gsh-digital-services/seaweedFS
5
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

more solid weed mount (#4089)

Browse Source 
* 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
This commit is contained in:
Chris Lu
2023-01-02 23:20:45 -08:00
committed by GitHub
parent 367353b936
commit d4566d4aaa
45 changed files with 1834 additions and 805 deletions
Download Patch File Download Diff File Expand all files Collapse all files

43
weed/mount/filehandle_read.go
View File

@@ -17,18 +17,20 @@ func (fh *FileHandle) unlockForRead(startOffset int64, size int) {
fh.dirtyPages.UnlockForRead(startOffset, startOffset+int64(size))
}
func (fh *FileHandle) readFromDirtyPages(buff []byte, startOffset int64) (maxStop int64) {
maxStop = fh.dirtyPages.ReadDirtyDataAt(buff, startOffset)
func (fh *FileHandle) readFromDirtyPages(buff []byte, startOffset int64, tsNs int64) (maxStop int64) {
maxStop = fh.dirtyPages.ReadDirtyDataAt(buff, startOffset, tsNs)
return
}
func (fh *FileHandle) readFromChunks(buff []byte, offset int64) (int64, error) {
func (fh *FileHandle) readFromChunks(buff []byte, offset int64) (int64, int64, error) {
fh.entryLock.Lock()
defer fh.entryLock.Unlock()
fileFullPath := fh.FullPath()
entry := fh.GetEntry()
if entry == nil {
return 0, io.EOF
return 0, 0, io.EOF
}
if entry.IsInRemoteOnly() {
@@ -36,43 +38,28 @@ func (fh *FileHandle) readFromChunks(buff []byte, offset int64) (int64, error) {
newEntry, err := fh.downloadRemoteEntry(entry)
if err != nil {
glog.V(1).Infof("download remote entry %s: %v", fileFullPath, err)
return 0, err
return 0, 0, err
}
entry = newEntry
}
fileSize := int64(filer.FileSize(entry))
fileSize := int64(entry.Attributes.FileSize)
if fileSize == 0 {
fileSize = int64(filer.FileSize(entry))
}
if fileSize == 0 {
glog.V(1).Infof("empty fh %v", fileFullPath)
return 0, io.EOF
return 0, 0, io.EOF
}
if offset+int64(len(buff)) <= int64(len(entry.Content)) {
totalRead := copy(buff, entry.Content[offset:])
glog.V(4).Infof("file handle read cached %s [%d,%d] %d", fileFullPath, offset, offset+int64(totalRead), totalRead)
return int64(totalRead), nil
return int64(totalRead), 0, nil
}
var chunkResolveErr error
if fh.entryViewCache == nil {
fh.entryViewCache, chunkResolveErr = filer.NonOverlappingVisibleIntervals(fh.wfs.LookupFn(), entry.GetChunks(), 0, fileSize)
if chunkResolveErr != nil {
return 0, fmt.Errorf("fail to resolve chunk manifest: %v", chunkResolveErr)
}
fh.CloseReader()
}
if fh.reader == nil {
chunkViews := filer.ViewFromVisibleIntervals(fh.entryViewCache, 0, fileSize)
glog.V(4).Infof("file handle read %s [%d,%d) from %d views", fileFullPath, offset, offset+int64(len(buff)), len(chunkViews))
for _, chunkView := range chunkViews {
glog.V(4).Infof(" read %s [%d,%d) from chunk %+v", fileFullPath, chunkView.LogicOffset, chunkView.LogicOffset+int64(chunkView.Size), chunkView.FileId)
}
fh.reader = filer.NewChunkReaderAtFromClient(fh.wfs.LookupFn(), chunkViews, fh.wfs.chunkCache, fileSize)
}
totalRead, err := fh.reader.ReadAt(buff, offset)
totalRead, ts, err := fh.entryChunkGroup.ReadDataAt(fileSize, buff, offset)
if err != nil && err != io.EOF {
glog.Errorf("file handle read %s: %v", fileFullPath, err)
@@ -80,7 +67,7 @@ func (fh *FileHandle) readFromChunks(buff []byte, offset int64) (int64, error) {
// glog.V(4).Infof("file handle read %s [%d,%d] %d : %v", fileFullPath, offset, offset+int64(totalRead), totalRead, err)
return int64(totalRead), err
return int64(totalRead), ts, err
}
func (fh *FileHandle) downloadRemoteEntry(entry *filer_pb.Entry) (*filer_pb.Entry, error) {