From 6c46afb087c198f7d9116af4d4aea52296ab78bc Mon Sep 17 00:00:00 2001 From: Aliaksandr Valialkin Date: Fri, 6 Sep 2019 00:45:26 +0300 Subject: [PATCH] vendor: update github.com/klauspost/compress from v1.7.6 to v1.8.2 --- go.mod | 2 +- go.sum | 4 +- .../klauspost/compress/huff0/decompress.go | 15 ++ .../klauspost/compress/huff0/huff0.go | 11 ++ .../klauspost/compress/snappy/decode_amd64.s | 16 +- .../klauspost/compress/snappy/decode_other.go | 22 ++- .../klauspost/compress/zstd/blockdec.go | 2 + .../klauspost/compress/zstd/decoder.go | 15 +- .../compress/zstd/decoder_options.go | 10 +- .../klauspost/compress/zstd/encoder.go | 22 ++- .../compress/zstd/encoder_options.go | 13 +- .../klauspost/compress/zstd/framedec.go | 3 + .../klauspost/compress/zstd/frameenc.go | 5 +- .../klauspost/compress/zstd/fse_predefined.go | 153 +++++++++--------- .../klauspost/compress/zstd/hash.go | 2 +- .../klauspost/compress/zstd/snappy.go | 1 + vendor/modules.txt | 2 +- 17 files changed, 189 insertions(+), 109 deletions(-) diff --git a/go.mod b/go.mod index 0a0022990..fa7437210 100644 --- a/go.mod +++ b/go.mod @@ -6,7 +6,7 @@ require ( github.com/cespare/xxhash/v2 v2.0.1-0.20190104013014-3767db7a7e18 github.com/golang/snappy v0.0.1 github.com/google/go-cmp v0.3.0 // indirect - github.com/klauspost/compress v1.7.6 + github.com/klauspost/compress v1.8.2 github.com/spaolacci/murmur3 v1.1.0 // indirect github.com/valyala/fastjson v1.4.1 github.com/valyala/gozstd v1.6.1 diff --git a/go.sum b/go.sum index cdbb0b48f..0e98c75d9 100644 --- a/go.sum +++ b/go.sum @@ -20,8 +20,8 @@ github.com/google/go-cmp v0.3.0 h1:crn/baboCvb5fXaQ0IJ1SGTsTVrWpDsCWC8EGETZijY= github.com/google/go-cmp v0.3.0/go.mod h1:8QqcDgzrUqlUb/G2PQTWiueGozuR1884gddMywk6iLU= github.com/klauspost/compress v1.4.0/go.mod h1:RyIbtBH6LamlWaDj8nUwkbUhJ87Yi3uG0guNDohfE1A= github.com/klauspost/compress v1.4.1/go.mod h1:RyIbtBH6LamlWaDj8nUwkbUhJ87Yi3uG0guNDohfE1A= -github.com/klauspost/compress v1.7.6 h1:GH2karLOcuZtA5a3+KuzSU33A2cvcHGbtEWM6K4t7oU= -github.com/klauspost/compress v1.7.6/go.mod h1:RyIbtBH6LamlWaDj8nUwkbUhJ87Yi3uG0guNDohfE1A= +github.com/klauspost/compress v1.8.2 h1:Bx0qjetmNjdFXASH02NSAREKpiaDwkO1DRZ3dV2KCcs= +github.com/klauspost/compress v1.8.2/go.mod h1:RyIbtBH6LamlWaDj8nUwkbUhJ87Yi3uG0guNDohfE1A= github.com/klauspost/cpuid v0.0.0-20180405133222-e7e905edc00e/go.mod h1:Pj4uuM528wm8OyEC2QMXAi2YiTZ96dNQPGgoMS4s3ek= github.com/klauspost/cpuid v1.2.0 h1:NMpwD2G9JSFOE1/TJjGSo5zG7Yb2bTe7eq1jH+irmeE= github.com/klauspost/cpuid v1.2.0/go.mod h1:Pj4uuM528wm8OyEC2QMXAi2YiTZ96dNQPGgoMS4s3ek= diff --git a/vendor/github.com/klauspost/compress/huff0/decompress.go b/vendor/github.com/klauspost/compress/huff0/decompress.go index 61ef96ae4..43b4815b3 100644 --- a/vendor/github.com/klauspost/compress/huff0/decompress.go +++ b/vendor/github.com/klauspost/compress/huff0/decompress.go @@ -193,14 +193,26 @@ func (s *Scratch) Decompress1X(in []byte) (out []byte, err error) { tmp[off+3] = hasDec(dt[br.peekBitsFast(s.actualTableLog)&tlMask]) off += 4 if off == 0 { + if len(s.Out)+256 > s.MaxDecodedSize { + br.close() + return nil, ErrMaxDecodedSizeExceeded + } s.Out = append(s.Out, tmp...) } } + if len(s.Out)+int(off) > s.MaxDecodedSize { + br.close() + return nil, ErrMaxDecodedSizeExceeded + } s.Out = append(s.Out, tmp[:off]...) for !br.finished() { br.fill() + if len(s.Out) >= s.MaxDecodedSize { + br.close() + return nil, ErrMaxDecodedSizeExceeded + } s.Out = append(s.Out, decode()) } return s.Out, br.close() @@ -218,6 +230,9 @@ func (s *Scratch) Decompress4X(in []byte, dstSize int) (out []byte, err error) { if len(in) < 6+(4*1) { return nil, errors.New("input too small") } + if dstSize > s.MaxDecodedSize { + return nil, ErrMaxDecodedSizeExceeded + } // TODO: We do not detect when we overrun a buffer, except if the last one does. var br [4]bitReader diff --git a/vendor/github.com/klauspost/compress/huff0/huff0.go b/vendor/github.com/klauspost/compress/huff0/huff0.go index 50d02e440..6f823f94d 100644 --- a/vendor/github.com/klauspost/compress/huff0/huff0.go +++ b/vendor/github.com/klauspost/compress/huff0/huff0.go @@ -35,6 +35,9 @@ var ( // ErrTooBig is return if input is too large for a single block. ErrTooBig = errors.New("input too big") + + // ErrMaxDecodedSizeExceeded is return if input is too large for a single block. + ErrMaxDecodedSizeExceeded = errors.New("maximum output size exceeded") ) type ReusePolicy uint8 @@ -86,6 +89,11 @@ type Scratch struct { // Reuse will specify the reuse policy Reuse ReusePolicy + // MaxDecodedSize will set the maximum allowed output size. + // This value will automatically be set to BlockSizeMax if not set. + // Decoders will return ErrMaxDecodedSizeExceeded is this limit is exceeded. + MaxDecodedSize int + br byteReader symbolLen uint16 // Length of active part of the symbol table. maxCount int // count of the most probable symbol @@ -116,6 +124,9 @@ func (s *Scratch) prepare(in []byte) (*Scratch, error) { if s.TableLog > tableLogMax { return nil, fmt.Errorf("tableLog (%d) > maxTableLog (%d)", s.TableLog, tableLogMax) } + if s.MaxDecodedSize <= 0 || s.MaxDecodedSize > BlockSizeMax { + s.MaxDecodedSize = BlockSizeMax + } if s.clearCount && s.maxCount == 0 { for i := range s.count { s.count[i] = 0 diff --git a/vendor/github.com/klauspost/compress/snappy/decode_amd64.s b/vendor/github.com/klauspost/compress/snappy/decode_amd64.s index e6179f65e..1c66e3723 100644 --- a/vendor/github.com/klauspost/compress/snappy/decode_amd64.s +++ b/vendor/github.com/klauspost/compress/snappy/decode_amd64.s @@ -184,9 +184,7 @@ tagLit60Plus: // checks. In the asm version, we code it once instead of once per switch case. ADDQ CX, SI SUBQ $58, SI - MOVQ SI, BX - SUBQ R11, BX - CMPQ BX, R12 + CMPQ SI, R13 JA errCorrupt // case x == 60: @@ -232,9 +230,7 @@ tagCopy4: ADDQ $5, SI // if uint(s) > uint(len(src)) { etc } - MOVQ SI, BX - SUBQ R11, BX - CMPQ BX, R12 + CMPQ SI, R13 JA errCorrupt // length = 1 + int(src[s-5])>>2 @@ -251,9 +247,7 @@ tagCopy2: ADDQ $3, SI // if uint(s) > uint(len(src)) { etc } - MOVQ SI, BX - SUBQ R11, BX - CMPQ BX, R12 + CMPQ SI, R13 JA errCorrupt // length = 1 + int(src[s-3])>>2 @@ -277,9 +271,7 @@ tagCopy: ADDQ $2, SI // if uint(s) > uint(len(src)) { etc } - MOVQ SI, BX - SUBQ R11, BX - CMPQ BX, R12 + CMPQ SI, R13 JA errCorrupt // offset = int(uint32(src[s-2])&0xe0<<3 | uint32(src[s-1])) diff --git a/vendor/github.com/klauspost/compress/snappy/decode_other.go b/vendor/github.com/klauspost/compress/snappy/decode_other.go index 8c9f2049b..94a96c5d7 100644 --- a/vendor/github.com/klauspost/compress/snappy/decode_other.go +++ b/vendor/github.com/klauspost/compress/snappy/decode_other.go @@ -85,14 +85,28 @@ func decode(dst, src []byte) int { if offset <= 0 || d < offset || length > len(dst)-d { return decodeErrCodeCorrupt } - // Copy from an earlier sub-slice of dst to a later sub-slice. Unlike - // the built-in copy function, this byte-by-byte copy always runs + // Copy from an earlier sub-slice of dst to a later sub-slice. + // If no overlap, use the built-in copy: + if offset > length { + copy(dst[d:d+length], dst[d-offset:]) + d += length + continue + } + + // Unlike the built-in copy function, this byte-by-byte copy always runs // forwards, even if the slices overlap. Conceptually, this is: // // d += forwardCopy(dst[d:d+length], dst[d-offset:]) - for end := d + length; d != end; d++ { - dst[d] = dst[d-offset] + // + // We align the slices into a and b and show the compiler they are the same size. + // This allows the loop to run without bounds checks. + a := dst[d : d+length] + b := dst[d-offset:] + b = b[:len(a)] + for i := range a { + a[i] = b[i] } + d += length } if d != len(dst) { return decodeErrCodeCorrupt diff --git a/vendor/github.com/klauspost/compress/zstd/blockdec.go b/vendor/github.com/klauspost/compress/zstd/blockdec.go index b87046f28..3e161ea15 100644 --- a/vendor/github.com/klauspost/compress/zstd/blockdec.go +++ b/vendor/github.com/klauspost/compress/zstd/blockdec.go @@ -448,6 +448,7 @@ func (b *blockDec) decodeCompressed(hist *history) error { } // Use our out buffer. huff.Out = b.literalBuf[:0] + huff.MaxDecodedSize = litRegenSize if fourStreams { literals, err = huff.Decompress4X(literals, litRegenSize) } else { @@ -610,6 +611,7 @@ func (b *blockDec) decodeCompressed(hist *history) error { // Use our out buffer. huff = hist.huffTree huff.Out = b.literalBuf[:0] + huff.MaxDecodedSize = litRegenSize if fourStreams { literals, err = huff.Decompress4X(literals, litRegenSize) } else { diff --git a/vendor/github.com/klauspost/compress/zstd/decoder.go b/vendor/github.com/klauspost/compress/zstd/decoder.go index a32a34cb6..098f9bf99 100644 --- a/vendor/github.com/klauspost/compress/zstd/decoder.go +++ b/vendor/github.com/klauspost/compress/zstd/decoder.go @@ -75,6 +75,7 @@ var ( // The Reset function can be used to initiate a new stream, which is will considerably // reduce the allocations normally caused by NewReader. func NewReader(r io.Reader, opts ...DOption) (*Decoder, error) { + initPredefined() var d Decoder d.o.setDefault() for _, o := range opts { @@ -285,10 +286,6 @@ func (d *Decoder) DecodeAll(input, dst []byte) ([]byte, error) { d.frames <- frame }() frame.bBuf = input - if cap(dst) == 0 { - // Allocate 1MB by default if nothing is provided. - dst = make([]byte, 0, 1<<20) - } for { err := frame.reset(&frame.bBuf) @@ -309,6 +306,16 @@ func (d *Decoder) DecodeAll(input, dst []byte) ([]byte, error) { dst = dst2 } } + if cap(dst) == 0 { + // Allocate window size * 2 by default if nothing is provided and we didn't get frame content size. + size := frame.WindowSize * 2 + // Cap to 1 MB. + if size > 1<<20 { + size = 1 << 20 + } + dst = make([]byte, 0, frame.WindowSize) + } + dst, err = frame.runDecoder(dst, block) if err != nil { return dst, err diff --git a/vendor/github.com/klauspost/compress/zstd/decoder_options.go