From 8081d9c74552612ad73fef27e5775fcfb371e812 Mon Sep 17 00:00:00 2001 From: Daniel J Walsh Date: Tue, 17 Mar 2020 09:52:11 -0400 Subject: Update containers/storage to v1.16.5 Signed-off-by: Daniel J Walsh --- vendor/github.com/klauspost/pgzip/README.md | 17 +++---- vendor/github.com/klauspost/pgzip/gzip.go | 76 ++++++++++++++++++----------- 2 files changed, 55 insertions(+), 38 deletions(-) (limited to 'vendor/github.com/klauspost/pgzip') diff --git a/vendor/github.com/klauspost/pgzip/README.md b/vendor/github.com/klauspost/pgzip/README.md index 81000996c..171b978fd 100644 --- a/vendor/github.com/klauspost/pgzip/README.md +++ b/vendor/github.com/klauspost/pgzip/README.md @@ -39,7 +39,6 @@ You might need to get/update the dependencies: ``` go get -u github.com/klauspost/compress -go get -u github.com/klauspost/crc32 ``` Usage @@ -65,7 +64,7 @@ Changes in [github.com/klauspost/compress](https://github.com/klauspost/compress ## Compression The simplest way to use this is to simply do the same as you would when using [compress/gzip](http://golang.org/pkg/compress/gzip). -To change the block size, use the added (*pgzip.Writer).SetConcurrency(blockSize, blocks int) function. With this you can control the approximate size of your blocks, as well as how many you want to be processing in parallel. Default values for this is SetConcurrency(250000, 16), meaning blocks are split at 250000 bytes and up to 16 blocks can be processing at once before the writer blocks. +To change the block size, use the added (*pgzip.Writer).SetConcurrency(blockSize, blocks int) function. With this you can control the approximate size of your blocks, as well as how many you want to be processing in parallel. Default values for this is SetConcurrency(1MB, runtime.GOMAXPROCS(0)), meaning blocks are split at 1 MB and up to the number of CPU threads blocks can be processing at once before the writer blocks. Example: @@ -99,19 +98,19 @@ See my blog post in [Benchmarks of Golang Gzip](https://blog.klauspost.com/go-gz Compression cost is usually about 0.2% with default settings with a block size of 250k. -Example with GOMAXPROC set to 8 (quad core with 8 hyperthreads) +Example with GOMAXPROC set to 32 (16 core CPU) Content is [Matt Mahoneys 10GB corpus](http://mattmahoney.net/dc/10gb.html). Compression level 6. Compressor | MB/sec | speedup | size | size overhead (lower=better) ------------|----------|---------|------|--------- -[gzip](http://golang.org/pkg/compress/gzip) (golang) | 7.21MB/s | 1.0x | 4786608902 | 0% -[gzip](http://github.com/klauspost/compress/gzip) (klauspost) | 10.98MB/s | 1.52x | 4781331645 | -0.11% -[pgzip](https://github.com/klauspost/pgzip) (klauspost) | 50.76MB/s|7.04x | 4784121440 | -0.052% -[bgzf](https://godoc.org/github.com/biogo/hts/bgzf) (biogo) | 38.65MB/s | 5.36x | 4924899484 | 2.889% -[pargzip](https://godoc.org/github.com/golang/build/pargzip) (builder) | 32.00MB/s | 4.44x | 4791226567 | 0.096% +[gzip](http://golang.org/pkg/compress/gzip) (golang) | 15.44MB/s (1 thread) | 1.0x | 4781329307 | 0% +[gzip](http://github.com/klauspost/compress/gzip) (klauspost) | 135.04MB/s (1 thread) | 8.74x | 4894858258 | +2.37% +[pgzip](https://github.com/klauspost/pgzip) (klauspost) | 1573.23MB/s| 101.9x | 4902285651 | +2.53% +[bgzf](https://godoc.org/github.com/biogo/hts/bgzf) (biogo) | 361.40MB/s | 23.4x | 4869686090 | +1.85% +[pargzip](https://godoc.org/github.com/golang/build/pargzip) (builder) | 306.01MB/s | 19.8x | 4786890417 | +0.12% -pgzip also contains a [linear time compression](https://github.com/klauspost/compress#linear-time-compression) mode, that will allow compression at ~150MB per core per second, independent of the content. +pgzip also contains a [linear time compression](https://github.com/klauspost/compress#linear-time-compression-huffman-only) mode, that will allow compression at ~250MB per core per second, independent of the content. See the [complete sheet](https://docs.google.com/spreadsheets/d/1nuNE2nPfuINCZJRMt6wFWhKpToF95I47XjSsc-1rbPQ/edit?usp=sharing) for different content types and compression settings. diff --git a/vendor/github.com/klauspost/pgzip/gzip.go b/vendor/github.com/klauspost/pgzip/gzip.go index 85d14e9cb..bb2e33941 100644 --- a/vendor/github.com/klauspost/pgzip/gzip.go +++ b/vendor/github.com/klauspost/pgzip/gzip.go @@ -11,6 +11,7 @@ import ( "hash" "hash/crc32" "io" + "runtime" "sync" "time" @@ -18,9 +19,9 @@ import ( ) const ( - defaultBlockSize = 256 << 10 + defaultBlockSize = 1 << 20 tailSize = 16384 - defaultBlocks = 16 + defaultBlocks = 4 ) // These constants are copied from the flate package, so that code that imports @@ -68,8 +69,8 @@ type result struct { // With this you can control the approximate size of your blocks, // as well as how many you want to be processing