more solid weed mount (#4089)

* compare chunks by timestamp * fix slab clearing error * fix test compilation * move oldest chunk to sealed, instead of by fullness * lock on fh.entryViewCache * remove verbose logs * revert slat clearing * less logs * less logs * track write and read by timestamp * remove useless logic * add entry lock on file handle release * use mem chunk only, swap file chunk has problems * comment out code that maybe used later * add debug mode to compare data read and write * more efficient readResolvedChunks with linked list * small optimization * fix test compilation * minor fix on writer * add SeparateGarbageChunks * group chunks into sections * turn off debug mode * fix tests * fix tests * tmp enable swap file chunk * Revert "tmp enable swap file chunk" This reverts commit 985137ec472924e4815f258189f6ca9f2168a0a7. * simple refactoring * simple refactoring * do not re-use swap file chunk. Sealed chunks should not be re-used. * comment out debugging facilities * either mem chunk or swap file chunk is fine now * remove orderedMutex as *semaphore.Weighted not found impactful * optimize size calculation for changing large files * optimize performance to avoid going through the long list of chunks * still problems with swap file chunk * rename * tiny optimization * swap file chunk save only successfully read data * fix * enable both mem and swap file chunk * resolve chunks with range * rename * fix chunk interval list * also change file handle chunk group when adding chunks * pick in-active chunk with time-decayed counter * fix compilation * avoid nil with empty fh.entry * refactoring * rename * rename * refactor visible intervals to *list.List * refactor chunkViews to *list.List * add IntervalList for generic interval list * change visible interval to use IntervalList in generics * cahnge chunkViews to *IntervalList[*ChunkView] * use NewFileChunkSection to create * rename variables * refactor * fix renaming leftover * renaming * renaming * add insert interval * interval list adds lock * incrementally add chunks to readers Fixes: 1. set start and stop offset for the value object 2. clone the value object 3. use pointer instead of copy-by-value when passing to interval.Value 4. use insert interval since adding chunk could be out of order * fix tests compilation * fix tests compilation
2023-01-02 23:20:45 -08:00
parent 367353b936
commit d4566d4aaa
45 changed files with 1834 additions and 805 deletions
--- a/weed/filer/filechunk_group.go
+++ b/weed/filer/filechunk_group.go
@@ -0,0 +1,148 @@
+package filer
+
+import (
+	"github.com/seaweedfs/seaweedfs/weed/pb/filer_pb"
+	"github.com/seaweedfs/seaweedfs/weed/util/chunk_cache"
+	"github.com/seaweedfs/seaweedfs/weed/wdclient"
+	"sync"
+)
+
+type ChunkGroup struct {
+	lookupFn       wdclient.LookupFileIdFunctionType
+	chunkCache     chunk_cache.ChunkCache
+	manifestChunks []*filer_pb.FileChunk
+	sections       map[SectionIndex]*FileChunkSection
+	sectionsLock   sync.RWMutex
+}
+
+func NewChunkGroup(lookupFn wdclient.LookupFileIdFunctionType, chunkCache chunk_cache.ChunkCache, chunks []*filer_pb.FileChunk) (*ChunkGroup, error) {
+	group := &ChunkGroup{
+		lookupFn:   lookupFn,
+		chunkCache: chunkCache,
+		sections:   make(map[SectionIndex]*FileChunkSection),
+	}
+
+	err := group.SetChunks(chunks)
+	return group, err
+}
+
+func (group *ChunkGroup) AddChunk(chunk *filer_pb.FileChunk) error {
+
+	group.sectionsLock.Lock()
+	defer group.sectionsLock.Unlock()
+
