Vendor Update (#16121)

* update github.com/PuerkitoBio/goquery

* update github.com/alecthomas/chroma

* update github.com/blevesearch/bleve/v2

* update github.com/caddyserver/certmagic

* update github.com/go-enry/go-enry/v2

* update github.com/go-git/go-billy/v5

* update github.com/go-git/go-git/v5

* update github.com/go-redis/redis/v8

* update github.com/go-testfixtures/testfixtures/v3

* update github.com/jaytaylor/html2text

* update github.com/json-iterator/go

* update github.com/klauspost/compress

* update github.com/markbates/goth

* update github.com/mattn/go-isatty

* update github.com/mholt/archiver/v3

* update github.com/microcosm-cc/bluemonday

* update github.com/minio/minio-go/v7

* update github.com/prometheus/client_golang

* update github.com/unrolled/render

* update github.com/xanzy/go-gitlab

* update github.com/yuin/goldmark

* update github.com/yuin/goldmark-highlighting

Co-authored-by: techknowlogick <techknowlogick@gitea.io>

24 files changed, 722 insertions, 306 deletions

diff --git a/vendor/github.com/klauspost/compress/flate/deflate.go b/vendor/github.com/klauspost/compress/flate/deflate.go
index 40b5802deb..5283ac5a53 100644
--- a/vendor/github.com/klauspost/compress/flate/deflate.go
+++ b/vendor/github.com/klauspost/compress/flate/deflate.go
@@ -644,7 +644,7 @@ func (d *compressor) init(w io.Writer, level int) (err error) {
 		d.fill = (*compressor).fillBlock
 		d.step = (*compressor).store
 	case level == ConstantCompression:
-		d.w.logNewTablePenalty = 8
+		d.w.logNewTablePenalty = 10
 		d.window = make([]byte, 32<<10)
 		d.fill = (*compressor).fillBlock
 		d.step = (*compressor).storeHuff
diff --git a/vendor/github.com/klauspost/compress/flate/fast_encoder.go b/vendor/github.com/klauspost/compress/flate/fast_encoder.go
index 678f081052..347ac2c902 100644
--- a/vendor/github.com/klauspost/compress/flate/fast_encoder.go
+++ b/vendor/github.com/klauspost/compress/flate/fast_encoder.go
@@ -45,7 +45,7 @@ const (
 
 	bTableBits   = 17                                               // Bits used in the big tables
 	bTableSize   = 1 << bTableBits                                  // Size of the table
-	allocHistory = maxStoreBlockSize * 10                           // Size to preallocate for history.
+	allocHistory = maxStoreBlockSize * 5                            // Size to preallocate for history.
 	bufferReset  = (1 << 31) - allocHistory - maxStoreBlockSize - 1 // Reset the buffer offset when reaching this.
 )
 
diff --git a/vendor/github.com/klauspost/compress/flate/huffman_bit_writer.go b/vendor/github.com/klauspost/compress/flate/huffman_bit_writer.go
index db54be1398..3ad5e98072 100644
--- a/vendor/github.com/klauspost/compress/flate/huffman_bit_writer.go
+++ b/vendor/github.com/klauspost/compress/flate/huffman_bit_writer.go
@@ -6,6 +6,7 @@ package flate
 
 import (
 	"encoding/binary"
+	"fmt"
 	"io"
 )
 
@@ -27,7 +28,7 @@ const (
 	// after which bytes are flushed to the writer.
 	// Should preferably be a multiple of 6, since
 	// we accumulate 6 bytes between writes to the buffer.
-	bufferFlushSize = 240
+	bufferFlushSize = 246
 
 	// bufferSize is the actual output byte buffer size.
 	// It must have additional headroom for a flush
@@ -59,19 +60,31 @@ var offsetExtraBits = [64]int8{
 	14, 14, 15, 15, 16, 16, 17, 17, 18, 18, 19, 19, 20, 20,
 }
 
-var offsetBase = [64]uint32{
-	/* normal deflate */
-	0x000000, 0x000001, 0x000002, 0x000003, 0x000004,
-	0x000006, 0x000008, 0x00000c, 0x000010, 0x000018,
-	0x000020, 0x000030, 0x000040, 0x000060, 0x000080,
-	0x0000c0, 0x000100, 0x000180, 0x000200, 0x000300,
-	0x000400, 0x000600, 0x000800, 0x000c00, 0x001000,
-	0x001800, 0x002000, 0x003000, 0x004000, 0x006000,
+var offsetCombined = [32]uint32{}
 
-	/* extended window */
-	0x008000, 0x00c000, 0x010000, 0x018000, 0x020000,
-	0x030000, 0x040000, 0x060000, 0x080000, 0x0c0000,
-	0x100000, 0x180000, 0x200000, 0x300000,
+func init() {
+	var offsetBase = [64]uint32{
+		/* normal deflate */
+		0x000000, 0x000001, 0x000002, 0x000003, 0x000004,
+		0x000006, 0x000008, 0x00000c, 0x000010, 0x000018,
+		0x000020, 0x000030, 0x000040, 0x000060, 0x000080,
+		0x0000c0, 0x000100, 0x000180, 0x000200, 0x000300,
+		0x000400, 0x000600, 0x000800, 0x000c00, 0x001000,
+		0x001800, 0x002000, 0x003000, 0x004000, 0x006000,
+
+		/* extended window */
+		0x008000, 0x00c000, 0x010000, 0x018000, 0x020000,
+		0x030000, 0x040000, 0x060000, 0x080000, 0x0c0000,
+		0x100000, 0x180000, 0x200000, 0x300000,
+	}
+
+	for i := range offsetCombined[:] {
+		// Don't use extended window values...
+		if offsetBase[i] > 0x006000 {
+			continue
+		}
+		offsetCombined[i] = uint32(offsetExtraBits[i])<<16 | (offsetBase[i])
+	}
 }
 
 // The odd order in which the codegen code sizes are written.
@@ -88,15 +101,16 @@ type huffmanBitWriter struct {
 	bits            uint64
 	nbits           uint16
 	nbytes          uint8
+	lastHuffMan     bool
 	literalEncoding *huffmanEncoder
+	tmpLitEncoding  *huffmanEncoder
 	offsetEncoding  *huffmanEncoder
 	codegenEncoding *huffmanEncoder
 	err             error
 	lastHeader      int
 	// Set between 0 (reused block can be up to 2x the size)
 	logNewTablePenalty uint
-	lastHuffMan        bool
-	bytes              [256]byte
+	bytes              [256 + 8]byte
 	literalFreq        [lengthCodesStart + 32]uint16
 	offsetFreq         [32]uint16
 	codegenFreq        [codegenCodeCount]uint16
@@ -128,6 +142,7 @@ func newHuffmanBitWriter(w io.Writer) *huffmanBitWriter {
 	return &huffmanBitWriter{
 		writer:          w,
 		literalEncoding: newHuffmanEncoder(literalCount),
+		tmpLitEncoding:  newHuffmanEncoder(literalCount),
 		codegenEncoding: newHuffmanEncoder(codegenCodeCount),
 		offsetEncoding:  newHuffmanEncoder(offsetCodeCount),
 	}
@@ -745,9 +760,31 @@ func (w *huffmanBitWriter) writeTokens(tokens []token, leCodes, oeCodes []hcode)
 	offs := oeCodes[:32]
 	lengths := leCodes[lengthCodesStart:]
 	lengths = lengths[:32]
+
+	// Go 1.16 LOVES having these on stack.
+	bits, nbits, nbytes := w.bits, w.nbits, w.nbytes
+
 	for _, t := range tokens {
 		if t < matchType {
-			w.writeCode(lits[t.literal()])
+			//w.writeCode(lits[t.literal()])
+			c := lits[t.literal()]
+			bits |= uint64(c.code) << nbits
+			nbits += c.len
+			if nbits >= 48 {
+				binary.LittleEndian.PutUint64(w.bytes[nbytes:], bits)
+				//*(*uint64)(unsafe.Pointer(&w.bytes[nbytes])) = bits
+				bits >>= 48
+				nbits -= 48
+				nbytes += 6
+				if nbytes >= bufferFlushSize {
+					if w.err != nil {
+						nbytes = 0
+						return
+					}
+					_, w.err = w.writer.Write(w.bytes[:nbytes])
+					nbytes = 0
+				}
+			}
 			continue
 		}
 
@@ -759,38 +796,99 @@ func (w *huffmanBitWriter) writeTokens(tokens []token, leCodes, oeCodes []hcode)
 		} else {
 			// inlined
 			c := lengths[lengthCode&31]
-			w.bits |= uint64(c.code) << w.nbits
-			w.nbits += c.len
-			if w.nbits >= 48 {
-				w.writeOutBits()
+			bits |= uint64(c.code) << nbits
+			nbits += c.len
+			if nbits >= 48 {
+				binary.LittleEndian.PutUint64(w.bytes[nbytes:], bits)
+				//*(*uint64)(unsafe.Pointer(&w.bytes[nbytes])) = bits
+				bits >>= 48
+				nbits -= 48
+				nbytes += 6
+				if nbytes >= bufferFlushSize {
+					if w.err != nil {
+						nbytes = 0
+						return
+					}
+					_, w.err = w.writer.Write(w.bytes[:nbytes])
+					nbytes = 0
+				}
 			}
 		}
 
 		extraLengthBits := uint16(lengthExtraBits[lengthCode&31])
 		if extraLengthBits > 0 {
+			//w.writeBits(extraLength, extraLengthBits)
 			extraLength := int32(length - lengthBase[lengthCode&31])
-			w.writeBits(extraLength, extraLengthBits)
+			bits |= uint64(extraLength) << nbits
+			nbits += extraLengthBits
+			if nbits >= 48 {
+				binary.LittleEndian.PutUint64(w.bytes[nbytes:], bits)
+				//*(*uint64)(unsafe.Pointer(&w.bytes[nbytes])) = bits
+				bits >>= 48
+				nbits -= 48
+				nbytes += 6
+				if nbytes >= bufferFlushSize {
+					if w.err != nil {
+						nbytes = 0
+						return
+					}
+					_, w.err = w.writer.Write(w.bytes[:nbytes])
+					nbytes = 0
+				}
+			}
 		}
 		// Write the offset
 		offset := t.offset()
-		offsetCode := offsetCode(offset)
+		offsetCode := offset >> 16
+		offset &= matchOffsetOnlyMask
 		if false {
 			w.writeCode(offs[offsetCode&31])
 		} else {
 			// inlined
-			c := offs[offsetCode&31]
-			w.bits |= uint64(c.code) << w.nbits
-			w.nbits += c.len
-			if w.nbits >= 48 {
-				w.writeOutBits()
+			c := offs[offsetCode]
+			bits |= uint64(c.code) << nbits
+			nbits += c.len
+			if nbits >= 48 {
+				binary.LittleEndian.PutUint64(w.bytes[nbytes:], bits)
+				//*(*uint64)(unsafe.Pointer(&w.bytes[nbytes])) = bits
+				bits >>= 48
+				nbits -= 48
+				nbytes += 6
+				if nbytes >= bufferFlushSize {
+					if w.err != nil {
+						nbytes = 0
+						return
+					}
+					_, w.err = w.writer.Write(w.bytes[:nbytes])
+					nbytes = 0
+				}
 			}
 		}
-		extraOffsetBits := uint16(offsetExtraBits[offsetCode&63])
-		if extraOffsetBits > 0 {
-			extraOffset := int32(offset - offsetBase[offsetCode&63])
-			w.writeBits(extraOffset, extraOffsetBits)
+		offsetComb := offsetCombined[offsetCode]
+		if offsetComb > 1<<16 {
+			//w.writeBits(extraOffset, extraOffsetBits)
+			bits |= uint64(offset&matchOffsetOnlyMask-(offsetComb&0xffff)) << nbits
+			nbits += uint16(offsetComb >> 16)
+			if nbits >= 48 {
+				binary.LittleEndian.PutUint64(w.bytes[nbytes:], bits)
+				//*(*uint64)(unsafe.Pointer(&w.bytes[nbytes])) = bits
+				bits >>= 48
+				nbits -= 48
+				nbytes += 6
+				if nbytes >= bufferFlushSize {
+					if w.err != nil {
+						nbytes = 0
+						return
+					}
+					_, w.err = w.writer.Write(w.bytes[:nbytes])
+					nbytes = 0
+				}
+			}
 		}
 	}
+	// Restore...
+	w.bits, w.nbits, w.nbytes = bits, nbits, nbytes
+
 	if deferEOB {
 		w.writeCode(leCodes[endBlockMarker])
 	}
@@ -825,13 +923,28 @@ func (w *huffmanBitWriter) writeBlockHuff(eof bool, input []byte, sync bool) {
 		}
 	}
 
