Skip to content

Commit

Permalink
Custom inline cache implementation
Browse files Browse the repository at this point in the history
  • Loading branch information
mikejholly committed May 16, 2024
1 parent be3e419 commit 750aa06
Show file tree
Hide file tree
Showing 17 changed files with 812 additions and 236 deletions.
1 change: 1 addition & 0 deletions cache/manager.go
Original file line number Diff line number Diff line change
Expand Up @@ -136,6 +136,7 @@ func (cm *cacheManager) GetByBlob(ctx context.Context, desc ocispecs.Descriptor,
if err != nil {
return nil, err
}

chainID := diffID
blobChainID := imagespecidentity.ChainID([]digest.Digest{desc.Digest, diffID})

Expand Down
5 changes: 3 additions & 2 deletions cache/remotecache/import.go
Original file line number Diff line number Diff line change
Expand Up @@ -302,8 +302,9 @@ type image struct {
Rootfs struct {
DiffIDs []digest.Digest `json:"diff_ids"`
} `json:"rootfs"`
Cache []byte `json:"moby.buildkit.cache.v0"`
History []struct {
Cache []byte `json:"moby.buildkit.cache.v0"`
EarthlyInlineCache []byte `json:"earthly.inlinecache.v0"`
History []struct {
Created *time.Time `json:"created,omitempty"`
CreatedBy string `json:"created_by,omitempty"`
EmptyLayer bool `json:"empty_layer,omitempty"`
Expand Down
239 changes: 239 additions & 0 deletions cache/remotecache/inline.go
Original file line number Diff line number Diff line change
@@ -0,0 +1,239 @@
package remotecache

import (
"context"
"encoding/json"
"fmt"

"github.com/containerd/containerd/content"
"github.com/containerd/containerd/images"
"github.com/containerd/containerd/labels"
v1 "github.com/moby/buildkit/cache/remotecache/v1"
"github.com/moby/buildkit/solver"
"github.com/moby/buildkit/util/bklog"
"github.com/moby/buildkit/util/imageutil"
"github.com/moby/buildkit/util/progress"
"github.com/moby/buildkit/worker"
digest "github.com/opencontainers/go-digest"
ocispecs "github.com/opencontainers/image-spec/specs-go/v1"
"github.com/pkg/errors"
)

// earthlyInlineCacheItem stores a relation between a simple solver cache key &
// a remote image descriptor. Used for inline caching.
type earthlyInlineCacheItem struct {
Key digest.Digest `json:"cacheKey"`
Descriptor digest.Digest `json:"descriptor"`
}

// EarthlyInlineCacheRemotes produces a map of cache keys to remote sources by
// parsing inline-cache metadata from a remote image's config data.
func EarthlyInlineCacheRemotes(ctx context.Context, provider content.Provider, desc ocispecs.Descriptor, w worker.Worker) (map[digest.Digest]*solver.Remote, error) {
dt, err := readBlob(ctx, provider, desc)
if err != nil {
return nil, err
}

manifestType, err := imageutil.DetectManifestBlobMediaType(dt)
if err != nil {
return nil, err
}

layerDone := progress.OneOff(ctx, fmt.Sprintf("inferred cache manifest type: %s", manifestType))
layerDone(nil)

configDesc, err := configDescriptor(dt, manifestType)
if err != nil {
return nil, err
}

if configDesc.Digest != "" {
return nil, errors.New("expected empty digest value")
}

m := map[digest.Digest][]byte{}

if err := allDistributionManifests(ctx, provider, dt, m); err != nil {
return nil, err
}

remotes := map[digest.Digest]*solver.Remote{}

for _, dt := range m {
var m ocispecs.Manifest

if err := json.Unmarshal(dt, &m); err != nil {
return nil, errors.Wrap(err, "failed to unmarshal manifest")
}

if m.Config.Digest == "" || len(m.Layers) == 0 {
continue
}

p, err := content.ReadBlob(ctx, provider, m.Config)
if err != nil {
return nil, errors.Wrap(err, "failed to read blob")
}

var img image

if err := json.Unmarshal(p, &img); err != nil {
return nil, errors.Wrap(err, "failed to unmarshal image")
}

if len(img.Rootfs.DiffIDs) != len(m.Layers) {
bklog.G(ctx).Warnf("invalid image with mismatching manifest and config")
continue
}

if img.EarthlyInlineCache == nil {
continue
}

cacheItems := []earthlyInlineCacheItem{}
if err := json.Unmarshal(img.EarthlyInlineCache, &cacheItems); err != nil {
return nil, errors.Wrap(err, "failed to unmarshal cache items")
}

layers, err := preprocessLayers(img, m)
if err != nil {
return nil, err
}

found := extractRemotes(provider, cacheItems, layers)
for key, remote := range found {
remotes[key] = remote
}
}