b/vendor/github.com/klauspost/compress/zstd/decoder_options.go index 52c1eb066..2ac9cd2dd 100644 --- a/vendor/github.com/klauspost/compress/zstd/decoder_options.go +++ b/vendor/github.com/klauspost/compress/zstd/decoder_options.go @@ -50,15 +50,17 @@ func WithDecoderConcurrency(n int) DOption { } // WithDecoderMaxMemory allows to set a maximum decoded size for in-memory -// (non-streaming) operations. -// Maxmimum and default is 1 << 63 bytes. +// non-streaming operations or maximum window size for streaming operations. +// This can be used to control memory usage of potentially hostile content. +// For streaming operations, the maximum window size is capped at 1<<30 bytes. +// Maximum and default is 1 << 63 bytes. func WithDecoderMaxMemory(n uint64) DOption { return func(o *decoderOptions) error { if n == 0 { - return errors.New("WithDecoderMaxmemory must be at least 1") + return errors.New("WithDecoderMaxMemory must be at least 1") } if n > 1<<63 { - return fmt.Errorf("WithDecoderMaxmemorymust be less than 1 << 63") + return fmt.Errorf("WithDecoderMaxmemory must be less than 1 << 63") } o.maxDecodedSize = n return nil diff --git a/vendor/github.com/klauspost/compress/zstd/encoder.go b/vendor/github.com/klauspost/compress/zstd/encoder.go index b7011be29..a855655c4 100644 --- a/vendor/github.com/klauspost/compress/zstd/encoder.go +++ b/vendor/github.com/klauspost/compress/zstd/encoder.go @@ -59,6 +59,7 @@ type encoderState struct { // NewWriter will create a new Zstandard encoder. // If the encoder will be used for encoding blocks a nil writer can be used. func NewWriter(w io.Writer, opts ...EOption) (*Encoder, error) { + initPredefined() var e Encoder e.o.setDefault() for _, o := range opts { @@ -393,12 +394,31 @@ func (e *Encoder) Close() error { // EncodeAll will encode all input in src and append it to dst. // This function can be called concurrently, but each call will only run on a single goroutine. -// If empty input is given, nothing is returned. +// If empty input is given, nothing is returned, unless WithZeroFrames is specified. // Encoded blocks can be concatenated and the result will be the combined input stream. // Data compressed with EncodeAll can be decoded with the Decoder, // using either a stream or DecodeAll. func (e *Encoder) EncodeAll(src, dst []byte) []byte { if len(src) == 0 { + if e.o.fullZero { + // Add frame header. + fh := frameHeader{ + ContentSize: 0, + WindowSize: minWindowSize, + SingleSegment: true, + // Adding a checksum would be a waste of space. + Checksum: false, + DictID: 0, + } + dst, _ = fh.appendTo(dst) + + // Write raw block as last one only. + var blk blockHeader + blk.setSize(0) + blk.setType(blockTypeRaw) + blk.setLast(true) + dst = blk.appendTo(dst) + } return dst } e.init.Do(func() { diff --git a/vendor/github.com/klauspost/compress/zstd/encoder_options.go b/vendor/github.com/klauspost/compress/zstd/encoder_options.go index 6e210c4a0..93548391a 100644 --- a/vendor/github.com/klauspost/compress/zstd/encoder_options.go +++ b/vendor/github.com/klauspost/compress/zstd/encoder_options.go @@ -6,7 +6,7 @@ import ( "strings" ) -// DOption is an option for creating a encoder. +// EOption is an option for creating a encoder. type EOption func(*encoderOptions) error // options retains accumulated state of multiple options. @@ -18,6 +18,7 @@ type encoderOptions struct { blockSize int windowSize int level EncoderLevel + fullZero bool } func (o *encoderOptions) setDefault() { @@ -166,6 +167,16 @@ func WithEncoderLevel(l EncoderLevel) EOption { } } +// WithZeroFrames will encode 0 length input as full frames. +// This can be needed for compatibility with zstandard usage, +// but is not needed for this package. +func WithZeroFrames(b bool) EOption { + return func(o *encoderOptions) error { + o.fullZero = b + return nil + } +} + // WithSingleSegment will set the "single segment" flag when EncodeAll is used. // If this flag is set, data must be regenerated within a single continuous memory segment. // In this case, Window_Descriptor byte is skipped, but Frame_Content_Size is necessarily present. diff --git a/vendor/github.com/klauspost/compress/zstd/framedec.go b/vendor/github.com/klauspost/compress/zstd/framedec.go index 0c2a62307..839a95fbf 100644 --- a/vendor/github.com/klauspost/compress/zstd/framedec.go +++ b/vendor/github.com/klauspost/compress/zstd/framedec.go @@ -62,6 +62,9 @@ func newFrameDec(o decoderOptions) *frameDec { o: o, maxWindowSize: 1 << 30, } + if d.maxWindowSize > o.maxDecodedSize { + d.maxWindowSize = o.maxDecodedSize + } return &d } diff --git a/vendor/github.com/klauspost/compress/zstd/frameenc.go b/vendor/github.com/klauspost/compress/zstd/frameenc.go index acac32527..4479cfe18 100644 --- a/vendor/github.com/klauspost/compress/zstd/frameenc.go +++ b/vendor/github.com/klauspost/compress/zstd/frameenc.go @@ -5,7 +5,6 @@ package zstd import ( - "errors" "fmt" "io" "math" @@ -49,9 +48,7 @@ func (f frameHeader) appendTo(dst []byte) ([]byte, error) { windowLog := (bits.Len32(f.WindowSize-1) - winLogMin) << 3 dst = append(dst, uint8(windowLog)) } - if f.SingleSegment && f.ContentSize == 0 { - return nil, errors.New("single segment, but no size set") - } + switch fcs { case 0: if f.SingleSegment { diff --git a/vendor/github.com/klauspost/compress/zstd/fse_predefined.go b/vendor/github.com/klauspost/compress/zstd/fse_predefined.go index 5186de802..6c17dc17f 100644 --- a/vendor/github.com/klauspost/compress/zstd/fse_predefined.go +++ b/vendor/github.com/klauspost/compress/zstd/fse_predefined.go @@ -7,6 +7,7 @@ package zstd import ( "fmt" "math" + "sync" ) var ( @@ -69,85 +70,89 @@ func fillBase(dst []baseOffset, base uint32, bits ...uint8) { } } -func init() { - // Literals length codes - tmp := make([]baseOffset, 36) - for i := range tmp[:16] { - tmp[i] = baseOffset{ - baseLine: uint32(i), - addBits: 0, - } - } - fillBase(tmp[16:], 16, 1, 1, 1, 1, 2, 2, 3, 3, 4, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16) - symbolTableX[tableLiteralLengths] = tmp +var predef sync.Once - // Match length codes - tmp = make([]baseOffset, 53) - for i := range tmp[:32] { - tmp[i] = baseOffset{ - // The transformation adds the 3 length. - baseLine: uint32(i) + 3, - addBits: 0, +func initPredefined() { + predef.Do(func() { + // Literals length codes + tmp := make([]baseOffset, 36) + for i := range tmp[:16] { + tmp[i] = baseOffset{ + baseLine: uint32(i), + addBits: 0, + } } - } - fillBase(tmp[32:], 35, 1, 1, 1, 1, 2, 2, 3, 3, 4, 4, 5, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16) - symbolTableX[tableMatchLengths] = tmp + fillBase(tmp[16:], 16, 1, 1, 1, 1, 2, 2, 3, 3, 4, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16) + symbolTableX[tableLiteralLengths] = tmp - // Offset codes - tmp = make([]baseOffset, maxOffsetBits+1) - tmp[1] = baseOffset{ - baseLine: 1, - addBits: 1, - } - fillBase(tmp[2:], 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30) - symbolTableX[tableOffsets] = tmp + // Match length codes + tmp = make([]baseOffset, 53) + for i := range tmp[:32] { + tmp[i] = baseOffset{ + // The transformation adds the 3 length. + baseLine: uint32(i) + 3, + addBits: 0, + } + } + fillBase(tmp[32:], 35, 1, 1, 1, 1, 2, 2, 3, 3, 4, 4, 5, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16) + symbolTableX[tableMatchLengths] = tmp - // Fill predefined tables and transform them. - // https://github.com/facebook/zstd/blob/dev/doc/zstd_compression_format.md#default-distributions - for i := range fsePredef[:] { - f := &fsePredef[i] - switch tableIndex(i) { - case tableLiteralLengths: - // https://github.com/facebook/zstd/blob/ededcfca57366461021c922720878c81a5854a0a/lib/decompress/zstd_decompress_block.c#L243 - f.actualTableLog = 6 - copy(f.norm[:], []int16{4, 3, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 1, 1, 1, - 2, 2, 2, 2, 2, 2, 2, 2, 2, 3, 2, 1, 1, 1, 1, 1, - -1, -1, -1, -1}) - f.symbolLen = 36 - case tableOffsets: - // https://github.com/facebook/zstd/blob/ededcfca57366461021c922720878c81a5854a0a/lib/decompress/zstd_decompress_block.c#L281 - f.actualTableLog = 5 - copy(f.norm[:], []int16{ - 1, 1, 1, 1, 1, 1, 2, 2, 2, 1, 1, 1, 1, 1, 1, 1, - 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, -1, -1, -1}) - f.symbolLen = 29 - case tableMatchLengths: - //https://github.com/facebook/zstd/blob/ededcfca57366461021c922720878c81a5854a0a/lib/decompress/zstd_decompress_block.c#L304 - f.actualTableLog = 6 - copy(f.norm[:], []int16{ - 1, 4, 3, 2, 2, 2, 2, 2, 2, 1, 1, 1, 1, 1, 1, 1, - 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, - 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, - -1, -1, -1, -1, -1}) - f.symbolLen = 53 + // Offset codes + tmp = make([]baseOffset, maxOffsetBits+1) + tmp[1] = baseOffset{ + baseLine: 1, + addBits: 1, } - if err := f.buildDtable(); err != nil { - panic(fmt.Errorf("building table %v: %v", tableIndex(i), err)) - } - if err := f.transform(symbolTableX[i]); err != nil { - panic(fmt.Errorf("building table %v: %v", tableIndex(i), err)) - } - f.preDefined = true + fillBase(tmp[2:], 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30) + symbolTableX[tableOffsets] = tmp - // Create encoder as well - enc := &fsePredefEnc[i] - copy(enc.norm[:], f.norm[:]) - enc.symbolLen = f.symbolLen - enc.actualTableLog = f.actualTableLog - if err := enc.buildCTable(); err != nil { - panic(fmt.Errorf("building encoding table %v: %v", tableIndex(i), err)) + // Fill predefined tables and transform them. + // https://github.com/facebook/zstd/blob/dev/doc/zstd_compression_format.md#default-distributions + for i := range fsePredef[:] { + f := &fsePredef[i] + switch tableIndex(i) { + case tableLiteralLengths: + // https://github.com/facebook/zstd/blob/ededcfca57366461021c922720878c81a5854a0a/lib/decompress/zstd_decompress_block.c#L243 + f.actualTableLog = 6 + copy(f.norm[:], []int16{4, 3, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 1, 1, 1, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 3, 2, 1, 1, 1, 1, 1, + -1, -1, -1, -1}) + f.symbolLen = 36 + case tableOffsets: + // https://github.com/facebook/zstd/blob/ededcfca57366461021c922720878c81a5854a0a/lib/decompress/zstd_decompress_block.c#L281 + f.actualTableLog = 5 + copy(f.norm[:], []int16{ + 1, 1, 1, 1, 1, 1, 2, 2, 2, 1, 1, 1, 1, 1, 1, 1, + 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, -1, -1, -1}) + f.symbolLen = 29 + case tableMatchLengths: + //https://github.com/facebook/zstd/blob/ededcfca57366461021c922720878c81a5854a0a/lib/decompress/zstd_decompress_block.c#L304 + f.actualTableLog = 6 + copy(f.norm[:], []int16{ + 1, 4, 3, 2, 2, 2, 2, 2, 2, 1, 1, 1, 1, 1, 1, 1, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, + -1, -1, -1, -1, -1}) + f.symbolLen = 53 + } + if err := f.buildDtable(); err != nil { + panic(fmt.Errorf("building table %v: %v", tableIndex(i), err)) + } + if err := f.transform(symbolTableX[i]); err != nil { + panic(fmt.Errorf("building table %v: %v", tableIndex(i), err)) + } + f.preDefined = true + + // Create encoder as well + enc := &fsePredefEnc[i] + copy(enc.norm[:], f.norm[:]) + enc.symbolLen = f.symbolLen + enc.actualTableLog = f.actualTableLog + if err := enc.buildCTable(); err != nil { + panic(fmt.Errorf("building encoding table %v: %v", tableIndex(i), err)) + } + enc.setBits(bitTables[i]) + enc.preDefined = true } - enc.setBits(bitTables[i]) - enc.preDefined = true - } + }) } diff --git a/vendor/github.com/klauspost/compress/zstd/hash.go b/vendor/github.com/klauspost/compress/zstd/hash.go index 819d87f88..4a752067f 100644 --- a/vendor/github.com/klauspost/compress/zstd/hash.go +++ b/vendor/github.com/klauspost/compress/zstd/hash.go @@ -64,7 +64,7 @@ func hash6(u uint64, h uint8) uint32 { return uint32(((u << (64 - 48)) * prime6bytes) >> ((64 - h) & 63)) } -// hash6 returns the hash of the lowest 7 bytes of u to fit in a hash table with h bits. +// hash7 returns the hash of the lowest 7 bytes of u to fit in a hash table with h bits. // Preferably h should be a constant and should always be <64. func hash7(u uint64, h uint8) uint32 { return uint32(((u << (64 - 56)) * prime7bytes) >> ((64 - h) & 63)) diff --git a/vendor/github.com/klauspost/compress/zstd/snappy.go b/vendor/github.com/klauspost/compress/zstd/snappy.go index e9e518570..a048818f9 100644 --- a/vendor/github.com/klauspost/compress/zstd/snappy.go +++ b/vendor/github.com/klauspost/compress/zstd/snappy.go @@ -80,6 +80,7 @@ type SnappyConverter struct { // If any error is detected on the Snappy stream it is returned. // The number of bytes written is returned. func (r *SnappyConverter) Convert(in io.Reader, w io.Writer) (int64, error) { + initPredefined() r.err = nil r.r = in if r.block == nil { diff --git a/vendor/modules.txt b/vendor/modules.txt index 20e8a3ea0..5ef13d63b 100644 --- a/vendor/modules.txt +++ b/vendor/modules.txt @@ -6,7 +6,7 @@ github.com/VictoriaMetrics/metrics github.com/cespare/xxhash/v2 # github.com/golang/snappy v0.0.1 github.com/golang/snappy -# github.com/klauspost/compress v1.7.6 +# github.com/klauspost/compress v1.8.2 github.com/klauspost/compress/fse github.com/klauspost/compress/huff0 github.com/klauspost/compress/snappy