+	// Fill is rarely better...
+	const fill = false
+	const numLiterals = endBlockMarker + 1
+	const numOffsets = 1
+
 	// Add everything as literals
 	// We have to estimate the header size.
 	// Assume header is around 70 bytes:
 	// https://stackoverflow.com/a/25454430
 	const guessHeaderSizeBits = 70 * 8
-	estBits := histogramSize(input, w.literalFreq[:], !eof && !sync)
-	estBits += w.lastHeader + len(input)/32
+	histogram(input, w.literalFreq[:numLiterals], fill)
+	w.literalFreq[endBlockMarker] = 1
+	w.tmpLitEncoding.generate(w.literalFreq[:numLiterals], 15)
+	if fill {
+		// Clear fill...
+		for i := range w.literalFreq[:numLiterals] {
+			w.literalFreq[i] = 0
+		}
+		histogram(input, w.literalFreq[:numLiterals], false)
+	}
+	estBits := w.tmpLitEncoding.canReuseBits(w.literalFreq[:numLiterals])
+	estBits += w.lastHeader
 	if w.lastHeader == 0 {
 		estBits += guessHeaderSizeBits
 	}
@@ -839,33 +952,31 @@ func (w *huffmanBitWriter) writeBlockHuff(eof bool, input []byte, sync bool) {
 
 	// Store bytes, if we don't get a reasonable improvement.
 	ssize, storable := w.storedSize(input)
-	if storable && ssize < estBits {
+	if storable && ssize <= estBits {
 		w.writeStoredHeader(len(input), eof)
 		w.writeBytes(input)
 		return
 	}
 
-	reuseSize := 0
 	if w.lastHeader > 0 {
-		reuseSize = w.literalEncoding.bitLength(w.literalFreq[:256])
+		reuseSize := w.literalEncoding.canReuseBits(w.literalFreq[:256])
 
 		if estBits < reuseSize {
+			if debugDeflate {
+				//fmt.Println("not reusing, reuse:", reuseSize/8, "> new:", estBits/8, "- header est:", w.lastHeader/8)
+			}
 			// We owe an EOB
 			w.writeCode(w.literalEncoding.codes[endBlockMarker])
 			w.lastHeader = 0
+		} else if debugDeflate {
+			fmt.Println("reusing, reuse:", reuseSize/8, "> new:", estBits/8, "- header est:", w.lastHeader/8)
 		}
 	}
 
-	const numLiterals = endBlockMarker + 1
-	const numOffsets = 1
+	count := 0
 	if w.lastHeader == 0 {
-		if !eof && !sync {
-			// Generate a slightly suboptimal tree that can be used for all.
-			fillHist(w.literalFreq[:numLiterals])
-		}
-		w.literalFreq[endBlockMarker] = 1
-		w.literalEncoding.generate(w.literalFreq[:numLiterals], 15)
-
+		// Use the temp encoding, so swap.
+		w.literalEncoding, w.tmpLitEncoding = w.tmpLitEncoding, w.literalEncoding
 		// Generate codegen and codegenFrequencies, which indicates how to encode
 		// the literalEncoding and the offsetEncoding.
 		w.generateCodegen(numLiterals, numOffsets, w.literalEncoding, huffOffset)
@@ -876,34 +987,47 @@ func (w *huffmanBitWriter) writeBlockHuff(eof bool, input []byte, sync bool) {
 		w.writeDynamicHeader(numLiterals, numOffsets, numCodegens, eof)
 		w.lastHuffMan = true
 		w.lastHeader, _ = w.headerSize()
+		if debugDeflate {
+			count += w.lastHeader
+			fmt.Println("header:", count/8)
+		}
 	}
 
-	encoding := w.literalEncoding.codes[:257]
+	encoding := w.literalEncoding.codes[:256]
+	// Go 1.16 LOVES having these on stack. At least 1.5x the speed.
+	bits, nbits, nbytes := w.bits, w.nbits, w.nbytes
 	for _, t := range input {
 		// Bitwriting inlined, ~30% speedup
 		c := encoding[t]
-		w.bits |= uint64(c.code) << w.nbits
-		w.nbits += c.len
-		if w.nbits >= 48 {
-			bits := w.bits
-			w.bits >>= 48
-			w.nbits -= 48
-			n := w.nbytes
-			binary.LittleEndian.PutUint64(w.bytes[n:], bits)
-			n += 6
-			if n >= bufferFlushSize {
+		bits |= uint64(c.code) << nbits
+		nbits += c.len
+		if debugDeflate {
+			count += int(c.len)
+		}
+		if nbits >= 48 {
+			binary.LittleEndian.PutUint64(w.bytes[nbytes:], bits)
+			//*(*uint64)(unsafe.Pointer(&w.bytes[nbytes])) = bits
+			bits >>= 48
+			nbits -= 48
+			nbytes += 6
+			if nbytes >= bufferFlushSize {
 				if w.err != nil {
-					n = 0
+					nbytes = 0
 					return
 				}
-				w.write(w.bytes[:n])
-				n = 0
+				_, w.err = w.writer.Write(w.bytes[:nbytes])
+				nbytes = 0
 			}
-			w.nbytes = n
 		}
 	}
+	// Restore...
+	w.bits, w.nbits, w.nbytes = bits, nbits, nbytes
+
+	if debugDeflate {
+		fmt.Println("wrote", count/8, "bytes")
+	}
 	if eof || sync {
-		w.writeCode(encoding[endBlockMarker])
+		w.writeCode(w.literalEncoding.codes[endBlockMarker])
 		w.lastHeader = 0
 		w.lastHuffMan = false
 	}
diff --git a/vendor/github.com/klauspost/compress/flate/huffman_code.go b/vendor/github.com/klauspost/compress/flate/huffman_code.go
index 0d3445a1cc..67b2b38728 100644
--- a/vendor/github.com/klauspost/compress/flate/huffman_code.go
+++ b/vendor/github.com/klauspost/compress/flate/huffman_code.go
@@ -21,9 +21,13 @@ type hcode struct {
 }
 
 type huffmanEncoder struct {
-	codes     []hcode
-	freqcache []literalNode
-	bitCount  [17]int32
+	codes    []hcode
+	bitCount [17]int32
+
+	// Allocate a reusable buffer with the longest possible frequency table.
+	// Possible lengths are codegenCodeCount, offsetCodeCount and literalCount.
+	// The largest of these is literalCount, so we allocate for that case.
+	freqcache [literalCount + 1]literalNode
 }
 
 type literalNode struct {
@@ -132,6 +136,21 @@ func (h *huffmanEncoder) bitLengthRaw(b []byte) int {
 	return total
 }
 
+// canReuseBits returns the number of bits or math.MaxInt32 if the encoder cannot be reused.
+func (h *huffmanEncoder) canReuseBits(freq []uint16) int {
+	var total int
+	for i, f := range freq {
+		if f != 0 {
+			code := h.codes[i]
+			if code.len == 0 {
+				return math.MaxInt32
+			}
+			total += int(f) * int(code.len)
+		}
+	}
+	return total
+}
+
 // Return the number of literals assigned to each bit size in the Huffman encoding
 //
 // This method is only called when list.length >= 3
@@ -291,12 +310,6 @@ func (h *huffmanEncoder) assignEncodingAndSize(bitCount []int32, list []literalN
 // freq  An array of frequencies, in which frequency[i] gives the frequency of literal i.
 // maxBits  The maximum number of bits to use for any literal.
 func (h *huffmanEncoder) generate(freq []uint16, maxBits int32) {
-	if h.freqcache == nil {
-		// Allocate a reusable buffer with the longest possible frequency table.
-		// Possible lengths are codegenCodeCount, offsetCodeCount and literalCount.
-		// The largest of these is literalCount, so we allocate for that case.
-		h.freqcache = make([]literalNode, literalCount+1)
-	}
 	list := h.freqcache[:len(freq)+1]
 	// Number of non-zero literals
 	count := 0
@@ -330,10 +343,14 @@ func (h *huffmanEncoder) generate(freq []uint16, maxBits int32) {
 	h.assignEncodingAndSize(bitCount, list)
 }
 
+// atLeastOne clamps the result between 1 and 15.
 func atLeastOne(v float32) float32 {
 	if v < 1 {
 		return 1
 	}
+	if v > 15 {
+		return 15
+	}
 	return v
 }
 
@@ -346,31 +363,12 @@ func fillHist(b []uint16) {
 	}
 }
 
-// histogramSize accumulates a histogram of b in h.
-// An estimated size in bits is returned.
-// len(h) must be >= 256, and h's elements must be all zeroes.
-func histogramSize(b []byte, h []uint16, fill bool) (bits int) {
+func histogram(b []byte, h []uint16, fill bool) {
 	h = h[:256]
 	for _, t := range b {
 		h[t]++
 	}
-	total := len(b)
 	if fill {
-		for _, v := range h {
-			if v == 0 {
-				total++
-			}
-		}
+		fillHist(h)
 	}
-
-	invTotal := 1.0 / float32(total)
-	shannon := float32(0.0)
-	for _, v := range h {
-		if v > 0 {
-			n := float32(v)
-			shannon += atLeastOne(-mFastLog2(n*invTotal)) * n
-		}
-	}
-
-	return int(shannon + 0.99)
 }
diff --git a/vendor/github.com/klauspost/compress/flate/level5.go b/vendor/github.com/klauspost/compress/flate/level5.go
index d513f1ffd3..293a3a320b 100644
--- a/vendor/github.com/klauspost/compress/flate/level5.go
+++ b/vendor/github.com/klauspost/compress/flate/level5.go
@@ -182,12 +182,27 @@ func (e *fastEncL5) Encode(dst *tokens, src []byte) {
 		// match. But, prior to the match, src[nextEmit:s] are unmatched. Emit
 		// them as literal bytes.
 
-		// Extend the 4-byte match as long as possible.
 		if l == 0 {
+			// Extend the 4-byte match as long as possible.
 			l = e.matchlenLong(s+4, t+4, src) + 4
 		} else if l == maxMatchLength {
 			l += e.matchlenLong(s+l, t+l, src)
 		}
+
+		// Try to locate a better match by checking the end of best match...
+		if sAt := s + l; l < 30 && sAt < sLimit {
+			eLong := e.bTable[hash7(load6432(src, sAt), tableBits)].Cur.offset
+			// Test current
+			t2 := eLong - e.cur - l
+			off := s - t2
+			if t2 >= 0 && off < maxMatchOffset && off > 0 {
+				if l2 := e.matchlenLong(s, t2, src); l2 > l {
+					t = t2
+					l = l2
+				}
+			}
+		}
+
 		// Extend backwards
 		for t > 0 && s > nextEmit && src[t-1] == src[s-1] {
 			s--
diff --git a/vendor/github.com/klauspost/compress/flate/level6.go b/vendor/github.com/klauspost/compress/flate/level6.go
index a52c80ea45..a709977ec4 100644
--- a/vendor/github.com/klauspost/compress/flate/level6.go
+++ b/vendor/github.com/klauspost/compress/flate/level6.go
@@ -211,6 +211,31 @@ func (e *fastEncL6) Encode(dst *tokens, src []byte) {
 			l += e.matchlenLong(s+l, t+l, src)
 		}
 