return remotes, nil
}

// extractRemotes constructs a list of descriptors--which represent the layer
// chain for the given digest--for each of the items discovered in the inline
// metadata.
func extractRemotes(provider content.Provider, cacheItems []earthlyInlineCacheItem, layers []ocispecs.Descriptor) map[digest.Digest]*solver.Remote {

remotes := map[digest.Digest]*solver.Remote{}

for _, cacheItem := range cacheItems {
descs := []ocispecs.Descriptor{}

found := false
for _, layer := range layers {
descs = append(descs, layer)
if layer.Digest == cacheItem.Descriptor {
found = true
break
}
}

if found {
remote := &solver.Remote{
Descriptors: descs,
Provider: provider,
}

remotes[cacheItem.Key] = remote
}
}

return remotes
}

// preprocessLayers adds custom annotations which are used later when
// reconstructing the ref.
func preprocessLayers(img image, m ocispecs.Manifest) ([]ocispecs.Descriptor, error) {
createdDates, createdMsg, err := parseCreatedLayerInfo(img)
if err != nil {
return nil, err
}

ret := []ocispecs.Descriptor{}

for i, layer := range m.Layers {
if layer.Annotations == nil {
layer.Annotations = map[string]string{}
}

if createdAt := createdDates[i]; createdAt != "" {
layer.Annotations["buildkit/createdat"] = createdAt
}

if createdBy := createdMsg[i]; createdBy != "" {
layer.Annotations["buildkit/description"] = createdBy
}

layer.Annotations[labels.LabelUncompressed] = img.Rootfs.DiffIDs[i].String()

ret = append(ret, layer)
}

return ret, nil
}

// configDescriptor parses and returns the correct manifest for the given manifest type.
func configDescriptor(dt []byte, manifestType string) (ocispecs.Descriptor, error) {
var configDesc ocispecs.Descriptor

switch manifestType {
case images.MediaTypeDockerSchema2ManifestList, ocispecs.MediaTypeImageIndex:
var mfst ocispecs.Index
if err := json.Unmarshal(dt, &mfst); err != nil {
return ocispecs.Descriptor{}, err
}

for _, m := range mfst.Manifests {
if m.MediaType == v1.CacheConfigMediaTypeV0 {
configDesc = m
continue
}
}
case images.MediaTypeDockerSchema2Manifest, ocispecs.MediaTypeImageManifest:
var mfst ocispecs.Manifest
if err := json.Unmarshal(dt, &mfst); err != nil {
return ocispecs.Descriptor{}, err
}

if mfst.Config.MediaType == v1.CacheConfigMediaTypeV0 {
configDesc = mfst.Config
}
default:
return ocispecs.Descriptor{}, errors.Errorf("unsupported or uninferrable manifest type %s", manifestType)
}

return configDesc, nil
}

// allDistributionManifests pulls all manifest data & linked manifests using the provider.
func allDistributionManifests(ctx context.Context, provider content.Provider, dt []byte, m map[digest.Digest][]byte) error {
mt, err := imageutil.DetectManifestBlobMediaType(dt)
if err != nil {
return err
}

switch mt {
case images.MediaTypeDockerSchema2Manifest, ocispecs.MediaTypeImageManifest:
m[digest.FromBytes(dt)] = dt
case images.MediaTypeDockerSchema2ManifestList, ocispecs.MediaTypeImageIndex:
var index ocispecs.Index
if err := json.Unmarshal(dt, &index); err != nil {
return errors.WithStack(err)
}

for _, d := range index.Manifests {
if _, ok := m[d.Digest]; ok {
continue
}
p, err := content.ReadBlob(ctx, provider, d)
if err != nil {
return errors.WithStack(err)
}
if err := allDistributionManifests(ctx, provider, p, m); err != nil {
return err
}
}
}

return nil
}
58 changes: 58 additions & 0 deletions cache/remotecache/registry/inline.go
Original file line number Diff line number Diff line change
@@ -0,0 +1,58 @@
package registry

import (
"context"
"strconv"

"github.com/containerd/containerd/remotes/docker"
"github.com/moby/buildkit/cache/remotecache"
"github.com/moby/buildkit/session"
"github.com/moby/buildkit/solver"
"github.com/moby/buildkit/util/contentutil"
"github.com/moby/buildkit/util/resolver"
"github.com/moby/buildkit/util/resolver/limited"
"github.com/moby/buildkit/worker"
digest "github.com/opencontainers/go-digest"
"github.com/pkg/errors"
)

