klauspost/compress/zstd/enc_best.go b/vendor/github.com/klauspost/compress/zstd/enc_best.go index c81a153..87f4287 100644 --- a/vendor/github.com/klauspost/compress/zstd/enc_best.go +++ b/vendor/github.com/klauspost/compress/zstd/enc_best.go @@ -201,14 +201,6 @@ encodeLoop: if delta >= e.maxMatchOff || delta <= 0 || load3232(src, offset) != first { return } - if debugAsserts { - if offset >= s { - panic(fmt.Sprintf("offset: %d - s:%d - rep: %d - cur :%d - max: %d", offset, s, rep, e.cur, e.maxMatchOff)) - } - if !bytes.Equal(src[s:s+4], src[offset:offset+4]) { - panic(fmt.Sprintf("first match mismatch: %v != %v, first: %08x", src[s:s+4], src[offset:offset+4], first)) - } - } // Try to quick reject if we already have a long match. if m.length > 16 { left := len(src) - int(m.s+m.length) @@ -227,8 +219,10 @@ encodeLoop: } } l := 4 + e.matchlen(s+4, offset+4, src) - if true { + if m.rep <= 0 { // Extend candidate match backwards as far as possible. + // Do not extend repeats as we can assume they are optimal + // and offsets change if s == nextEmit. tMin := s - e.maxMatchOff if tMin < 0 { tMin = 0 @@ -239,7 +233,14 @@ encodeLoop: l++ } } - + if debugAsserts { + if offset >= s { + panic(fmt.Sprintf("offset: %d - s:%d - rep: %d - cur :%d - max: %d", offset, s, rep, e.cur, e.maxMatchOff)) + } + if !bytes.Equal(src[s:s+l], src[offset:offset+l]) { + panic(fmt.Sprintf("second match mismatch: %v != %v, first: %08x", src[s:s+4], src[offset:offset+4], first)) + } + } cand := match{offset: offset, s: s, length: l, rep: rep} cand.estBits(bitsPerByte) if m.est >= highScore || cand.est-m.est+(cand.s-m.s)*bitsPerByte>>10 < 0 { @@ -336,24 +337,31 @@ encodeLoop: } if debugAsserts { + if best.offset >= best.s { + panic(fmt.Sprintf("best.offset > s: %d >= %d", best.offset, best.s)) + } + if best.s < nextEmit { + panic(fmt.Sprintf("s %d < nextEmit %d", best.s, nextEmit)) + } + if best.offset < s-e.maxMatchOff { + panic(fmt.Sprintf("best.offset < s-e.maxMatchOff: %d < %d", best.offset, s-e.maxMatchOff)) + } if !bytes.Equal(src[best.s:best.s+best.length], src[best.offset:best.offset+best.length]) { panic(fmt.Sprintf("match mismatch: %v != %v", src[best.s:best.s+best.length], src[best.offset:best.offset+best.length])) } } // We have a match, we can store the forward value + s = best.s if best.rep > 0 { var seq seq seq.matchLen = uint32(best.length - zstdMinMatch) - if debugAsserts && s < nextEmit { - panic("s < nextEmit") - } addLiterals(&seq, best.s) // Repeat. If bit 4 is set, this is a non-lit repeat. seq.offset = uint32(best.rep & 3) if debugSequences { - println("repeat sequence", seq, "next s:", s) + println("repeat sequence", seq, "next s:", best.s, "off:", best.s-best.offset) } blk.sequences = append(blk.sequences, seq) @@ -396,7 +404,6 @@ encodeLoop: // A 4-byte