+		// Try to locate a better match by checking the end-of-match...
+		if sAt := s + l; sAt < sLimit {
+			eLong := &e.bTable[hash7(load6432(src, sAt), tableBits)]
+			// Test current
+			t2 := eLong.Cur.offset - e.cur - l
+			off := s - t2
+			if off < maxMatchOffset {
+				if off > 0 && t2 >= 0 {
+					if l2 := e.matchlenLong(s, t2, src); l2 > l {
+						t = t2
+						l = l2
+					}
+				}
+				// Test next:
+				t2 = eLong.Prev.offset - e.cur - l
+				off := s - t2
+				if off > 0 && off < maxMatchOffset && t2 >= 0 {
+					if l2 := e.matchlenLong(s, t2, src); l2 > l {
+						t = t2
+						l = l2
+					}
+				}
+			}
+		}
+
 		// Extend backwards
 		for t > 0 && s > nextEmit && src[t-1] == src[s-1] {
 			s--
diff --git a/vendor/github.com/klauspost/compress/flate/token.go b/vendor/github.com/klauspost/compress/flate/token.go
index f9abf606d6..eb862d7a92 100644
--- a/vendor/github.com/klauspost/compress/flate/token.go
+++ b/vendor/github.com/klauspost/compress/flate/token.go
@@ -13,14 +13,17 @@ import (
 )
 
 const (
+	// From top
 	// 2 bits:   type   0 = literal  1=EOF  2=Match   3=Unused
 	// 8 bits:   xlength = length - MIN_MATCH_LENGTH
-	// 22 bits   xoffset = offset - MIN_OFFSET_SIZE, or literal
-	lengthShift = 22
-	offsetMask  = 1<<lengthShift - 1
-	typeMask    = 3 << 30
-	literalType = 0 << 30
-	matchType   = 1 << 30
+	// 5 bits    offsetcode
+	// 16 bits   xoffset = offset - MIN_OFFSET_SIZE, or literal
+	lengthShift         = 22
+	offsetMask          = 1<<lengthShift - 1
+	typeMask            = 3 << 30
+	literalType         = 0 << 30
+	matchType           = 1 << 30
+	matchOffsetOnlyMask = 0xffff
 )
 
 // The length code for length X (MIN_MATCH_LENGTH <= X <= MAX_MATCH_LENGTH)
@@ -187,7 +190,7 @@ func (t *tokens) indexTokens(in []token) {
 			t.AddLiteral(tok.literal())
 			continue
 		}
-		t.AddMatch(uint32(tok.length()), tok.offset())
+		t.AddMatch(uint32(tok.length()), tok.offset()&matchOffsetOnlyMask)
 	}
 }
 
@@ -232,7 +235,7 @@ func (t *tokens) EstimatedBits() int {
 		for _, v := range t.litHist[:] {
 			if v > 0 {
 				n := float32(v)
-				shannon += -mFastLog2(n*invTotal) * n
+				shannon += atLeastOne(-mFastLog2(n*invTotal)) * n
 			}
 		}
 		// Just add 15 for EOB
@@ -240,7 +243,7 @@ func (t *tokens) EstimatedBits() int {
 		for i, v := range t.extraHist[1 : literalCount-256] {
 			if v > 0 {
 				n := float32(v)
-				shannon += -mFastLog2(n*invTotal) * n
+				shannon += atLeastOne(-mFastLog2(n*invTotal)) * n
 				bits += int(lengthExtraBits[i&31]) * int(v)
 				nMatches += int(v)
 			}
@@ -251,7 +254,7 @@ func (t *tokens) EstimatedBits() int {
 		for i, v := range t.offHist[:offsetCodeCount] {
 			if v > 0 {
 				n := float32(v)
-				shannon += -mFastLog2(n*invTotal) * n
+				shannon += atLeastOne(-mFastLog2(n*invTotal)) * n
 				bits += int(offsetExtraBits[i&31]) * int(v)
 			}
 		}
@@ -270,11 +273,13 @@ func (t *tokens) AddMatch(xlength uint32, xoffset uint32) {
 			panic(fmt.Errorf("invalid offset: %v", xoffset))
 		}
 	}
+	oCode := offsetCode(xoffset)
+	xoffset |= oCode << 16
 	t.nLits++
-	lengthCode := lengthCodes1[uint8(xlength)] & 31
+
+	t.extraHist[lengthCodes1[uint8(xlength)]]++
+	t.offHist[oCode]++
 	t.tokens[t.n] = token(matchType | xlength<<lengthShift | xoffset)
-	t.extraHist[lengthCode]++
-	t.offHist[offsetCode(xoffset)&31]++
 	t.n++
 }
 
@@ -286,7 +291,8 @@ func (t *tokens) AddMatchLong(xlength int32, xoffset uint32) {
 			panic(fmt.Errorf("invalid offset: %v", xoffset))
 		}
 	}
-	oc := offsetCode(xoffset) & 31
+	oc := offsetCode(xoffset)
+	xoffset |= oc << 16
 	for xlength > 0 {
 		xl := xlength
 		if xl > 258 {
@@ -294,12 +300,11 @@ func (t *tokens) AddMatchLong(xlength int32, xoffset uint32) {
 			xl = 258 - baseMatchLength
 		}
 		xlength -= xl
-		xl -= 3
+		xl -= baseMatchLength
 		t.nLits++
-		lengthCode := lengthCodes1[uint8(xl)] & 31
-		t.tokens[t.n] = token(matchType | uint32(xl)<<lengthShift | xoffset)
-		t.extraHist[lengthCode]++
+		t.extraHist[lengthCodes1[uint8(xl)]]++
 		t.offHist[oc]++
+		t.tokens[t.n] = token(matchType | uint32(xl)<<lengthShift | xoffset)
 		t.n++
 	}
 }
diff --git a/vendor/github.com/klauspost/compress/gzip/gunzip.go b/vendor/github.com/klauspost/compress/gzip/gunzip.go
index 568b5d4fb8..21e768b360 100644
--- a/vendor/github.com/klauspost/compress/gzip/gunzip.go
+++ b/vendor/github.com/klauspost/compress/gzip/gunzip.go
@@ -75,6 +75,7 @@ type Header struct {
 type Reader struct {
 	Header       // valid after NewReader or Reader.Reset
 	r            flate.Reader
+	br           *bufio.Reader
 	decompressor io.ReadCloser
 	digest       uint32 // CRC-32, IEEE polynomial (section 8)
 	size         uint32 // Uncompressed size (section 2.3.1)
@@ -109,7 +110,13 @@ func (z *Reader) Reset(r io.Reader) error {
 	if rr, ok := r.(flate.Reader); ok {
 		z.r = rr
 	} else {
-		z.r = bufio.NewReader(r)
+		// Reuse if we can.
+		if z.br != nil {
+			z.br.Reset(r)
+		} else {
+			z.br = bufio.NewReader(r)
+		}
+		z.r = z.br
 	}
 	z.Header, z.err = z.readHeader()
 	return z.err
diff --git a/vendor/github.com/klauspost/compress/zstd/README.md b/vendor/github.com/klauspost/compress/zstd/README.md
index 7680bfe1dd..787813fa9e 100644
--- a/vendor/github.com/klauspost/compress/zstd/README.md
+++ b/vendor/github.com/klauspost/compress/zstd/README.md
@@ -16,8 +16,7 @@ Currently the package is heavily optimized for 64 bit processors and will be sig
 
 Install using `go get -u github.com/klauspost/compress`. The package is located in `github.com/klauspost/compress/zstd`.
 
-Godoc Documentation: https://godoc.org/github.com/klauspost/compress/zstd
-
+[![Go Reference](https://pkg.go.dev/badge/github.com/klauspost/compress/zstd.svg)](https://pkg.go.dev/github.com/klauspost/compress/zstd)
 
 ## Compressor
 
@@ -152,8 +151,8 @@ This package:
 file    out     level   insize      outsize     millis  mb/s
 silesia.tar zskp    1   211947520   73101992    643     313.87
 silesia.tar zskp    2   211947520   67504318    969     208.38
-silesia.tar zskp    3   211947520   65177448    1899    106.44
-silesia.tar zskp    4   211947520   61381950    8115    24.91
+silesia.tar zskp    3   211947520   64595893    2007    100.68
+silesia.tar zskp    4   211947520   60995370    7691    26.28
 
 cgo zstd:
 silesia.tar zstd    1   211947520   73605392    543     371.56
@@ -171,8 +170,8 @@ https://files.klauspost.com/compress/gob-stream.7z
 file        out     level   insize  outsize     millis  mb/s
 gob-stream  zskp    1   1911399616  235022249   3088    590.30
 gob-stream  zskp    2   1911399616  205669791   3786    481.34
-gob-stream  zskp    3   1911399616  185792019   9324    195.48
-gob-stream  zskp    4   1911399616  171537212   32113   56.76
+gob-stream  zskp    3   1911399616  175034659   9636    189.17
+gob-stream  zskp    4   1911399616  167273881   29337   62.13
 gob-stream  zstd    1   1911399616  249810424   2637    691.26
 gob-stream  zstd    3   1911399616  208192146   3490    522.31
 gob-stream  zstd    6   1911399616  193632038   6687    272.56
@@ -187,8 +186,8 @@ http://mattmahoney.net/dc/textdata.html
 file    out level   insize      outsize     millis  mb/s
 enwik9  zskp    1   1000000000  343848582   3609    264.18
 enwik9  zskp    2   1000000000  317276632   5746    165.97
-enwik9  zskp    3   1000000000  294540704   11725   81.34
-enwik9  zskp    4   1000000000  276609671   44029   21.66
+enwik9  zskp    3   1000000000  292243069   12162   78.41
+enwik9  zskp    4   1000000000  275241169   36430   26.18
 enwik9  zstd    1   1000000000  358072021   3110    306.65
 enwik9  zstd    3   1000000000  313734672   4784    199.35
 enwik9  zstd    6   1000000000  295138875   10290   92.68
@@ -202,8 +201,8 @@ https://files.klauspost.com/compress/github-june-2days-2019.json.zst
 file                        out level   insize      outsize     millis  mb/s
 github-june-2days-2019.json zskp    1   6273951764  699045015   10620   563.40
 github-june-2days-2019.json zskp    2   6273951764  617881763   11687   511.96
-github-june-2days-2019.json zskp    3   6273951764  537511906   29252   204.54
-github-june-2days-2019.json zskp    4   6273951764  512796117   97791   61.18
+github-june-2days-2019.json zskp    3   6273951764  524340691   34043   175.75
+github-june-2days-2019.json zskp    4   6273951764  503314661   93811   63.78
 github-june-2days-2019.json zstd    1   6273951764  766284037   8450    708.00
 github-june-2days-2019.json zstd    3   6273951764  661889476   10927   547.57
 github-june-2days-2019.json zstd    6   6273951764  642756859   22996   260.18
@@ -217,8 +216,8 @@ https://files.klauspost.com/compress/rawstudio-mint14.7z
 file                    out level   insize      outsize     millis  mb/s
 rawstudio-mint14.tar    zskp    1   8558382592  3667489370  20210   403.84
 rawstudio-mint14.tar    zskp    2   8558382592  3364592300  31873   256.07
-rawstudio-mint14.tar    zskp    3   8558382592  3224594213  71751   113.75
-rawstudio-mint14.tar    zskp    4   8558382592  3027332295  486243  16.79
+rawstudio-mint14.tar    zskp    3   8558382592  3158085214  77675   105.08
+rawstudio-mint14.tar    zskp    4   8558382592  3020370044  404956  20.16
 rawstudio-mint14.tar    zstd    1   8558382592  3609250104  17136   476.27
 rawstudio-mint14.tar    zstd    3   8558382592  3341679997  29262   278.92
 rawstudio-mint14.tar    zstd    6   8558382592  3235846406  77904   104.77
@@ -232,8 +231,8 @@ https://files.klauspost.com/compress/nyc-taxi-data-10M.csv.zst
 file                    out level   insize      outsize     millis  mb/s
 nyc-taxi-data-10M.csv   zskp    1   3325605752  641339945   8925    355.35
 nyc-taxi-data-10M.csv   zskp    2   3325605752  591748091   11268   281.44
-nyc-taxi-data-10M.csv   zskp    3   3325605752  538490114   19880   159.53
-nyc-taxi-data-10M.csv   zskp    4   3325605752  495986829   89368   35.49
+nyc-taxi-data-10M.csv   zskp    3   3325605752  530289687   25239   125.66
+nyc-taxi-data-10M.csv   zskp    4   3325605752  490907191   65939   48.10
 nyc-taxi-data-10M.csv   zstd    1   3325605752  687399637   8233    385.18
 nyc-taxi-data-10M.csv   zstd    3   3325605752  598514411   10065   315.07
 nyc-taxi-data-10M.csv   zstd    6   3325605752  570522953   20038   158.27
@@ -405,13 +404,28 @@ BenchmarkDecoder_DecodeAllParallelCgo/comp-data.bin.zst-16        749938
 
 This reflects the performance around May 2020, but this may be out of date.
 