// EarthlyInlineCacheRemotes fetches a group of remote sources based on values
// discovered in a remote image's inline-cache metadata field.
func EarthlyInlineCacheRemotes(ctx context.Context, sm *session.Manager, w worker.Worker, hosts docker.RegistryHosts, g session.Group, attrs map[string]string) (map[digest.Digest]*solver.Remote, error) {
ref, err := canonicalizeRef(attrs[attrRef])
if err != nil {
return nil, err
}

refString := ref.String()

insecure := false
if v, ok := attrs[attrInsecure]; ok {
val, err := strconv.ParseBool(v)
if err != nil {
return nil, errors.Wrapf(err, "failed to parse %s", attrInsecure)
}
insecure = val
}

scope, hosts := registryConfig(hosts, ref, "pull", insecure)
remote := resolver.DefaultPool.GetResolver(hosts, refString, scope, sm, g)

xref, desc, err := remote.Resolve(ctx, refString)
if err != nil {
return nil, err
}

fetcher, err := remote.Fetcher(ctx, xref)
if err != nil {
return nil, err
}

src := &withDistributionSourceLabel{
Provider: contentutil.FromFetcher(limited.Default.WrapFetcher(fetcher, refString)),
ref: refString,
source: w.ContentStore(),
}

return remotecache.EarthlyInlineCacheRemotes(ctx, src, desc, w)
}
1 change: 1 addition & 0 deletions cache/remotecache/v1/cachestorage.go
Original file line number Diff line number Diff line change
Expand Up @@ -261,6 +261,7 @@ func (cs *cacheResultStorage) Load(ctx context.Context, res solver.CacheResult)
return nil, errors.WithStack(solver.ErrNotFound)
}

// MH: Load from remote is here.
ref, err := cs.w.FromRemote(ctx, item.result)
if err != nil {
return nil, errors.Wrap(err, "failed to load result from remote")
Expand Down
3 changes: 3 additions & 0 deletions cmd/buildkitd/main.go
Original file line number Diff line number Diff line change
Expand Up @@ -822,13 +822,15 @@ func newController(c *cli.Context, cfg *config.Config, shutdownCh chan struct{})
"s3": s3remotecache.ResolveCacheExporterFunc(),
"azblob": azblob.ResolveCacheExporterFunc(),
}

remoteCacheImporterFuncs := map[string]remotecache.ResolveCacheImporterFunc{
"registry": registryremotecache.ResolveCacheImporterFunc(sessionManager, w.ContentStore(), resolverFn),
"local": localremotecache.ResolveCacheImporterFunc(sessionManager),
"gha": gha.ResolveCacheImporterFunc(),
"s3": s3remotecache.ResolveCacheImporterFunc(),
"azblob": azblob.ResolveCacheImporterFunc(),
}

return control.NewController(control.Opt{
SessionManager: sessionManager,
WorkerController: wc,
Expand All @@ -844,6 +846,7 @@ func newController(c *cli.Context, cfg *config.Config, shutdownCh chan struct{})
ContentStore: w.ContentStore(),
HistoryConfig: cfg.History,
RootDir: cfg.Root,
RegistryHosts: resolverFn,
})
}

Expand Down
3 changes: 3 additions & 0 deletions control/control.go
Original file line number Diff line number Diff line change
Expand Up @@ -11,6 +11,7 @@ import (

contentapi "github.com/containerd/containerd/api/services/content/v1"
"github.com/containerd/containerd/content"
"github.com/containerd/containerd/remotes/docker"
"github.com/containerd/containerd/services/content/contentserver"
"github.com/distribution/reference"
"github.com/hashicorp/go-multierror"
Expand Down Expand Up @@ -69,6 +70,7 @@ type Opt struct {
ContentStore *containerdsnapshot.Store
HistoryConfig *config.HistoryConfig
RootDir string
RegistryHosts docker.RegistryHosts
}

type Controller struct { // TODO: ControlService
Expand Down Expand Up @@ -107,6 +109,7 @@ func NewController(opt Opt) (*Controller, error) {
Entitlements: opt.Entitlements,
HistoryQueue: hq,
RootDir: opt.RootDir,
RegistryHosts: opt.RegistryHosts,
})
if err != nil {
return nil, errors.Wrap(err, "failed to create solver")
Expand Down
2 changes: 1 addition & 1 deletion executor/runcexecutor/monitor_stats.go
Original file line number Diff line number Diff line change
Expand Up @@ -53,7 +53,7 @@ func (w *runcExecutor) monitorContainerStats(ctx context.Context, id string, sam
for {
select {
case <-ctx.Done():
bklog.G(ctx).Infof("stats collection context done: %v", ctx.Err())
bklog.G(ctx).Debugf("stats collection context done: %v", ctx.Err())
return
case <-timer.C: // Initial sleep will give container the chance to start.
stats, err := w.runc.Stats(ctx, id)
Expand Down
1 change: 1 addition & 0 deletions exporter/containerimage/exptypes/types.go
Original file line number Diff line number Diff line change
Expand Up @@ -11,6 +11,7 @@ const (
ExporterImageConfigDigestKey = "containerimage.config.digest"
ExporterImageDescriptorKey = "containerimage.descriptor"
ExporterInlineCache = "containerimage.inlinecache"
EarthlyInlineCache = "earthly.inlinecache"
ExporterPlatformsKey = "refs.platforms"
)

Expand Down
Loading

0 comments on commit 750aa06

Please sign in to comment.