match has been found. Update recent offsets. // We'll later see if more than 4 bytes. - s = best.s t := best.offset offset1, offset2, offset3 = s-t, offset1, offset2 diff --git a/vendor/github.com/klauspost/compress/zstd/encoder_options.go b/vendor/github.com/klauspost/compress/zstd/encoder_options.go index faaf819..20671dc 100644 --- a/vendor/github.com/klauspost/compress/zstd/encoder_options.go +++ b/vendor/github.com/klauspost/compress/zstd/encoder_options.go @@ -94,7 +94,7 @@ func WithEncoderConcurrency(n int) EOption { // The value must be a power of two between MinWindowSize and MaxWindowSize. // A larger value will enable better compression but allocate more memory and, // for above-default values, take considerably longer. -// The default value is determined by the compression level. +// The default value is determined by the compression level and max 8MB. func WithWindowSize(n int) EOption { return func(o *encoderOptions) error { switch { @@ -232,9 +232,9 @@ func WithEncoderLevel(l EncoderLevel) EOption { case SpeedDefault: o.windowSize = 8 << 20 case SpeedBetterCompression: - o.windowSize = 16 << 20 + o.windowSize = 8 << 20 case SpeedBestCompression: - o.windowSize = 32 << 20 + o.windowSize = 8 << 20 } } if !o.customALEntropy { diff --git a/vendor/github.com/klauspost/compress/zstd/frameenc.go b/vendor/github.com/klauspost/compress/zstd/frameenc.go index 2f5d5ed..667ca06 100644 --- a/vendor/github.com/klauspost/compress/zstd/frameenc.go +++ b/vendor/github.com/klauspost/compress/zstd/frameenc.go @@ -76,7 +76,7 @@ func (f frameHeader) appendTo(dst []byte) []byte { if f.SingleSegment { dst = append(dst, uint8(f.ContentSize)) } - // Unless SingleSegment is set, framessizes < 256 are nto stored. + // Unless SingleSegment is set, framessizes < 256 are not stored. case 1: f.ContentSize -= 256 dst = append(dst, uint8(f.ContentSize), uint8(f.ContentSize>>8)) diff --git a/vendor/github.com/klauspost/compress/zstd/fse_decoder_generic.go b/vendor/github.com/klauspost/compress/zstd/fse_decoder_generic.go index 332e51f..8adfebb 100644 --- a/vendor/github.com/klauspost/compress/zstd/fse_decoder_generic.go +++ b/vendor/github.com/klauspost/compress/zstd/fse_decoder_generic.go @@ -20,10 +20,9 @@ func (s *fseDecoder) buildDtable() error { if v == -1 { s.dt[highThreshold].setAddBits(uint8(i)) highThreshold-- - symbolNext[i] = 1 - } else { - symbolNext[i] = uint16(v) + v = 1 } + symbolNext[i] = uint16(v) } } @@ -35,10 +34,12 @@ func (s *fseDecoder) buildDtable() error { for ss, v := range s.norm[:s.symbolLen] { for i := 0; i < int(v); i++ { s.dt[position].setAddBits(uint8(ss)) - position = (position + step) & tableMask - for position > highThreshold { + for { // lowprob area position = (position + step) & tableMask + if position <= highThreshold { + break + } } } } diff --git a/vendor/github.com/klauspost/compress/zstd/seqdec_amd64.s b/vendor/github.com/klauspost/compress/zstd/seqdec_amd64.s index 974b997..5b06174 100644 --- a/vendor/github.com/klauspost/compress/zstd/seqdec_amd64.s +++ b/vendor/github.com/klauspost/compress/zstd/seqdec_amd64.s @@ -157,8 +157,7 @@ sequenceDecs_decode_amd64_ll_update_zero: // Update Literal Length State MOVBQZX DI, R14 - SHRQ $0x10, DI - MOVWQZX DI, DI + SHRL $0x10, DI LEAQ (BX)(R14*1), CX MOVQ DX, R15 MOVQ CX, BX @@ -177,8 +176,7 @@ sequenceDecs_decode_amd64_ll_update_zero: // Update Match Length State MOVBQZX R8, R14 - SHRQ $0x10, R8 - MOVWQZX R8, R8 + SHRL $0x10, R8 LEAQ (BX)(R14*1), CX MOVQ DX, R15 MOVQ CX, BX @@ -197,8 +195,7 @@ sequenceDecs_decode_amd64_ll_update_zero: // Update Offset State MOVBQZX R9, R14 - SHRQ $0x10, R9 - MOVWQZX R9, R9 + SHRL $0x10, R9 LEAQ (BX)(R14*1), CX MOVQ DX, R15 MOVQ CX, BX @@ -459,8 +456,7 @@ sequenceDecs_decode_56_amd64_ll_update_zero: // Update Literal Length State MOVBQZX DI, R14 - SHRQ $0x10, DI - MOVWQZX DI, DI + SHRL $0x10, DI LEAQ (BX)(R14*1), CX MOVQ DX, R15 MOVQ CX, BX @@ -479,8 +475,7 @@ sequenceDecs_decode_56_amd64_ll_update_zero: // Update Match Length State MOVBQZX R8, R14 - SHRQ $0x10, R8 - MOVWQZX R8, R8 + SHRL $0x10, R8 LEAQ (BX)(R14*1), CX MOVQ DX, R15 MOVQ CX, BX @@ -499,8 +494,7 @@ sequenceDecs_decode_56_amd64_ll_update_zero: // Update Offset State MOVBQZX R9, R14 - SHRQ $0x10, R9 - MOVWQZX R9, R9 + SHRL $0x10, R9 LEAQ (BX)(R14*1), CX MOVQ DX, R15 MOVQ CX, BX @@ -772,11 +766,10 @@ sequenceDecs_decode_bmi2_fill_2_end: BZHIQ R14, R15, R15 // Update Offset State - BZHIQ R8, R15, CX - SHRXQ R8, R15, R15 - MOVQ $0x00001010, R14 - BEXTRQ R14, R8, R8 - ADDQ CX, R8 + BZHIQ R8, R15, CX + SHRXQ R8, R15, R15 + SHRL $0x10, R8 + ADDQ CX, R8 // Load ctx.ofTable MOVQ ctx+16(FP), CX @@ -784,11 +777,10 @@ sequenceDecs_decode_bmi2_fill_2_end: MOVQ (CX)(R8*8), R8 // Update Match Length State - BZHIQ DI, R15, CX - SHRXQ DI, R15, R15 - MOVQ $0x00001010, R14 - BEXTRQ R14, DI, DI - ADDQ CX, DI + BZHIQ DI, R15, CX + SHRXQ DI, R15, R15 + SHRL $0x10, DI + ADDQ CX, DI // Load ctx.mlTable MOVQ ctx+16(FP), CX @@ -796,10 +788,9 @@ sequenceDecs_decode_bmi2_fill_2_end: MOVQ (CX)(DI*8), DI // Update Literal Length State - BZHIQ SI, R15, CX - MOVQ $0x00001010, R14 - BEXTRQ R14, SI, SI - ADDQ CX, SI + BZHIQ SI, R15, CX + SHRL $0x10, SI + ADDQ CX, SI // Load ctx.llTable MOVQ ctx+16(FP), CX @@ -1032,11 +1023,10 @@ sequenceDecs_decode_56_bmi2_fill_end: BZHIQ R14, R15, R15 // Update Offset State - BZHIQ R8, R15, CX - SHRXQ R8, R15, R15 - MOVQ $0x00001010, R14 - BEXTRQ R14, R8, R8 - ADDQ CX, R8 + BZHIQ R8, R15, CX + SHRXQ R8, R15, R15 + SHRL $0x10, R8 + ADDQ CX, R8 // Load ctx.ofTable MOVQ ctx+16(FP), CX @@ -1044,11 +1034,10 @@ sequenceDecs_decode_56_bmi2_fill_end: MOVQ (CX)(R8*8), R8 // Update Match Length State - BZHIQ DI, R15, CX - SHRXQ DI, R15, R15 - MOVQ $0x00001010, R14 - BEXTRQ R14, DI, DI - ADDQ CX, DI + BZHIQ DI, R15, CX + SHRXQ DI, R15, R15 + SHRL $0x10, DI + ADDQ CX, DI // Load ctx.mlTable MOVQ ctx+16(FP), CX @@ -1056,10 +1045,9 @@ sequenceDecs_decode_56_bmi2_fill_end: MOVQ (CX)(DI*8), DI // Update Literal Length State - BZHIQ SI, R15, CX - MOVQ $0x00001010, R14 - BEXTRQ R14, SI, SI - ADDQ CX, SI + BZHIQ SI, R15, CX + SHRL $0x10, SI + ADDQ CX, SI // Load ctx.llTable MOVQ ctx+16(FP), CX @@ -1967,8 +1955,7 @@ sequenceDecs_decodeSync_amd64_ll_update_zero: // Update Literal Length State MOVBQZX DI, R13 - SHRQ $0x10, DI - MOVWQZX DI, DI + SHRL $0x10, DI LEAQ (BX)(R13*1), CX MOVQ DX, R14 MOVQ CX, BX @@ -1987,8 +1974,7 @@ sequenceDecs_decodeSync_amd64_ll_update_zero: // Update Match Length State MOVBQZX R8, R13 - SHRQ $0x10, R8 - MOVWQZX R8, R8 + SHRL $0x10, R8 LEAQ (BX)(R13*1), CX MOVQ DX, R14 MOVQ CX, BX @@ -2007,8 +1993,7 @@ sequenceDecs_decodeSync_amd64_ll_update_zero: // Update Offset State MOVBQZX R9, R13 - SHRQ $0x10, R9 - MOVWQZX R9, R9 + SHRL $0x10, R9 LEAQ (BX)(R13*1), CX MOVQ DX, R14 MOVQ CX, BX @@ -2514,11 +2499,10 @@ sequenceDecs_decodeSync_bmi2_fill_2_end: BZHIQ R13, R14, R14 // Update Offset State - BZHIQ R8, R14, CX - SHRXQ R8, R14, R14 - MOVQ $0x00001010, R13 - BEXTRQ R13, R8, R8 - ADDQ CX, R8 + BZHIQ R8, R14, CX + SHRXQ R8, R14, R14 + SHRL $0x10, R8 + ADDQ CX, R8 // Load ctx.ofTable MOVQ ctx+16(FP), CX @@ -2526,11 +2510,10 @@ sequenceDecs_decodeSync_bmi2_fill_2_end: MOVQ (CX)(R8*8), R8 // Update Match Length State - BZHIQ DI, R14, CX - SHRXQ DI, R14, R14 - MOVQ $0x00001010, R13 - BEXTRQ R13, DI, DI - ADDQ CX, DI + BZHIQ DI, R14, CX + SHRXQ DI, R14, R14 + SHRL $0x10, DI + ADDQ CX, DI // Load ctx.mlTable MOVQ ctx+16(FP), CX @@ -2538,10 +2521,9 @@ sequenceDecs_decodeSync_bmi2_fill_2_end: MOVQ (CX)(DI*8), DI // Update Literal Length State - BZHIQ SI, R14, CX - MOVQ $0x00001010, R13 - BEXTRQ R13, SI, SI - ADDQ CX, SI + BZHIQ SI, R14, CX + SHRL $0x10, SI + ADDQ CX, SI // Load ctx.llTable MOVQ ctx+16(FP), CX @@ -3055,8 +3037,7 @@ sequenceDecs_decodeSync_safe_amd64_ll_update_zero: // Update Literal Length State MOVBQZX DI, R13 - SHRQ $0x10, DI - MOVWQZX DI, DI + SHRL $0x10, DI LEAQ (BX)(R13*1), CX MOVQ DX, R14 MOVQ CX, BX @@ -3075,8 +3056,7 @@ sequenceDecs_decodeSync_safe_amd64_ll_update_zero: // Update Match Length State MOVBQZX R8, R13 - SHRQ $0x10, R8 - MOVWQZX R8, R8 + SHRL $0x10, R8 LEAQ (BX)(R13*1), CX MOVQ DX, R14 MOVQ CX, BX @@ -3095,8 +3075,7 @@ sequenceDecs_decodeSync_safe_amd64_ll_update_zero: // Update Offset State MOVBQZX R9, R13 - SHRQ $0x10, R9 - MOVWQZX R9, R9 + SHRL $0x10, R9 LEAQ (BX)(R13*1), CX MOVQ DX, R14 MOVQ CX, BX @@ -3704,11 +3683,10 @@ sequenceDecs_decodeSync_safe_bmi2_fill_2_end: BZHIQ R13, R14, R14 // Update Offset State - BZHIQ R8, R14, CX - SHRXQ R8, R14, R14 - MOVQ $0x00001010, R13 - BEXTRQ R13, R8, R8 - ADDQ CX, R8 + BZHIQ R8, R14, CX + SHRXQ R8, R14, R14 + SHRL $0x10, R8 + ADDQ CX, R8 // Load ctx.ofTable MOVQ ctx+16(FP), CX @@ -3716,11 +3694,10 @@ sequenceDecs_decodeSync_safe_bmi2_fill_2_end: MOVQ (CX)(R8*8), R8 // Update Match Length State - BZHIQ DI, R14, CX - SHRXQ DI, R14, R14 - MOVQ $0x00001010, R13 - BEXTRQ R13, DI, DI - ADDQ CX, DI + BZHIQ DI, R14, CX + SHRXQ DI, R14, R14 + SHRL $0x10, DI + ADDQ CX, DI // Load ctx.mlTable MOVQ ctx+16(FP), CX @@ -3728,10 +3705,9 @@ sequenceDecs_decodeSync_safe_bmi2_fill_2_end: MOVQ (CX)(DI*8), DI // Update Literal Length State - BZHIQ SI, R14, CX - MOVQ $0x00001010, R13 - BEXTRQ R13, SI, SI - ADDQ CX, SI + BZHIQ SI, R14, CX + SHRL $0x10, SI + ADDQ CX, SI // Load ctx.llTable MOVQ ctx+16(FP), CX diff --git a/vendor/golang.org/x/crypto/internal/poly1305/bits_compat.go b/vendor/golang.org/x/crypto/internal/poly1305/bits_compat.go deleted file mode 100644 index d33c889..0000000 --- a/vendor/golang.org/x/crypto/internal/poly1305/bits_compat.go +++ /dev/null @@ -1,39 +0,0 @@ -// 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. - -//go: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 -} diff --git a/vendor/golang.org/x/crypto/internal/poly1305/bits_go1.13.go b/vendor/golang.org/x/crypto/internal/poly1305/bits_go1.13.go deleted file mode 100644 index 495c1fa..0000000 --- a/vendor/golang.org/x/crypto/internal/poly1305/bits_go1.13.go +++ /dev/null @@ -1,21 +0,0 @@ -// 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. - -//go: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) -} diff --git a/vendor/golang.org/x/crypto/internal/poly1305/sum_generic.go b/vendor/golang.org/x/crypto/internal/poly1305/sum_generic.go index e041da5..ec2202b 100644 --- a/vendor/golang.org/x/crypto/internal/poly1305/sum_generic.go +++ b/vendor/golang.org/x/crypto/internal/poly1305/sum_generic.go @@ -7,7 +7,10 @@ package poly1305 -import "encoding/binary" +import ( + "encoding/binary" + "math/bits" +) // Poly1305 [RFC 7539] is a relatively simple algorithm: the authentication tag // for a 64 bytes message is approximately @@ -114,13 +117,13 @@ type uint128 struct { } func mul64(a, b uint64) uint128 { - hi, lo := bitsMul64(a, b) + hi, lo := bits.Mul64(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) + lo, c := bits.Add64(a.lo, b.lo, 0) + hi, c := bits.Add64(a.hi, b.hi, c) if c != 0 { panic("poly1305: unexpected overflow") } @@ -155,8 +158,8 @@ func updateGeneric(state *macState, msg []byte) { // 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) + h0, c = bits.Add64(h0, binary.LittleEndian.Uint64(msg[0:8]), 0) + h1, c = bits.Add64(h1, binary.LittleEndian.Uint64(msg[8:16]), c) h2 += c + 1 msg = msg[TagSize:] @@ -165,8 +168,8 @@ func updateGeneric(state *macState, msg []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) + h0, c = bits.Add64(h0, binary.LittleEndian.Uint64(buf[0:8]), 0) + h1, c = bits.Add64(h1, binary.LittleEndian.Uint64(buf[8:16]), c) h2 += c msg = nil @@ -219,9 +222,9 @@ func updateGeneric(state *macState, msg []byte) { 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) + t1, c := bits.Add64(m1.lo, m0.hi, 0) + t2, c := bits.Add64(m2.lo, m1.hi, c) + t3, _ := bits.Add64(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 @@ -243,14 +246,14 @@ func updateGeneric(state *macState, msg []byte) { // 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) + h0, c = bits.Add64(h0, cc.lo, 0) + h1, c = bits.Add64(h1, cc.hi, c) h2 += c cc = shiftRightBy2(cc) - h0, c = bitsAdd64(h0, cc.lo, 0) - h1, c = bitsAdd64(h1, cc.hi, c) + h0, c = bits.Add64(h0, cc.lo, 0) + h1, c = bits.Add64(h1, cc.hi, c) h2 += c // h2 is at most 3 + 1 + 1 = 5, making the whole of h at most @@ -287,9 +290,9 @@ func finalize(out *[TagSize]byte, h *[3]uint64, s *[2]uint64) { // 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) + hMinusP0, b := bits.Sub64(h0, p0, 0) + hMinusP1, b := bits.Sub64(h1, p1, b) + _, b = bits.Sub64(h2, p2, b) // h = h if h < p else h - p h0 = select64(b, h0, hMinusP0) @@ -301,8 +304,8 @@ func finalize(out *[TagSize]byte, h *[3]uint64, s *[2]uint64) { // // 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) + h0, c := bits.Add64(h0, s[0], 0) + h1, _ = bits.Add64(h1, s[1], c) binary.LittleEndian.PutUint64(out[0:8], h0) binary.LittleEndian.PutUint64(out[8:16], h1) diff --git a/vendor/modules.txt b/vendor/modules.txt index a6c5e0c..3710e65 100644 --- a/vendor/modules.txt +++ b/vendor/modules.txt @@ -4,9 +4,10 @@ filippo.io/age filippo.io/age/internal/bech32 filippo.io/age/internal/format filippo.io/age/internal/stream -# github.com/andybalholm/brotli v1.0.6 -## explicit; go 1.12 +# github.com/andybalholm/brotli v1.1.0 +## explicit; go 1.13 github.com/andybalholm/brotli +github.com/andybalholm/brotli/matchfinder # github.com/cloudflare/circl v1.3.7 ## explicit; go 1.19 github.com/cloudflare/circl/ecc/goldilocks @@ -20,7 +21,7 @@ github.com/cloudflare/circl/sign/ed448 # github.com/davecgh/go-spew v1.1.1 ## explicit github.com/davecgh/go-spew/spew -# github.com/klauspost/compress v1.17.4 +# github.com/klauspost/compress v1.17.6 ## explicit; go 1.19 github.com/klauspost/compress github.com/klauspost/compress/flate @@ -39,7 +40,7 @@ github.com/pmezard/go-difflib/difflib ## explicit; go 1.20 github.com/stretchr/testify/assert github.com/stretchr/testify/require -# golang.org/x/crypto v0.17.0 +# golang.org/x/crypto v0.18.0 ## explicit; go 1.18 golang.org/x/crypto/argon2 golang.org/x/crypto/blake2b