+## Zstd inside ZIP files
+
+It is possible to use zstandard to compress individual files inside zip archives.
+While this isn't widely supported it can be useful for internal files.
+
+To support the compression and decompression of these files you must register a compressor and decompressor.
+
+It is highly recommended registering the (de)compressors on individual zip Reader/Writer and NOT
+use the global registration functions. The main reason for this is that 2 registrations from 
+different packages will result in a panic.
+
+It is a good idea to only have a single compressor and decompressor, since they can be used for multiple zip
+files concurrently, and using a single instance will allow reusing some resources.
+
+See [this example](https://pkg.go.dev/github.com/klauspost/compress/zstd#example-ZipCompressor) for 
+how to compress and decompress files inside zip archives.
+
 # Contributions
 
 Contributions are always welcome. 
 For new features/fixes, remember to add tests and for performance enhancements include benchmarks.
 
-For sending files for reproducing errors use a service like [goobox](https://goobox.io/#/upload) or similar to share your files.
-
 For general feedback and experience reports, feel free to open an issue or write me on [Twitter](https://twitter.com/sh0dan).
 
 This package includes the excellent [`github.com/cespare/xxhash`](https://github.com/cespare/xxhash) package Copyright (c) 2016 Caleb Spare.
diff --git a/vendor/github.com/klauspost/compress/zstd/blockdec.go b/vendor/github.com/klauspost/compress/zstd/blockdec.go
index b51d922bda..e30af505ca 100644
--- a/vendor/github.com/klauspost/compress/zstd/blockdec.go
+++ b/vendor/github.com/klauspost/compress/zstd/blockdec.go
@@ -123,12 +123,10 @@ func newBlockDec(lowMem bool) *blockDec {
 // Input must be a start of a block and will be at the end of the block when returned.
 func (b *blockDec) reset(br byteBuffer, windowSize uint64) error {
 	b.WindowSize = windowSize
-	tmp := br.readSmall(3)
-	if tmp == nil {
-		if debug {
-			println("Reading block header:", io.ErrUnexpectedEOF)
-		}
-		return io.ErrUnexpectedEOF
+	tmp, err := br.readSmall(3)
+	if err != nil {
+		println("Reading block header:", err)
+		return err
 	}
 	bh := uint32(tmp[0]) | (uint32(tmp[1]) << 8) | (uint32(tmp[2]) << 16)
 	b.Last = bh&1 != 0
@@ -146,7 +144,7 @@ func (b *blockDec) reset(br byteBuffer, windowSize uint64) error {
 		}
 		cSize = 1
 	case blockTypeCompressed:
-		if debug {
+		if debugDecoder {
 			println("Data size on stream:", cSize)
 		}
 		b.RLESize = 0
@@ -155,7 +153,7 @@ func (b *blockDec) reset(br byteBuffer, windowSize uint64) error {
 			maxSize = int(windowSize)
 		}
 		if cSize > maxCompressedBlockSize || uint64(cSize) > b.WindowSize {
-			if debug {
+			if debugDecoder {
 				printf("compressed block too big: csize:%d block: %+v\n", uint64(cSize), b)
 			}
 			return ErrCompressedSizeTooBig
@@ -179,10 +177,9 @@ func (b *blockDec) reset(br byteBuffer, windowSize uint64) error {
 	if cap(b.dst) <= maxSize {
 		b.dst = make([]byte, 0, maxSize+1)
 	}
-	var err error
 	b.data, err = br.readBig(cSize, b.dataStorage)
 	if err != nil {
-		if debug {
+		if debugDecoder {
 			println("Reading block:", err, "(", cSize, ")", len(b.data))
 			printf("%T", br)
 		}
@@ -252,7 +249,7 @@ func (b *blockDec) startDecoder() {
 				b:   b.dst,
 				err: err,
 			}
-			if debug {
+			if debugDecoder {
 				println("Decompressed to", len(b.dst), "bytes, error:", err)
 			}
 			b.result <- o
@@ -267,7 +264,7 @@ func (b *blockDec) startDecoder() {
 		default:
 			panic("Invalid block type")
 		}
-		if debug {
+		if debugDecoder {
 			println("blockDec: Finished block")
 		}
 	}
@@ -300,7 +297,7 @@ func (b *blockDec) decodeBuf(hist *history) error {
 		b.dst = hist.b
 		hist.b = nil
 		err := b.decodeCompressed(hist)
-		if debug {
+		if debugDecoder {
 			println("Decompressed to total", len(b.dst), "bytes, hash:", xxhash.Sum64(b.dst), "error:", err)
 		}
 		hist.b = b.dst
@@ -393,7 +390,7 @@ func (b *blockDec) decodeCompressed(hist *history) error {
 			in = in[5:]
 		}
 	}
-	if debug {
+	if debugDecoder {
 		println("literals type:", litType, "litRegenSize:", litRegenSize, "litCompSize:", litCompSize, "sizeFormat:", sizeFormat, "4X:", fourStreams)
 	}
 	var literals []byte
@@ -431,7 +428,7 @@ func (b *blockDec) decodeCompressed(hist *history) error {
 			literals[i] = v
 		}
 		in = in[1:]
-		if debug {
+		if debugDecoder {
 			printf("Found %d RLE compressed literals\n", litRegenSize)
 		}
 	case literalsBlockTreeless:
@@ -442,7 +439,7 @@ func (b *blockDec) decodeCompressed(hist *history) error {
 		// Store compressed literals, so we defer decoding until we get history.
 		literals = in[:litCompSize]
 		in = in[litCompSize:]
-		if debug {
+		if debugDecoder {
 			printf("Found %d compressed literals\n", litCompSize)
 		}
 	case literalsBlockCompressed:
@@ -484,7 +481,7 @@ func (b *blockDec) decodeCompressed(hist *history) error {
 		if len(literals) != litRegenSize {
 			return fmt.Errorf("literal output size mismatch want %d, got %d", litRegenSize, len(literals))
 		}
-		if debug {
+		if debugDecoder {
 			printf("Decompressed %d literals into %d bytes\n", litCompSize, litRegenSize)
 		}
 	}
@@ -535,12 +532,12 @@ func (b *blockDec) decodeCompressed(hist *history) error {
 		br := byteReader{b: in, off: 0}
 		compMode := br.Uint8()
 		br.advance(1)
-		if debug {
+		if debugDecoder {
 			printf("Compression modes: 0b%b", compMode)
 		}
 		for i := uint(0); i < 3; i++ {
 			mode := seqCompMode((compMode >> (6 - i*2)) & 3)
-			if debug {
+			if debugDecoder {
 				println("Table", tableIndex(i), "is", mode)
 			}
 			var seq *sequenceDec
@@ -571,7 +568,7 @@ func (b *blockDec) decodeCompressed(hist *history) error {
 				}
 				dec.setRLE(symb)
 				seq.fse = dec
-				if debug {
+				if debugDecoder {
 					printf("RLE set to %+v, code: %v", symb, v)
 				}
 			case compModeFSE:
@@ -587,7 +584,7 @@ func (b *blockDec) decodeCompressed(hist *history) error {
 					println("Transform table error:", err)
 					return err
 				}
-				if debug {
+				if debugDecoder {
 					println("Read table ok", "symbolLen:", dec.symbolLen)
 				}
 				seq.fse = dec
@@ -655,7 +652,7 @@ func (b *blockDec) decodeCompressed(hist *history) error {
 	if huff != nil {
 		hist.huffTree = huff
 	}
-	if debug {
+	if debugDecoder {
 		println("Final literals:", len(literals), "hash:", xxhash.Sum64(literals), "and", nSeqs, "sequences.")
 	}
 
@@ -672,7 +669,7 @@ func (b *blockDec) decodeCompressed(hist *history) error {
 	if err != nil {
 		return err
 	}
-	if debug {
+	if debugDecoder {
 		println("History merged ok")
 	}
 	br := &bitReader{}
@@ -731,7 +728,7 @@ func (b *blockDec) decodeCompressed(hist *history) error {
 	}
 	hist.append(b.dst)
 	hist.recentOffsets = seqs.prevOffset
-	if debug {
+	if debugDecoder {
 		println("Finished block with literals:", len(literals), "and", nSeqs, "sequences.")
 	}
 
diff --git a/vendor/github.com/klauspost/compress/zstd/blockenc.go b/vendor/github.com/klauspost/compress/zstd/blockenc.go
index e1be092f32..3df185ee46 100644
--- a/vendor/github.com/klauspost/compress/zstd/blockenc.go
+++ b/vendor/github.com/klauspost/compress/zstd/blockenc.go
@@ -156,7 +156,7 @@ func (h *literalsHeader) setSize(regenLen int) {
 	switch {
 	case inBits < 5:
 		lh |= (uint64(regenLen) << 3) | (1 << 60)
-		if debug {
+		if debugEncoder {
 			got := int(lh>>3) & 0xff
 			if got != regenLen {
 				panic(fmt.Sprint("litRegenSize = ", regenLen, "(want) != ", got, "(got)"))
@@ -184,7 +184,7 @@ func (h *literalsHeader) setSizes(compLen, inLen int, single bool) {
 			lh |= 1 << 2
 		}
 		lh |= (uint64(inLen) << 4) | (uint64(compLen) << (10 + 4)) | (3 << 60)
-		if debug {
+		if debugEncoder {
 			const mmask = (1 << 24) - 1
 			n := (lh >> 4) & mmask
 			if int(n&1023) != inLen {
@@ -312,7 +312,7 @@ func (b *blockEnc) encodeRaw(a []byte) {
 	bh.setType(blockTypeRaw)
 	b.output = bh.appendTo(b.output[:0])
 	b.output = append(b.output, a...)
-	if debug {
+	if debugEncoder {
 		println("Adding RAW block, length", len(a), "last:", b.last)
 	}
 }
@@ -325,7 +325,7 @@ func (b *blockEnc) encodeRawTo(dst, src []byte) []byte {
 	bh.setType(blockTypeRaw)
 	dst = bh.appendTo(dst)
 	dst = append(dst, src...)
-	if debug {
+	if debugEncoder {
 		println("Adding RAW block, length", len(src), "last:", b.last)
 	}
 	return dst
@@ -339,7 +339,7 @@ func (b *blockEnc) encodeLits(lits []byte, raw bool) error {
 