in parallel. // -// Default values for this is SetConcurrency(250000, 16), -// meaning blocks are split at 250000 bytes and up to 16 blocks +// Default values for this is SetConcurrency(defaultBlockSize, runtime.GOMAXPROCS(0)), +// meaning blocks are split at 1 MB and up to the number of CPU threads // can be processing at once before the writer blocks. func (z *Writer) SetConcurrency(blockSize, blocks int) error { if blockSize <= tailSize { @@ -115,7 +116,7 @@ func NewWriterLevel(w io.Writer, level int) (*Writer, error) { return nil, fmt.Errorf("gzip: invalid compression level: %d", level) } z := new(Writer) - z.SetConcurrency(defaultBlockSize, defaultBlocks) + z.SetConcurrency(defaultBlockSize, runtime.GOMAXPROCS(0)) z.init(w, level) return z, nil } @@ -174,7 +175,7 @@ func (z *Writer) Reset(w io.Writer) { if z.results != nil && !z.closed { close(z.results) } - z.SetConcurrency(defaultBlockSize, defaultBlocks) + z.SetConcurrency(defaultBlockSize, runtime.GOMAXPROCS(0)) z.init(w, z.level) } @@ -239,36 +240,36 @@ func (z *Writer) writeString(s string) (err error) { // compressCurrent will compress the data currently buffered // This should only be called from the main writer/flush/closer func (z *Writer) compressCurrent(flush bool) { + c := z.currentBuffer + if len(c) > z.blockSize { + // This can never happen through the public interface. + panic("len(z.currentBuffer) > z.blockSize (most likely due to concurrent Write race)") + } + r := result{} r.result = make(chan []byte, 1) r.notifyWritten = make(chan struct{}, 0) + // Reserve a result slot select { case z.results <- r: case <-z.pushedErr: return } - // If block given is more than twice the block size, split it. - c := z.currentBuffer - if len(c) > z.blockSize*2 { - c = c[:z.blockSize] - z.wg.Add(1) - go z.compressBlock(c, z.prevTail, r, false) - z.prevTail = c[len(c)-tailSize:] - z.currentBuffer = z.currentBuffer[z.blockSize:] - z.compressCurrent(flush) - // Last one flushes if needed - return - } - z.wg.Add(1) - go z.compressBlock(c, z.prevTail, r, z.closed) + tail := z.prevTail if len(c) > tailSize { - z.prevTail = c[len(c)-tailSize:] + buf := z.dstPool.Get().([]byte) // Put in .compressBlock + // Copy tail from current buffer before handing the buffer over to the + // compressBlock goroutine. + buf = append(buf[:0], c[len(c)-tailSize:]...) + z.prevTail = buf } else { z.prevTail = nil } - z.currentBuffer = z.dstPool.Get().([]byte) + go z.compressBlock(c, tail, r, z.closed) + + z.currentBuffer = z.dstPool.Get().([]byte) // Put in .compressBlock z.currentBuffer = z.currentBuffer[:0] // Wait if flushing @@ -358,29 +359,37 @@ func (z *Writer) Write(p []byte) (int, error) { // Start receiving data from compressors go func() { listen := z.results + var failed bool for { r, ok := <-listen // If closed, we are finished. if !ok { return } + if failed { + close(r.notifyWritten) + continue + } buf := <-r.result n, err := z.w.Write(buf) if err != nil { z.pushError(err) close(r.notifyWritten) - return + failed = true + continue } if n != len(buf) { z.pushError(fmt.Errorf("gzip: short write %d should be %d", n, len(buf))) + failed = true close(r.notifyWritten) - return + continue } z.dstPool.Put(buf) close(r.notifyWritten) } }() - z.currentBuffer = make([]byte, 0, z.blockSize) + z.currentBuffer = z.dstPool.Get().([]byte) + z.currentBuffer = z.currentBuffer[:0] } q := p for len(q) > 0 { @@ -390,7 +399,10 @@ func (z *Writer) Write(p []byte) (int, error) { } z.digest.Write(q[:length]) z.currentBuffer = append(z.currentBuffer, q[:length]...) - if len(z.currentBuffer) >= z.blockSize { + if len(z.currentBuffer) > z.blockSize { + panic("z.currentBuffer too large (most likely due to concurrent Write race)") + } + if len(z.currentBuffer) == z.blockSize { z.compressCurrent(false) if err := z.checkError(); err != nil { return len(p) - len(q) - length, err @@ -410,12 +422,13 @@ func (z *Writer) compressBlock(p, prevTail []byte, r result, closed bool) { close(r.result) z.wg.Done() }() - buf := z.dstPool.Get().([]byte) + buf := z.dstPool.Get().([]byte) // Corresponding Put in .Write's result writer dest := bytes.NewBuffer(buf[:0]) - compressor := z.dictFlatePool.Get().(*flate.Writer) + compressor := z.dictFlatePool.Get().(*flate.Writer) // Put below compressor.ResetDict(dest, prevTail) compressor.Write(p) + z.dstPool.Put(p) // Corresponding Get in .Write and .compressCurrent err := compressor.Flush() if err != nil { @@ -429,7 +442,12 @@ func (z *Writer) compressBlock(p, prevTail []byte, r result, closed bool) { return } } - z.dictFlatePool.Put(compressor) + z.dictFlatePool.Put(compressor) // Get above + + if prevTail != nil { + z.dstPool.Put(prevTail) // Get in .compressCurrent + } + // Read back buffer buf = dest.Bytes() r.result <- buf -- cgit v1.2.3-54-g00ecf