2
2
mirror of https://github.com/octoleo/restic.git synced 2024-11-29 16:23:59 +00:00

parallel: report progress for StreamTrees

This assigns an id to each tree root and then keeps track of how many
tree loads (i.e. trees referenced for the first time) are pending per
tree root. Once a tree root and its subtrees were fully processed there
are no more pending tree loads and the tree root is reported as
processed.
This commit is contained in:
Michael Eischer 2020-11-17 22:37:57 +01:00 committed by Alexander Neumann
parent 3d6a3e2555
commit 258ce0c1e5
4 changed files with 44 additions and 18 deletions

View File

@ -183,7 +183,7 @@ func copyTree(ctx context.Context, srcRepo restic.Repository, dstRepo restic.Rep
visited := visitedTrees.Has(treeID) visited := visitedTrees.Has(treeID)
visitedTrees.Insert(treeID) visitedTrees.Insert(treeID)
return visited return visited
}) }, nil)
wg.Go(func() error { wg.Go(func() error {
// reused buffer // reused buffer

View File

@ -376,7 +376,7 @@ func (c *Checker) Structure(ctx context.Context, errChan chan<- error) {
c.blobRefs.M.Insert(h) c.blobRefs.M.Insert(h)
c.blobRefs.Unlock() c.blobRefs.Unlock()
return blobReferenced return blobReferenced
}) }, nil)
defer close(errChan) defer close(errChan)
for i := 0; i < defaultParallelism; i++ { for i := 0; i < defaultParallelism; i++ {

View File

@ -27,7 +27,7 @@ func FindUsedBlobs(ctx context.Context, repo TreeLoader, treeID ID, blobs BlobSe
blobs.Insert(h) blobs.Insert(h)
lock.Unlock() lock.Unlock()
return blobReferenced return blobReferenced
}) }, nil)
wg.Go(func() error { wg.Go(func() error {
for tree := range treeStream { for tree := range treeStream {

