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/mount/page_writer/page_chunk_mem.go
+++ b/weed/mount/page_writer/page_chunk_mem.go
@@ -19,6 +19,7 @@ type MemChunk struct {
 	usage           *ChunkWrittenIntervalList
 	chunkSize       int64
 	logicChunkIndex LogicChunkIndex
+	activityScore   *ActivityScore
 }

 func NewMemChunk(logicChunkIndex LogicChunkIndex, chunkSize int64) *MemChunk {
@@ -28,6 +29,7 @@ func NewMemChunk(logicChunkIndex LogicChunkIndex, chunkSize int64) *MemChunk {
 		chunkSize:       chunkSize,
 		buf:             mem.Allocate(int(chunkSize)),
 		usage:           newChunkWrittenIntervalList(),
+		activityScore:   NewActivityScore(),
 	}
 }

@@ -39,29 +41,37 @@ func (mc *MemChunk) FreeResource() {
 	mem.Free(mc.buf)
 }

-func (mc *MemChunk) WriteDataAt(src []byte, offset int64) (n int) {
+func (mc *MemChunk) WriteDataAt(src []byte, offset int64, tsNs int64) (n int) {
 	mc.Lock()
 	defer mc.Unlock()

 	innerOffset := offset % mc.chunkSize
 	n = copy(mc.buf[innerOffset:], src)
-	mc.usage.MarkWritten(innerOffset, innerOffset+int64(n))
+	mc.usage.MarkWritten(innerOffset, innerOffset+int64(n), tsNs)
+	mc.activityScore.MarkWrite()
+
 	return
 }

-func (mc *MemChunk) ReadDataAt(p []byte, off int64) (maxStop int64) {
+func (mc *MemChunk) ReadDataAt(p []byte, off int64, tsNs int64) (maxStop int64) {
 	mc.RLock()
 	defer mc.RUnlock()

 	memChunkBaseOffset := int64(mc.logicChunkIndex) * mc.chunkSize
 	for t := mc.usage.head.next; t != mc.usage.tail; t = t.next {
-		logicStart := max(off, int64(mc.logicChunkIndex)*mc.chunkSize+t.StartOffset)
+		logicStart := max(off, memChunkBaseOffset+t.StartOffset)
 		logicStop := min(off+int64(len(p)), memChunkBaseOffset+t.stopOffset)
 		if logicStart < logicStop {
-			copy(p[logicStart-off:logicStop-off], mc.buf[logicStart-memChunkBaseOffset:logicStop-memChunkBaseOffset])
-			maxStop = max(maxStop, logicStop)
+			if t.TsNs >= tsNs {
+				copy(p[logicStart-off:logicStop-off], mc.buf[logicStart-memChunkBaseOffset:logicStop-memChunkBaseOffset])
+				maxStop = max(maxStop, logicStop)
+			} else {
+				println("read old data1", tsNs-t.TsNs, "ns")
+			}
 		}
 	}
+	mc.activityScore.MarkRead()
+
 	return
 }

@@ -72,11 +82,8 @@ func (mc *MemChunk) IsComplete() bool {
 	return mc.usage.IsComplete(mc.chunkSize)
 }

-func (mc *MemChunk) WrittenSize() int64 {
-	mc.RLock()
-	defer mc.RUnlock()
-
-	return mc.usage.WrittenSize()
+func (mc *MemChunk) ActivityScore() int64 {
+	return mc.activityScore.ActivityScore()
 }

 func (mc *MemChunk) SaveContent(saveFn SaveToStorageFunc) {
@@ -88,7 +95,7 @@ func (mc *MemChunk) SaveContent(saveFn SaveToStorageFunc) {
 	}
 	for t := mc.usage.head.next; t != mc.usage.tail; t = t.next {
 		reader := util.NewBytesReader(mc.buf[t.StartOffset:t.stopOffset])
-		saveFn(reader, int64(mc.logicChunkIndex)*mc.chunkSize+t.StartOffset, t.Size(), func() {
+		saveFn(reader, int64(mc.logicChunkIndex)*mc.chunkSize+t.StartOffset, t.Size(), t.TsNs, func() {
 		})
 	}
 }