 	// Don't compress extremely small blocks
 	if len(lits) < 8 || (len(lits) < 32 && b.dictLitEnc == nil) || raw {
-		if debug {
+		if debugEncoder {
 			println("Adding RAW block, length", len(lits), "last:", b.last)
 		}
 		bh.setType(blockTypeRaw)
@@ -371,7 +371,7 @@ func (b *blockEnc) encodeLits(lits []byte, raw bool) error {
 
 	switch err {
 	case huff0.ErrIncompressible:
-		if debug {
+		if debugEncoder {
 			println("Adding RAW block, length", len(lits), "last:", b.last)
 		}
 		bh.setType(blockTypeRaw)
@@ -379,7 +379,7 @@ func (b *blockEnc) encodeLits(lits []byte, raw bool) error {
 		b.output = append(b.output, lits...)
 		return nil
 	case huff0.ErrUseRLE:
-		if debug {
+		if debugEncoder {
 			println("Adding RLE block, length", len(lits))
 		}
 		bh.setType(blockTypeRLE)
@@ -396,12 +396,12 @@ func (b *blockEnc) encodeLits(lits []byte, raw bool) error {
 	bh.setType(blockTypeCompressed)
 	var lh literalsHeader
 	if reUsed {
-		if debug {
+		if debugEncoder {
 			println("Reused tree, compressed to", len(out))
 		}
 		lh.setType(literalsBlockTreeless)
 	} else {
-		if debug {
+		if debugEncoder {
 			println("New tree, compressed to", len(out), "tree size:", len(b.litEnc.OutTable))
 		}
 		lh.setType(literalsBlockCompressed)
@@ -517,7 +517,7 @@ func (b *blockEnc) encode(org []byte, raw, rawAllLits bool) error {
 		lh.setSize(len(b.literals))
 		b.output = lh.appendTo(b.output)
 		b.output = append(b.output, b.literals...)
-		if debug {
+		if debugEncoder {
 			println("Adding literals RAW, length", len(b.literals))
 		}
 	case huff0.ErrUseRLE:
@@ -525,22 +525,22 @@ func (b *blockEnc) encode(org []byte, raw, rawAllLits bool) error {
 		lh.setSize(len(b.literals))
 		b.output = lh.appendTo(b.output)
 		b.output = append(b.output, b.literals[0])
-		if debug {
+		if debugEncoder {
 			println("Adding literals RLE")
 		}
 	case nil:
 		// Compressed litLen...
 		if reUsed {
-			if debug {
+			if debugEncoder {
 				println("reused tree")
 			}
 			lh.setType(literalsBlockTreeless)
 		} else {
-			if debug {
+			if debugEncoder {
 				println("new tree, size:", len(b.litEnc.OutTable))
 			}
 			lh.setType(literalsBlockCompressed)
-			if debug {
+			if debugEncoder {
 				_, _, err := huff0.ReadTable(out, nil)
 				if err != nil {
 					panic(err)
@@ -548,18 +548,18 @@ func (b *blockEnc) encode(org []byte, raw, rawAllLits bool) error {
 			}
 		}
 		lh.setSizes(len(out), len(b.literals), single)
-		if debug {
+		if debugEncoder {
 			printf("Compressed %d literals to %d bytes", len(b.literals), len(out))
 			println("Adding literal header:", lh)
 		}
 		b.output = lh.appendTo(b.output)
 		b.output = append(b.output, out...)
 		b.litEnc.Reuse = huff0.ReusePolicyAllow
-		if debug {
+		if debugEncoder {
 			println("Adding literals compressed")
 		}
 	default:
-		if debug {
+		if debugEncoder {
 			println("Adding literals ERROR:", err)
 		}
 		return err
@@ -577,7 +577,7 @@ func (b *blockEnc) encode(org []byte, raw, rawAllLits bool) error {
 		n := len(b.sequences) - 0x7f00
 		b.output = append(b.output, 255, uint8(n), uint8(n>>8))
 	}
-	if debug {
+	if debugEncoder {
 		println("Encoding", len(b.sequences), "sequences")
 	}
 	b.genCodes()
@@ -611,17 +611,17 @@ func (b *blockEnc) encode(org []byte, raw, rawAllLits bool) error {
 		nSize = nSize + (nSize+2*8*16)>>4
 		switch {
 		case predefSize <= prevSize && predefSize <= nSize || forcePreDef:
-			if debug {
+			if debugEncoder {
 				println("Using predefined", predefSize>>3, "<=", nSize>>3)
 			}
 			return preDef, compModePredefined
 		case prevSize <= nSize:
-			if debug {
+			if debugEncoder {
 				println("Using previous", prevSize>>3, "<=", nSize>>3)
 			}
 			return prev, compModeRepeat
 		default:
-			if debug {
+			if debugEncoder {
 				println("Using new, predef", predefSize>>3, ". previous:", prevSize>>3, ">", nSize>>3, "header max:", cur.maxHeaderSize()>>3, "bytes")
 				println("tl:", cur.actualTableLog, "symbolLen:", cur.symbolLen, "norm:", cur.norm[:cur.symbolLen], "hist", cur.count[:cur.symbolLen])
 			}
@@ -634,7 +634,7 @@ func (b *blockEnc) encode(org []byte, raw, rawAllLits bool) error {
 	if llEnc.useRLE {
 		mode |= uint8(compModeRLE) << 6
 		llEnc.setRLE(b.sequences[0].llCode)
-		if debug {
+		if debugEncoder {
 			println("llEnc.useRLE")
 		}
 	} else {
@@ -645,7 +645,7 @@ func (b *blockEnc) encode(org []byte, raw, rawAllLits bool) error {
 	if ofEnc.useRLE {
 		mode |= uint8(compModeRLE) << 4
 		ofEnc.setRLE(b.sequences[0].ofCode)
-		if debug {
+		if debugEncoder {
 			println("ofEnc.useRLE")
 		}
 	} else {
@@ -657,7 +657,7 @@ func (b *blockEnc) encode(org []byte, raw, rawAllLits bool) error {
 	if mlEnc.useRLE {
 		mode |= uint8(compModeRLE) << 2
 		mlEnc.setRLE(b.sequences[0].mlCode)
-		if debug {
+		if debugEncoder {
 			println("mlEnc.useRLE, code: ", b.sequences[0].mlCode, "value", b.sequences[0].matchLen)
 		}
 	} else {
@@ -666,7 +666,7 @@ func (b *blockEnc) encode(org []byte, raw, rawAllLits bool) error {
 		mode |= uint8(m) << 2
 	}
 	b.output = append(b.output, mode)
-	if debug {
+	if debugEncoder {
 		printf("Compression modes: 0b%b", mode)
 	}
 	b.output, err = llEnc.writeCount(b.output)
@@ -786,7 +786,7 @@ func (b *blockEnc) encode(org []byte, raw, rawAllLits bool) error {
 
 	// Size is output minus block header.
 	bh.setSize(uint32(len(b.output)-bhOffset) - 3)
-	if debug {
+	if debugEncoder {
 		println("Rewriting block header", bh)
 	}
 	_ = bh.appendTo(b.output[bhOffset:bhOffset])
diff --git a/vendor/github.com/klauspost/compress/zstd/bytebuf.go b/vendor/github.com/klauspost/compress/zstd/bytebuf.go
index 658ef78380..aab71c6cf8 100644
--- a/vendor/github.com/klauspost/compress/zstd/bytebuf.go
+++ b/vendor/github.com/klauspost/compress/zstd/bytebuf.go
@@ -12,8 +12,8 @@ import (
 
 type byteBuffer interface {
 	// Read up to 8 bytes.
-	// Returns nil if no more input is available.
-	readSmall(n int) []byte
+	// Returns io.ErrUnexpectedEOF if this cannot be satisfied.
+	readSmall(n int) ([]byte, error)
 
 	// Read >8 bytes.
 	// MAY use the destination slice.
@@ -29,17 +29,17 @@ type byteBuffer interface {
 // in-memory buffer
 type byteBuf []byte
 
-func (b *byteBuf) readSmall(n int) []byte {
+func (b *byteBuf) readSmall(n int) ([]byte, error) {
 	if debugAsserts && n > 8 {
 		panic(fmt.Errorf("small read > 8 (%d). use readBig", n))
 	}
 	bb := *b
 	if len(bb) < n {
-		return nil
+		return nil, io.ErrUnexpectedEOF
 	}
 	r := bb[:n]
 	*b = bb[n:]
-	return r
+	return r, nil
 }
 
 func (b *byteBuf) readBig(n int, dst []byte) ([]byte, error) {
@@ -81,19 +81,22 @@ type readerWrapper struct {
 	tmp [8]byte
 }
 
-func (r *readerWrapper) readSmall(n int) []byte {
+func (r *readerWrapper) readSmall(n int) ([]byte, error) {
 	if debugAsserts && n > 8 {
 		panic(fmt.Errorf("small read > 8 (%d). use readBig", n))
 	}
 	n2, err := io.ReadFull(r.r, r.tmp[:n])
 	// We only really care about the actual bytes read.
-	if n2 != n {
-		if debug {
+	if err != nil {
+		if err == io.EOF {
+			return nil, io.ErrUnexpectedEOF
+		}
+		if debugDecoder {
 			println("readSmall: got", n2, "want", n, "err", err)
 		}
-		return nil
+		return nil, err
 	}
-	return r.tmp[:n]
+	return r.tmp[:n], nil
 }
 
 func (r *readerWrapper) readBig(n int, dst []byte) ([]byte, error) {
diff --git a/vendor/github.com/klauspost/compress/zstd/decoder.go b/vendor/github.com/klauspost/compress/zstd/decoder.go
index f593e464b6..4d984c3b26 100644
--- a/vendor/github.com/klauspost/compress/zstd/decoder.go
+++ b/vendor/github.com/klauspost/compress/zstd/decoder.go
@@ -113,9 +113,6 @@ func NewReader(r io.Reader, opts ...DOption) (*Decoder, error) {
 // Returns the number of bytes written and any error that occurred.
 // When the stream is done, io.EOF will be returned.
 func (d *Decoder) Read(p []byte) (int, error) {
-	if d.stream == nil {
-		return 0, ErrDecoderNilInput
-	}
 	var n int
 	for {
 		if len(d.current.b) > 0 {
@@ -138,7 +135,7 @@ func (d *Decoder) Read(p []byte) (int, error) {
 		}
 	}
 	if len(d.current.b) > 0 {
-		if debug {
+		if debugDecoder {
 			println("returning", n, "still bytes left:", len(d.current.b))
 		}
 		// Only return error at end of block
@@ -147,7 +144,7 @@ func (d *Decoder) Read(p []byte) (int, error) {
 	if d.current.err != nil {
 		d.drainOutput()
 	}
-	if debug {
+	if debugDecoder {
 		println("returning", n, d.current.err, len(d.decoders))
 	}
 	return n, d.current.err
@@ -167,20 +164,17 @@ func (d *Decoder) Reset(r io.Reader) error {
 
 	if r == nil {
 		d.current.err = ErrDecoderNilInput
+		if len(d.current.b) > 0 {
+			d.current.b = d.current.b[:0]
+		}
 		d.current.flushed = true
 		return nil
 	}
 
-	if d.stream == nil {
-		d.stream = make(chan decodeStream, 1)
-		d.streamWg.Add(1)
-		go d.startStreamDecoder(d.stream)
-	}
-
-	// If bytes buffer and < 1MB, do sync decoding anyway.
-	if bb, ok := r.(byter); ok && bb.Len() < 1<<20 {
+	// If bytes buffer and < 5MB, do sync decoding anyway.
+	if bb, ok := r.(byter); ok && bb.Len() < 5<<20 {
 		bb2 := bb
-		if debug {
+		if debugDecoder {
 			println("*bytes.Buffer detected, doing sync decode, len:", bb.Len())
 		}
 		b := bb2.Bytes()
@@ -196,12 +190,18 @@ func (d *Decoder) Reset(r io.Reader) error {
 		d.current.b = dst
 		d.current.err = err
 		d.current.flushed = true
-		if debug {
+		if debugDecoder {
 			println("sync decode to", len(dst), "bytes, err:", err)
 		}
 		return nil
 	}
 