View File

@ -6,6 +6,7 @@ import (
"sync" "sync"
"github.com/restic/restic/internal/debug" "github.com/restic/restic/internal/debug"
"github.com/restic/restic/internal/ui/progress"
"golang.org/x/sync/errgroup" "golang.org/x/sync/errgroup"
) )
@ -18,14 +19,24 @@ type TreeItem struct {
*Tree *Tree
} }
type trackedTreeItem struct {
TreeItem
rootIdx int
}
type trackedID struct {
ID
rootIdx int
}
// loadTreeWorker loads trees from repo and sends them to out. // loadTreeWorker loads trees from repo and sends them to out.
func loadTreeWorker(ctx context.Context, repo TreeLoader, func loadTreeWorker(ctx context.Context, repo TreeLoader,
in <-chan ID, out chan<- TreeItem) { in <-chan trackedID, out chan<- trackedTreeItem) {
for treeID := range in { for treeID := range in {
tree, err := repo.LoadTree(ctx, treeID) tree, err := repo.LoadTree(ctx, treeID.ID)
debug.Log("load tree %v (%v) returned err: %v", tree, treeID, err) debug.Log("load tree %v (%v) returned err: %v", tree, treeID, err)
job := TreeItem{ID: treeID, Error: err, Tree: tree} job := trackedTreeItem{TreeItem: TreeItem{ID: treeID.ID, Error: err, Tree: tree}, rootIdx: treeID.rootIdx}
select { select {
case <-ctx.Done(): case <-ctx.Done():
@ -35,17 +46,23 @@ func loadTreeWorker(ctx context.Context, repo TreeLoader,
} }
} }
func filterTrees(ctx context.Context, backlog IDs, loaderChan chan<- ID, func filterTrees(ctx context.Context, trees IDs, loaderChan chan<- trackedID,
in <-chan TreeItem, out chan<- TreeItem, skip func(tree ID) bool) { in <-chan trackedTreeItem, out chan<- TreeItem, skip func(tree ID) bool, p *progress.Counter) {
var ( var (
inCh = in inCh = in
outCh chan<- TreeItem outCh chan<- TreeItem
loadCh chan<- ID loadCh chan<- trackedID
job TreeItem job TreeItem
nextTreeID ID nextTreeID trackedID
outstandingLoadTreeJobs = 0 outstandingLoadTreeJobs = 0
) )
rootCounter := make([]int, len(trees))
backlog := make([]trackedID, 0, len(trees))
for idx, id := range trees {
backlog = append(backlog, trackedID{ID: id, rootIdx: idx})
rootCounter[idx] = 1
}
for { for {
if loadCh == nil && len(backlog) > 0 { if loadCh == nil && len(backlog) > 0 {
@ -53,7 +70,11 @@ func filterTrees(ctx context.Context, backlog IDs, loaderChan chan<- ID,
ln := len(backlog) - 1 ln := len(backlog) - 1
nextTreeID, backlog = backlog[ln], backlog[:ln] nextTreeID, backlog = backlog[ln], backlog[:ln]
if skip(nextTreeID) { if skip(nextTreeID.ID) {
rootCounter[nextTreeID.rootIdx]--
if p != nil && rootCounter[nextTreeID.rootIdx] == 0 {
p.Add(1)
}
continue continue
} }
@ -82,6 +103,7 @@ func filterTrees(ctx context.Context, backlog IDs, loaderChan chan<- ID,
} }
outstandingLoadTreeJobs-- outstandingLoadTreeJobs--
rootCounter[j.rootIdx]--
debug.Log("input job tree %v", j.ID) debug.Log("input job tree %v", j.ID)
@ -90,7 +112,7 @@ func filterTrees(ctx context.Context, backlog IDs, loaderChan chan<- ID,
} else if j.Tree == nil { } else if j.Tree == nil {
debug.Log("received job with nil tree pointer: %v (ID %v)", j.Error, j.ID) debug.Log("received job with nil tree pointer: %v (ID %v)", j.Error, j.ID)
// send a new job with the new error instead of the old one // send a new job with the new error instead of the old one
j = TreeItem{ID: j.ID, Error: errors.New("tree is nil and error is nil")} j = trackedTreeItem{TreeItem: TreeItem{ID: j.ID, Error: errors.New("tree is nil and error is nil")}, rootIdx: j.rootIdx}
} else { } else {
subtrees := j.Tree.Subtrees() subtrees := j.Tree.Subtrees()
debug.Log("subtrees for tree %v: %v", j.ID, subtrees) debug.Log("subtrees for tree %v: %v", j.ID, subtrees)
@ -104,11 +126,15 @@ func filterTrees(ctx context.Context, backlog IDs, loaderChan chan<- ID,
debug.Log("tree %v has nil subtree", j.ID) debug.Log("tree %v has nil subtree", j.ID)
continue continue
} }
backlog = append(backlog, id) backlog = append(backlog, trackedID{ID: id, rootIdx: j.rootIdx})
rootCounter[j.rootIdx]++
} }
} }
if p != nil && rootCounter[j.rootIdx] == 0 {
p.Add(1)
}
job = j job = j.TreeItem
outCh = out outCh = out
inCh = nil inCh = nil
@ -122,9 +148,9 @@ func filterTrees(ctx context.Context, backlog IDs, loaderChan chan<- ID,
// StreamTrees iteratively loads the given trees and their subtrees. The skip method // StreamTrees iteratively loads the given trees and their subtrees. The skip method
// is guaranteed to always be called from the same goroutine. // is guaranteed to always be called from the same goroutine.
func StreamTrees(ctx context.Context, wg *errgroup.Group, repo TreeLoader, trees IDs, skip func(tree ID) bool) <-chan TreeItem { func StreamTrees(ctx context.Context, wg *errgroup.Group, repo TreeLoader, trees IDs, skip func(tree ID) bool, p *progress.Counter) <-chan TreeItem {
loaderChan := make(chan ID) loaderChan := make(chan trackedID)
loadedTreeChan := make(chan TreeItem) loadedTreeChan := make(chan trackedTreeItem)
treeStream := make(chan TreeItem) treeStream := make(chan TreeItem)
var loadTreeWg sync.WaitGroup var loadTreeWg sync.WaitGroup
@ -148,7 +174,7 @@ func StreamTrees(ctx context.Context, wg *errgroup.Group, repo TreeLoader, trees
wg.Go(func() error { wg.Go(func() error {
defer close(loaderChan) defer close(loaderChan)
defer close(treeStream) defer close(treeStream)
filterTrees(ctx, trees, loaderChan, loadedTreeChan, treeStream, skip) filterTrees(ctx, trees, loaderChan, loadedTreeChan, treeStream, skip, p)
return nil return nil
}) })
return treeStream return treeStream