+	sectionIndexStart, sectionIndexStop := SectionIndex(chunk.Offset/SectionSize), SectionIndex((chunk.Offset+int64(chunk.Size))/SectionSize)
+	for si := sectionIndexStart; si < sectionIndexStop+1; si++ {
+		section, found := group.sections[si]
+		if !found {
+			section = NewFileChunkSection(si)
+			group.sections[si] = section
+		}
+		section.addChunk(chunk)
+	}
+	return nil
+}
+
+func (group *ChunkGroup) ReadDataAt(fileSize int64, buff []byte, offset int64) (n int, tsNs int64, err error) {
+
+	group.sectionsLock.RLock()
+	defer group.sectionsLock.RUnlock()
+
+	sectionIndexStart, sectionIndexStop := SectionIndex(offset/SectionSize), SectionIndex((offset+int64(len(buff)))/SectionSize)
+	for si := sectionIndexStart; si < sectionIndexStop+1; si++ {
+		section, found := group.sections[si]
+		rangeStart, rangeStop := max(offset, int64(si*SectionSize)), min(offset+int64(len(buff)), int64((si+1)*SectionSize))
+		if !found {
+			for i := rangeStart; i < rangeStop; i++ {
+				buff[i-offset] = 0
+			}
+			continue
+		}
+		xn, xTsNs, xErr := section.readDataAt(group, fileSize, buff[rangeStart-offset:rangeStop-offset], rangeStart)
+		if xErr != nil {
+			err = xErr
+		}
+		n += xn
+		tsNs = max(tsNs, xTsNs)
+	}
+	return
+}
+
+func (group *ChunkGroup) SetChunks(chunks []*filer_pb.FileChunk) error {
+	var dataChunks []*filer_pb.FileChunk
+	for _, chunk := range chunks {
+
+		if !chunk.IsChunkManifest {
+			dataChunks = append(dataChunks, chunk)
+			continue
+		}
+
+		resolvedChunks, err := ResolveOneChunkManifest(group.lookupFn, chunk)
+		if err != nil {
+			return err
+		}
+
+		group.manifestChunks = append(group.manifestChunks, chunk)
+		dataChunks = append(dataChunks, resolvedChunks...)
+	}
+
+	for _, chunk := range dataChunks {
+		sectionIndexStart, sectionIndexStop := SectionIndex(chunk.Offset/SectionSize), SectionIndex((chunk.Offset+int64(chunk.Size))/SectionSize)
+		for si := sectionIndexStart; si < sectionIndexStop+1; si++ {
+			section, found := group.sections[si]
+			if !found {
+				section = NewFileChunkSection(si)
+				group.sections[si] = section
+			}
+			section.chunks = append(section.chunks, chunk)
+		}
+	}
+	return nil
+}
+
+const (
+	// see weedfs_file_lseek.go
+	SEEK_DATA uint32 = 3 // seek to next data after the offset
+	// SEEK_HOLE uint32 = 4 // seek to next hole after the offset
+)
+
+// FIXME: needa tests
+func (group *ChunkGroup) SearchChunks(offset, fileSize int64, whence uint32) (found bool, out int64) {
+	group.sectionsLock.RLock()
+	defer group.sectionsLock.RUnlock()
+
+	return group.doSearchChunks(offset, fileSize, whence)
+}
+
+func (group *ChunkGroup) doSearchChunks(offset, fileSize int64, whence uint32) (found bool, out int64) {
+
+	sectionIndex, maxSectionIndex := SectionIndex(offset/SectionSize), SectionIndex(fileSize/SectionSize)
+	if whence == SEEK_DATA {
+		for si := sectionIndex; si < maxSectionIndex+1; si++ {
+			section, foundSection := group.sections[si]
+			if !foundSection {
+				continue
+			}
+			sectionStart := section.DataStartOffset(group, offset, fileSize)
+			if sectionStart == -1 {
+				continue
+			}
+			return true, sectionStart
+		}
+		return false, 0
+	} else {
+		// whence == SEEK_HOLE
+		for si := sectionIndex; si < maxSectionIndex; si++ {
+			section, foundSection := group.sections[si]
+			if !foundSection {
+				return true, offset
+			}
+			holeStart := section.NextStopOffset(group, offset, fileSize)
+			if holeStart%SectionSize == 0 {
+				continue
+			}
+			return true, holeStart
+		}
+		return true, fileSize
+	}
+}