+	if d.stream == nil {
+		d.stream = make(chan decodeStream, 1)
+		d.streamWg.Add(1)
+		go d.startStreamDecoder(d.stream)
+	}
+
 	// Remove current block.
 	d.current.decodeOutput = decodeOutput{}
 	d.current.err = nil
@@ -225,7 +225,7 @@ func (d *Decoder) drainOutput() {
 		d.current.cancel = nil
 	}
 	if d.current.d != nil {
-		if debug {
+		if debugDecoder {
 			printf("re-adding current decoder %p, decoders: %d", d.current.d, len(d.decoders))
 		}
 		d.decoders <- d.current.d
@@ -238,7 +238,7 @@ func (d *Decoder) drainOutput() {
 	}
 	for v := range d.current.output {
 		if v.d != nil {
-			if debug {
+			if debugDecoder {
 				printf("re-adding decoder %p", v.d)
 			}
 			d.decoders <- v.d
@@ -255,9 +255,6 @@ func (d *Decoder) drainOutput() {
 // The return value n is the number of bytes written.
 // Any error encountered during the write is also returned.
 func (d *Decoder) WriteTo(w io.Writer) (int64, error) {
-	if d.stream == nil {
-		return 0, ErrDecoderNilInput
-	}
 	var n int64
 	for {
 		if len(d.current.b) > 0 {
@@ -297,7 +294,7 @@ func (d *Decoder) DecodeAll(input, dst []byte) ([]byte, error) {
 	block := <-d.decoders
 	frame := block.localFrame
 	defer func() {
-		if debug {
+		if debugDecoder {
 			printf("re-adding decoder: %p", block)
 		}
 		frame.rawInput = nil
@@ -310,7 +307,7 @@ func (d *Decoder) DecodeAll(input, dst []byte) ([]byte, error) {
 		frame.history.reset()
 		err := frame.reset(&frame.bBuf)
 		if err == io.EOF {
-			if debug {
+			if debugDecoder {
 				println("frame reset return EOF")
 			}
 			return dst, nil
@@ -355,7 +352,7 @@ func (d *Decoder) DecodeAll(input, dst []byte) ([]byte, error) {
 			return dst, err
 		}
 		if len(frame.bBuf) == 0 {
-			if debug {
+			if debugDecoder {
 				println("frame dbuf empty")
 			}
 			break
@@ -371,7 +368,7 @@ func (d *Decoder) DecodeAll(input, dst []byte) ([]byte, error) {
 // if no data was available without blocking.
 func (d *Decoder) nextBlock(blocking bool) (ok bool) {
 	if d.current.d != nil {
-		if debug {
+		if debugDecoder {
 			printf("re-adding current decoder %p", d.current.d)
 		}
 		d.decoders <- d.current.d
@@ -391,7 +388,7 @@ func (d *Decoder) nextBlock(blocking bool) (ok bool) {
 			return false
 		}
 	}
-	if debug {
+	if debugDecoder {
 		println("got", len(d.current.b), "bytes, error:", d.current.err)
 	}
 	return true
@@ -485,7 +482,7 @@ func (d *Decoder) startStreamDecoder(inStream chan decodeStream) {
 	defer d.streamWg.Done()
 	frame := newFrameDec(d.o)
 	for stream := range inStream {
-		if debug {
+		if debugDecoder {
 			println("got new stream")
 		}
 		br := readerWrapper{r: stream.r}
@@ -493,7 +490,7 @@ func (d *Decoder) startStreamDecoder(inStream chan decodeStream) {
 		for {
 			frame.history.reset()
 			err := frame.reset(&br)
-			if debug && err != nil {
+			if debugDecoder && err != nil {
 				println("Frame decoder returned", err)
 			}
 			if err == nil && frame.DictionaryID != nil {
@@ -510,7 +507,7 @@ func (d *Decoder) startStreamDecoder(inStream chan decodeStream) {
 				}
 				break
 			}
-			if debug {
+			if debugDecoder {
 				println("starting frame decoder")
 			}
 
diff --git a/vendor/github.com/klauspost/compress/zstd/dict.go b/vendor/github.com/klauspost/compress/zstd/dict.go
index fa25a18d86..a36ae83ef5 100644
--- a/vendor/github.com/klauspost/compress/zstd/dict.go
+++ b/vendor/github.com/klauspost/compress/zstd/dict.go
@@ -82,7 +82,7 @@ func loadDict(b []byte) (*dict, error) {
 			println("Transform table error:", err)
 			return err
 		}
-		if debug {
+		if debugDecoder || debugEncoder {
 			println("Read table ok", "symbolLen:", dec.symbolLen)
 		}
 		// Set decoders as predefined so they aren't reused.
diff --git a/vendor/github.com/klauspost/compress/zstd/enc_best.go b/vendor/github.com/klauspost/compress/zstd/enc_best.go
index fe3625c5f5..b7d4b90047 100644
--- a/vendor/github.com/klauspost/compress/zstd/enc_best.go
+++ b/vendor/github.com/klauspost/compress/zstd/enc_best.go
@@ -132,7 +132,7 @@ func (e *bestFastEncoder) Encode(blk *blockEnc, src []byte) {
 	}
 	_ = addLiterals
 
-	if debug {
+	if debugEncoder {
 		println("recent offsets:", blk.recentOffsets)
 	}
 
@@ -220,6 +220,20 @@ encodeLoop:
 			best = bestOf(best, matchAt(candidateL.prev-e.cur, s, uint32(cv), -1))
 			best = bestOf(best, matchAt(candidateL2.offset-e.cur, s+1, uint32(cv2), -1))
 			best = bestOf(best, matchAt(candidateL2.prev-e.cur, s+1, uint32(cv2), -1))
+
+			// See if we can find a better match by checking where the current best ends.
+			// Use that offset to see if we can find a better full match.
+			if sAt := best.s + best.length; sAt < sLimit {
+				nextHashL := hash8(load6432(src, sAt), bestLongTableBits)
+				candidateEnd := e.longTable[nextHashL]
+				if pos := candidateEnd.offset - e.cur - best.length; pos >= 0 {
+					bestEnd := bestOf(best, matchAt(pos, best.s, load3232(src, best.s), -1))
+					if pos := candidateEnd.prev - e.cur - best.length; pos >= 0 {
+						bestEnd = bestOf(bestEnd, matchAt(pos, best.s, load3232(src, best.s), -1))
+					}
+					best = bestEnd
+				}
+			}
 		}
 
 		// We have a match, we can store the forward value
@@ -260,7 +274,7 @@ encodeLoop:
 
 			nextEmit = s
 			if s >= sLimit {
-				if debug {
+				if debugEncoder {
 					println("repeat ended", s, best.length)
 
 				}
@@ -398,7 +412,7 @@ encodeLoop:
 	blk.recentOffsets[0] = uint32(offset1)
 	blk.recentOffsets[1] = uint32(offset2)
 	blk.recentOffsets[2] = uint32(offset3)
-	if debug {
+	if debugEncoder {
 		println("returning, recent offsets:", blk.recentOffsets, "extra literals:", blk.extraLits)
 	}
 }
diff --git a/vendor/github.com/klauspost/compress/zstd/enc_better.go b/vendor/github.com/klauspost/compress/zstd/enc_better.go
index c2ce4a2bac..eab7b5083e 100644
--- a/vendor/github.com/klauspost/compress/zstd/enc_better.go
+++ b/vendor/github.com/klauspost/compress/zstd/enc_better.go
@@ -138,7 +138,7 @@ func (e *betterFastEncoder) Encode(blk *blockEnc, src []byte) {
 		blk.literals = append(blk.literals, src[nextEmit:until]...)
 		s.litLen = uint32(until - nextEmit)
 	}
-	if debug {
+	if debugEncoder {
 		println("recent offsets:", blk.recentOffsets)
 	}
 
@@ -204,7 +204,7 @@ encodeLoop:
 
 					nextEmit = s
 					if s >= sLimit {
-						if debug {
+						if debugEncoder {
 							println("repeat ended", s, lenght)
 
 						}
@@ -264,7 +264,7 @@ encodeLoop:
 					s += lenght + repOff2
 					nextEmit = s
 					if s >= sLimit {
-						if debug {
+						if debugEncoder {
 							println("repeat ended", s, lenght)
 
 						}
@@ -412,8 +412,41 @@ encodeLoop:
 			cv = load6432(src, s)
 		}
 
-		// A 4-byte match has been found. Update recent offsets.
-		// We'll later see if more than 4 bytes.
+		// Try to find a better match by searching for a long match at the end of the current best match
+		if true && s+matched < sLimit {
+			nextHashL := hash8(load6432(src, s+matched), betterLongTableBits)
+			cv := load3232(src, s)
+			candidateL := e.longTable[nextHashL]
+			coffsetL := candidateL.offset - e.cur - matched
+			if coffsetL >= 0 && coffsetL < s && s-coffsetL < e.maxMatchOff && cv == load3232(src, coffsetL) {
+				// Found a long match, at least 4 bytes.
+				matchedNext := e.matchlen(s+4, coffsetL+4, src) + 4
+				if matchedNext > matched {
+					t = coffsetL
+					matched = matchedNext
+					if debugMatches {
+						println("long match at end-of-match")
+					}
+				}
+			}
+
+			// Check prev long...
+			if true {
+				coffsetL = candidateL.prev - e.cur - matched
+				if coffsetL >= 0 && coffsetL < s && s-coffsetL < e.maxMatchOff && cv == load3232(src, coffsetL) {
+					// Found a long match, at least 4 bytes.
+					matchedNext := e.matchlen(s+4, coffsetL+4, src) + 4
+					if matchedNext > matched {
+						t = coffsetL
+						matched = matchedNext
+						if debugMatches {
+							println("prev long match at end-of-match")
+						}
+					}
+				}
+			}
+		}
+		// A match has been found. Update recent offsets.
 		offset2 = offset1
 		offset1 = s - t
 
@@ -520,7 +553,7 @@ encodeLoop:
 	}
 	blk.recentOffsets[0] = uint32(offset1)
 	blk.recentOffsets[1] = uint32(offset2)
-	if debug {
+	if debugEncoder {
 		println("returning, recent offsets:", blk.recentOffsets, "extra literals:", blk.extraLits)
 	}
 }
@@ -623,7 +656,7 @@ func (e *betterFastEncoderDict) Encode(blk *blockEnc, src []byte) {
 		blk.literals = append(blk.literals, src[nextEmit:until]...)
 		s.litLen = uint32(until - nextEmit)
 	}
-	if debug {
+	if debugEncoder {
 		println("recent offsets:", blk.recentOffsets)
 	}
 
@@ -691,7 +724,7 @@ encodeLoop:
 
 					nextEmit = s
 					if s >= sLimit {
-						if debug {
+						if debugEncoder {
 							println("repeat ended", s, lenght)
 
 						}
@@ -754,7 +787,7 @@ encodeLoop:
 					s += lenght + repOff2
 					nextEmit = s
 					if s >= sLimit {
-						if debug {
+						if debugEncoder {
 							println("repeat ended", s, lenght)
 
 						}
@@ -905,9 +938,41 @@ encodeLoop:
 			}
 			cv = load6432(src, s)
 		}
+		// Try to find a better match by searching for a long match at the end of the current best match
+		if s+matched < sLimit {
+			nextHashL := hash8(load6432(src, s+matched), betterLongTableBits)
+			cv := load3232(src, s)
+			candidateL := e.longTable[nextHashL]
+			coffsetL := candidateL.offset - e.cur - matched
+			if coffsetL >= 0 && coffsetL < s && s-coffsetL < e.maxMatchOff && cv == load3232(src, coffsetL) {
+				// Found a long match, at least 4 bytes.
+				matchedNext := e.matchlen(s+4, coffsetL+4, src) + 4
+				if matchedNext > matched {
+					t = coffsetL
+					matched = matchedNext
+					if debugMatches {
+						println("long match at end-of-match")
+					}
+				}
+			}
 
-		// A 4-byte match has been found. Update recent offsets.
-		// We'll later see if more than 4 bytes.
+			// Check prev long...
+			if true {
+				coffsetL = candidateL.prev - e.cur - matched
+				if coffsetL >= 0 && coffsetL < s && s-coffsetL < e.maxMatchOff && cv == load3232(src, coffsetL) {
+					// Found a long match, at least 4 bytes.
+					matchedNext := e.matchlen(s+4, coffsetL+4, src) + 4
+					if matchedNext > matched {
+						t = coffsetL
+						matched = matchedNext
+						if debugMatches {
+							println("prev long match at end-of-match")
+						}
+					}
+				}
+			}
+		}
+		// A match has been found. Update recent offsets.
 		offset2 = offset1
 		offset1 = s - t
 
@@ -1019,7 +1084,7 @@ encodeLoop:
 	}
 	blk.recentOffsets[0] = uint32(offset1)
 	blk.recentOffsets[1] = uint32(offset2)
-	if debug {
+	if debugEncoder {
 		println("returning, recent offsets:", blk.recentOffsets, "extra literals:", blk.extraLits)
 	}
 }
diff --git a/vendor/github.com/klauspost/compress/zstd/enc_dfast.go b/vendor/github.com/klauspost/compress/zstd/enc_dfast.go
index 8629d43d86..96b21b90e8 100644
--- a/vendor/github.com/klauspost/compress/zstd/enc_dfast.go
+++ b/vendor/github.com/klauspost/compress/zstd/enc_dfast.go
@@ -109,7 +109,7 @@ func (e *doubleFastEncoder) Encode(blk *blockEnc, src []byte) {
 		blk.literals = append(blk.literals, src[nextEmit:until]...)
 		s.litLen = uint32(until - nextEmit)
 	}
-	if debug {
+	if debugEncoder {
 		println("recent offsets:", blk.recentOffsets)
 	}
 
@@ -170,7 +170,7 @@ encodeLoop:
 					s += lenght + repOff
 					nextEmit = s
 					if s >= sLimit {
-						if debug {
+						if debugEncoder {
 							println("repeat ended", s, lenght)
 
 						}
@@ -368,7 +368,7 @@ encodeLoop:
 	}
 	blk.recentOffsets[0] = uint32(offset1)
 	blk.recentOffsets[1] = uint32(offset2)
-	if debug {
+	if debugEncoder {
 		println("returning, recent offsets:", blk.recentOffsets, "extra literals:", blk.extraLits)
 	}
 }
@@ -427,7 +427,7 @@ func (e *doubleFastEncoder) EncodeNoHist(blk *blockEnc, src []byte) {
 		blk.literals = append(blk.literals, src[nextEmit:until]...)
 		s.litLen = uint32(until - nextEmit)
 	}
-	if debug {
+	if debugEncoder {
 		println("recent offsets:", blk.recentOffsets)
 	}
 
@@ -483,7 +483,7 @@ encodeLoop:
 					s += length + repOff
 					nextEmit = s
 					if s >= sLimit {
-						if debug {
+						if debugEncoder {
 							println("repeat ended", s, length)
 
 						}
@@ -677,7 +677,7 @@ encodeLoop:
 		blk.literals = append(blk.literals, src[nextEmit:]...)
 		blk.extraLits = len(src) - int(nextEmit)
 	}
-	if debug {
+	if debugEncoder {
 		println("returning, recent offsets:", blk.recentOffsets, "extra literals:", blk.extraLits)
 	}
 
@@ -767,7 +767,7 @@ func (e *doubleFastEncoderDict) Encode(blk *blockEnc, src []byte) {
 		blk.literals = append(blk.literals, src[nextEmit:until]...)
 		s.litLen = uint32(until - nextEmit)
 	}
-	if debug {
+	if debugEncoder {
 		println("recent offsets:", blk.recentOffsets)
 	}
 
@@ -830,7 +830,7 @@ encodeLoop:
 					s += lenght + repOff
 					nextEmit = s
 					if s >= sLimit {
-						if debug {
+						if debugEncoder {
 							println("repeat ended", s, lenght)
 
 						}
@@ -1039,7 +1039,7 @@ encodeLoop:
 	}
 	blk.recentOffsets[0] = uint32(offset1)
 	blk.recentOffsets[1] = uint32(offset2)
-	if debug {
+	if debugEncoder {
 		println("returning, recent offsets:", blk.recentOffsets, "extra literals:", blk.extraLits)
 	}
 	// If we encoded more than 64K mark all dirty.
diff --git a/vendor/github.com/klauspost/compress/zstd/enc_fast.go b/vendor/github.com/klauspost/compress/zstd/enc_fast.go
index ba4a17e106..2246d286dc 100644
--- a/vendor/github.com/klauspost/compress/zstd/enc_fast.go
+++ b/vendor/github.com/klauspost/compress/zstd/enc_fast.go
@@ -103,7 +103,7 @@ func (e *fastEncoder) Encode(blk *blockEnc, src []byte) {
 		blk.literals = append(blk.literals, src[nextEmit:until]...)
 		s.litLen = uint32(until - nextEmit)
 	}
-	if debug {
+	if debugEncoder {
 		println("recent offsets:", blk.recentOffsets)
 	}
 
@@ -178,7 +178,7 @@ encodeLoop:
 				s += length + 2
 				nextEmit = s
 				if s >= sLimit {
-					if debug {
+					if debugEncoder {
 						println("repeat ended", s, length)
 
 					}
@@ -330,7 +330,7 @@ encodeLoop:
 	}
 	blk.recentOffsets[0] = uint32(offset1)
 	blk.recentOffsets[1] = uint32(offset2)
-	if debug {
+	if debugEncoder {
 		println("returning, recent offsets:", blk.recentOffsets, "extra literals:", blk.extraLits)
 	}
 }
@@ -343,7 +343,7 @@ func (e *fastEncoder) EncodeNoHist(blk *blockEnc, src []byte) {
 		inputMargin            = 8
 		minNonLiteralBlockSize = 1 + 1 + inputMargin
 	)
-	if debug {
+	if debugEncoder {
 		if len(src) > maxBlockSize {
 			panic("src too big")
 		}
@@ -391,7 +391,7 @@ func (e *fastEncoder) EncodeNoHist(blk *blockEnc, src []byte) {
 		blk.literals = append(blk.literals, src[nextEmit:until]...)
 		s.litLen = uint32(until - nextEmit)
 	}
-	if debug {
+	if debugEncoder {
 		println("recent offsets:", blk.recentOffsets)
 	}
 
@@ -462,7 +462,7 @@ encodeLoop:
 				s += length + 2
 				nextEmit = s
 				if s >= sLimit {
-					if debug {
+					if debugEncoder {
 						println("repeat ended", s, length)
 
 					}
@@ -616,7 +616,7 @@ encodeLoop:
 		blk.literals = append(blk.literals, src[nextEmit:]...)
 		blk.extraLits = len(src) - int(nextEmit)
 	}
-	if debug {
+	if debugEncoder {
 		println("returning, recent offsets:", blk.recentOffsets, "extra literals:", blk.extraLits)
 	}
 	// We do not store history, so we must offset e.cur to avoid false matches for next user.
@@ -696,7 +696,7 @@ func (e *fastEncoderDict) Encode(blk *blockEnc, src []byte) {
 		blk.literals = append(blk.literals, src[nextEmit:until]...)
 		s.litLen = uint32(until - nextEmit)
 	}
-	if debug {
+	if debugEncoder {
 		println("recent offsets:", blk.recentOffsets)
 	}
 
@@ -773,7 +773,7 @@ encodeLoop:
 				s += length + 2
 				nextEmit = s
 				if s >= sLimit {
-					if debug {
+					if debugEncoder {
 						println("repeat ended", s, length)
 
 					}
@@ -926,7 +926,7 @@ encodeLoop:
 	}
 	blk.recentOffsets[0] = uint32(offset1)
 	blk.recentOffsets[1] = uint32(offset2)
-	if debug {
+	if debugEncoder {
 		println("returning, recent offsets:", blk.recentOffsets, "extra literals:", blk.extraLits)
 	}
 }
diff --git a/vendor/github.com/klauspost/compress/zstd/encoder.go b/vendor/github.com/klauspost/compress/zstd/encoder.go
index 4871dd03af..ea85548fc9 100644
--- a/vendor/github.com/klauspost/compress/zstd/encoder.go
+++ b/vendor/github.com/klauspost/compress/zstd/encoder.go
@@ -245,7 +245,7 @@ func (e *Encoder) nextBlock(final bool) error {
 	s.filling, s.current, s.previous = s.previous[:0], s.filling, s.current
 	s.wg.Add(1)
 	go func(src []byte) {
-		if debug {
+		if debugEncoder {
 			println("Adding block,", len(src), "bytes, final:", final)
 		}
 		defer func() {
@@ -290,7 +290,7 @@ func (e *Encoder) nextBlock(final bool) error {
 			}
 			switch err {
 			case errIncompressible:
-				if debug {
+				if debugEncoder {
 					println("Storing incompressible block as raw")
 				}
 				blk.encodeRaw(src)
@@ -313,7 +313,7 @@ func (e *Encoder) nextBlock(final bool) error {
 //
 // The Copy function uses ReaderFrom if available.
 func (e *Encoder) ReadFrom(r io.Reader) (n int64, err error) {
-	if debug {
+	if debugEncoder {
 		println("Using ReadFrom")
 	}
 
@@ -336,20 +336,20 @@ func (e *Encoder) ReadFrom(r io.Reader) (n int64, err error) {
 		switch err {
 		case io.EOF:
 			e.state.filling = e.state.filling[:len(e.state.filling)-len(src)]
-			if debug {
+			if debugEncoder {
 				println("ReadFrom: got EOF final block:", len(e.state.filling))
 			}
 			return n, nil
 		case nil:
 		default:
-			if debug {
+			if debugEncoder {
 				println("ReadFrom: got error:", err)
 			}
 			e.state.err = err
 			return n, err
 		}
 		if len(src) > 0 {
-			if debug {
+			if debugEncoder {
 				println("ReadFrom: got space left in source:", len(src))
 			}
 			continue
@@ -512,7 +512,7 @@ func (e *Encoder) EncodeAll(src, dst []byte) []byte {
 
 		switch err {
 		case errIncompressible:
-			if debug {
+			if debugEncoder {
 				println("Storing incompressible block as raw")
 			}
 			dst = blk.encodeRawTo(dst, src)
@@ -548,7 +548,7 @@ func (e *Encoder) EncodeAll(src, dst []byte) []byte {
 
 			switch err {
 			case errIncompressible:
-				if debug {
+				if debugEncoder {
 					println("Storing incompressible block as raw")
 				}
 				dst = blk.encodeRawTo(dst, todo)
diff --git a/vendor/github.com/klauspost/compress/zstd/framedec.go b/vendor/github.com/klauspost/compress/zstd/framedec.go
index 693c5f05d2..e8cc9a2c22 100644
--- a/vendor/github.com/klauspost/compress/zstd/framedec.go
+++ b/vendor/github.com/klauspost/compress/zstd/framedec.go
@@ -78,44 +78,68 @@ func newFrameDec(o decoderOptions) *frameDec {
 func (d *frameDec) reset(br byteBuffer) error {
 	d.HasCheckSum = false
 	d.WindowSize = 0
-	var b []byte
+	var signature [4]byte
 	for {
-		b = br.readSmall(4)
-		if b == nil {
+		var err error
+		// Check if we can read more...
+		b, err := br.readSmall(1)
+		switch err {
+		case io.EOF, io.ErrUnexpectedEOF:
 			return io.EOF
+		default:
+			return err
+		case nil:
+			signature[0] = b[0]
 		}
-		if !bytes.Equal(b[1:4], skippableFrameMagic) || b[0]&0xf0 != 0x50 {
-			if debug {
-				println("Not skippable", hex.EncodeToString(b), hex.EncodeToString(skippableFrameMagic))
+		// Read the rest, don't allow io.ErrUnexpectedEOF
+		b, err = br.readSmall(3)
+		switch err {
+		case io.EOF:
+			return io.EOF
+		default:
+			return err
+		case nil:
+			copy(signature[1:], b)
+		}
+
+		if !bytes.Equal(signature[1:4], skippableFrameMagic) || signature[0]&0xf0 != 0x50 {
+			if debugDecoder {
+				println("Not skippable", hex.EncodeToString(signature[:]), hex.EncodeToString(skippableFrameMagic))
 			}
 			// Break if not skippable frame.
 			break
 		}
 		// Read size to skip
-		b = br.readSmall(4)
-		if b == nil {
-			println("Reading Frame Size EOF")
-			return io.ErrUnexpectedEOF
+		b, err = br.readSmall(4)
+		if err != nil {
+			if debugDecoder {
+				println("Reading Frame Size", err)
+			}
+			return err
 		}
 		n := uint32(b[0]) | (uint32(b[1]) << 8) | (uint32(b[2]) << 16) | (uint32(b[3]) << 24)
 		println("Skipping frame with", n, "bytes.")
-		err := br.skipN(int(n))
+		err = br.skipN(int(n))
 		if err != nil {
-			if debug {
+			if debugDecoder {
 				println("Reading discarded frame", err)
 			}
 			return err
 		}
 	}
-	if !bytes.Equal(b, frameMagic) {
-		println("Got magic numbers: ", b, "want:", frameMagic)
+	if !bytes.Equal(signature[:], frameMagic) {
+		if debugDecoder {
+			println("Got magic numbers: ", signature, "want:", frameMagic)
+		}
 		return ErrMagicMismatch
 	}
 
 	// Read Frame_Header_Descriptor
 	fhd, err := br.readByte()
 	if err != nil {
-		println("Reading Frame_Header_Descriptor", err)
+		if debugDecoder {
+			println("Reading Frame_Header_Descriptor", err)
+		}
 		return err
 	}
 	d.SingleSegment = fhd&(1<<5) != 0
@@ -130,7 +154,9 @@ func (d *frameDec) reset(br byteBuffer) error {
 	if !d.SingleSegment {
 		wd, err := br.readByte()
 		if err != nil {
-			println("Reading Window_Descriptor", err)
+			if debugDecoder {
+				println("Reading Window_Descriptor", err)
+			}
 			return err
 		}
 		printf("raw: %x, mantissa: %d, exponent: %d\n", wd, wd&7, wd>>3)
@@ -147,12 +173,11 @@ func (d *frameDec) reset(br byteBuffer) error {
 		if size == 3 {
 			size = 4
 		}
-		b = br.readSmall(int(size))
-		if b == nil {
-			if debug {
-				println("Reading Dictionary_ID", io.ErrUnexpectedEOF)
-			}
-			return io.ErrUnexpectedEOF
+
+		b, err := br.readSmall(int(size))
+		if err != nil {
+			println("Reading Dictionary_ID", err)
+			return err
 		}
 		var id uint32
 		switch size {
@@ -163,7 +188,7 @@ func (d *frameDec) reset(br byteBuffer) error {
 		case 4:
 			id = uint32(b[0]) | (uint32(b[1]) << 8) | (uint32(b[2]) << 16) | (uint32(b[3]) << 24)
 		}
-		if debug {
+		if debugDecoder {
 			println("Dict size", size, "ID:", id)
 		}
 		if id > 0 {
@@ -187,10 +212,10 @@ func (d *frameDec) reset(br byteBuffer) error {
 	}
 	d.FrameContentSize = 0
 	if fcsSize > 0 {
-		b := br.readSmall(fcsSize)
-		if b == nil {
-			println("Reading Frame content", io.ErrUnexpectedEOF)
-			return io.ErrUnexpectedEOF
+		b, err := br.readSmall(fcsSize)
+		if err != nil {
+			println("Reading Frame content", err)
+			return err
 		}
 		switch fcsSize {
 		case 1:
@@ -205,7 +230,7 @@ func (d *frameDec) reset(br byteBuffer) error {
 			d2 := uint32(b[4]) | (uint32(b[5]) << 8) | (uint32(b[6]) << 16) | (uint32(b[7]) << 24)
 			d.FrameContentSize = uint64(d1) | (uint64(d2) << 32)
 		}
-		if debug {
+		if debugDecoder {
 			println("field size bits:", v, "fcsSize:", fcsSize, "FrameContentSize:", d.FrameContentSize, hex.EncodeToString(b[:fcsSize]), "singleseg:", d.SingleSegment, "window:", d.WindowSize)
 		}
 	}
@@ -248,7 +273,7 @@ func (d *frameDec) reset(br byteBuffer) error {
 
 // next will start decoding the next block from stream.
 func (d *frameDec) next(block *blockDec) error {
-	if debug {
+	if debugDecoder {
 		printf("decoding new block %p:%p", block, block.data)
 	}
 	err := block.reset(d.rawInput, d.WindowSize)
@@ -259,7 +284,7 @@ func (d *frameDec) next(block *blockDec) error {
 		return err
 	}
 	block.input <- struct{}{}
-	if debug {
+	if debugDecoder {
 		println("next block:", block)
 	}
 	d.asyncRunningMu.Lock()
@@ -307,19 +332,19 @@ func (d *frameDec) checkCRC() error {
 	tmp[3] = byte(got >> 24)
 
 	// We can overwrite upper tmp now
-	want := d.rawInput.readSmall(4)
-	if want == nil {
-		println("CRC missing?")
-		return io.ErrUnexpectedEOF
+	want, err := d.rawInput.readSmall(4)
+	if err != nil {
+		println("CRC missing?", err)
+		return err
 	}
 
 	if !bytes.Equal(tmp[:], want) {
-		if debug {
+		if debugDecoder {
 			println("CRC Check Failed:", tmp[:], "!=", want)
 		}
 		return ErrCRCMismatch
 	}
-	if debug {
+	if debugDecoder {
 		println("CRC ok", tmp[:])
 	}
 	return nil
@@ -340,7 +365,7 @@ func (d *frameDec) initAsync() {
 	if cap(d.decoding) < d.o.concurrent {
 		d.decoding = make(chan *blockDec, d.o.concurrent)
 	}
-	if debug {
+	if debugDecoder {
 		h := d.history
 		printf("history init. len: %d, cap: %d", len(h.b), cap(h.b))
 	}
@@ -388,7 +413,7 @@ func (d *frameDec) startDecoder(output chan decodeOutput) {
 			output <- r
 			return
 		}
-		if debug {
+		if debugDecoder {
 			println("got result, from ", d.offset, "to", d.offset+int64(len(r.b)))
 			d.offset += int64(len(r.b))
 		}
@@ -396,7 +421,7 @@ func (d *frameDec) startDecoder(output chan decodeOutput) {
 			// Send history to next block
 			select {
 			case next = <-d.decoding:
-				if debug {
+				if debugDecoder {
 					println("Sending ", len(d.history.b), "bytes as history")
 				}
 				next.history <- &d.history
@@ -434,7 +459,7 @@ func (d *frameDec) startDecoder(output chan decodeOutput) {
 		output <- r
 		if next == nil {
 			// There was no decoder available, we wait for one now that we have sent to the writer.
-			if debug {
+			if debugDecoder {
 				println("Sending ", len(d.history.b), " bytes as history")
 			}
 			next = <-d.decoding
@@ -458,7 +483,7 @@ func (d *frameDec) runDecoder(dst []byte, dec *blockDec) ([]byte, error) {
 		if err != nil {
 			break
 		}
-		if debug {
+		if debugDecoder {
 			println("next block:", dec)
 		}
 		err = dec.decodeBuf(&d.history)
diff --git a/vendor/github.com/klauspost/compress/zstd/fse_encoder.go b/vendor/github.com/klauspost/compress/zstd/fse_encoder.go
index c74681b999..b4757ee3f0 100644
--- a/vendor/github.com/klauspost/compress/zstd/fse_encoder.go
+++ b/vendor/github.com/klauspost/compress/zstd/fse_encoder.go
@@ -229,7 +229,7 @@ func (s *fseEncoder) setRLE(val byte) {
 		deltaFindState: 0,
 		deltaNbBits:    0,
 	}
-	if debug {
+	if debugEncoder {
 		println("setRLE: val", val, "symbolTT", s.ct.symbolTT[val])
 	}
 	s.rleVal = val
diff --git a/vendor/github.com/klauspost/compress/zstd/snappy.go b/vendor/github.com/klauspost/compress/zstd/snappy.go
index 9d9d1d567e..0372b1714a 100644
--- a/vendor/github.com/klauspost/compress/zstd/snappy.go
+++ b/vendor/github.com/klauspost/compress/zstd/snappy.go
@@ -203,7 +203,7 @@ func (r *SnappyConverter) Convert(in io.Reader, w io.Writer) (int64, error) {
 			written += int64(n)
 			continue
 		case chunkTypeUncompressedData:
-			if debug {
+			if debugEncoder {
 				println("Uncompressed, chunklen", chunkLen)
 			}
 			// Section 4.3. Uncompressed data (chunk type 0x01).
@@ -246,7 +246,7 @@ func (r *SnappyConverter) Convert(in io.Reader, w io.Writer) (int64, error) {
 			continue
 
 		case chunkTypeStreamIdentifier:
-			if debug {
+			if debugEncoder {
 				println("stream id", chunkLen, len(snappyMagicBody))
 			}
 			// Section 4.1. Stream identifier (chunk type 0xff).
diff --git a/vendor/github.com/klauspost/compress/zstd/zip.go b/vendor/github.com/klauspost/compress/zstd/zip.go
new file mode 100644
index 0000000000..9325b928ae
--- /dev/null
+++ b/vendor/github.com/klauspost/compress/zstd/zip.go
@@ -0,0 +1,121 @@
+// Copyright 2019+ Klaus Post. All rights reserved.
+// License information can be found in the LICENSE file.
+
+package zstd
+
+import (
+	"errors"
+	"io"
+	"sync"
+)
+
+// ZipMethodWinZip is the method for Zstandard compressed data inside Zip files for WinZip.
+// See https://www.winzip.com/win/en/comp_info.html
+const ZipMethodWinZip = 93
+
+// ZipMethodPKWare is the original method number used by PKWARE to indicate Zstandard compression.
+// Deprecated: This has been deprecated by PKWARE, use ZipMethodWinZip instead for compression.
+// See https://pkware.cachefly.net/webdocs/APPNOTE/APPNOTE-6.3.9.TXT
+const ZipMethodPKWare = 20
+
+var zipReaderPool sync.Pool
+
+// newZipReader cannot be used since we would leak goroutines...
+func newZipReader(r io.Reader) io.ReadCloser {
+	dec, ok := zipReaderPool.Get().(*Decoder)
+	if ok {
+		dec.Reset(r)
+	} else {
+		d, err := NewReader(r, WithDecoderConcurrency(1), WithDecoderLowmem(true))
+		if err != nil {
+			panic(err)
+		}
+		dec = d
+	}
+	return &pooledZipReader{dec: dec}
+}
+
+type pooledZipReader struct {
+	mu  sync.Mutex // guards Close and Read
+	dec *Decoder
+}
+
+func (r *pooledZipReader) Read(p []byte) (n int, err error) {
+	r.mu.Lock()
+	defer r.mu.Unlock()
+	if r.dec == nil {
+		return 0, errors.New("Read after Close")
+	}
+	dec, err := r.dec.Read(p)
+
+	return dec, err
+}
+
+func (r *pooledZipReader) Close() error {
+	r.mu.Lock()
+	defer r.mu.Unlock()
+	var err error
+	if r.dec != nil {
+		err = r.dec.Reset(nil)
+		zipReaderPool.Put(r.dec)
+		r.dec = nil
+	}
+	return err
+}
+
+type pooledZipWriter struct {
+	mu  sync.Mutex // guards Close and Read
+	enc *Encoder
+}
+
+func (w *pooledZipWriter) Write(p []byte) (n int, err error) {
+	w.mu.Lock()
+	defer w.mu.Unlock()
+	if w.enc == nil {
+		return 0, errors.New("Write after Close")
+	}
+	return w.enc.Write(p)
+}
+
+func (w *pooledZipWriter) Close() error {
+	w.mu.Lock()
+	defer w.mu.Unlock()
+	var err error
+	if w.enc != nil {
+		err = w.enc.Close()
+		zipReaderPool.Put(w.enc)
+		w.enc = nil
+	}
+	return err
+}
+
+// ZipCompressor returns a compressor that can be registered with zip libraries.
+// The provided encoder options will be used on all encodes.
+func ZipCompressor(opts ...EOption) func(w io.Writer) (io.WriteCloser, error) {
+	var pool sync.Pool
+	return func(w io.Writer) (io.WriteCloser, error) {
+		enc, ok := pool.Get().(*Encoder)
+		if ok {
+			enc.Reset(w)
+		} else {
+			var err error
+			enc, err = NewWriter(w, opts...)
+			if err != nil {
+				return nil, err
+			}
+		}
+		return &pooledZipWriter{enc: enc}, nil
+	}
+}
+
+// ZipDecompressor returns a decompressor that can be registered with zip libraries.
+// See ZipCompressor for example.
+func ZipDecompressor() func(r io.Reader) io.ReadCloser {
+	return func(r io.Reader) io.ReadCloser {
+		d, err := NewReader(r, WithDecoderConcurrency(1), WithDecoderLowmem(true))
+		if err != nil {
+			panic(err)
+		}
+		return d.IOReadCloser()
+	}
+}
diff --git a/vendor/github.com/klauspost/compress/zstd/zstd.go b/vendor/github.com/klauspost/compress/zstd/zstd.go
index 1ba308c8bf..ef1d49a009 100644
--- a/vendor/github.com/klauspost/compress/zstd/zstd.go
+++ b/vendor/github.com/klauspost/compress/zstd/zstd.go
@@ -15,6 +15,12 @@ import (
 // enable debug printing
 const debug = false
 
+// enable encoding debug printing
+const debugEncoder = debug
+
+// enable decoding debug printing
+const debugDecoder = debug
+
 // Enable extra assertions.
 const debugAsserts = debug || false
 
@@ -82,13 +88,13 @@ var (
 )
 
 func println(a ...interface{}) {
-	if debug {
+	if debug || debugDecoder || debugEncoder {
 		log.Println(a...)
 	}
 }
 
 func printf(format string, a ...interface{}) {
-	if debug {
+	if debug || debugDecoder || debugEncoder {
 		log.Printf(format, a...)
 	}
 }