Merge pull request #660 from kakkoyun/refactor_debugfile

symbolizer: Introduce DebugInfo file

go.mod

@@ -17,15 +17,15 @@ require (
 	github.com/go-kit/log v0.2.0
 	github.com/go-ozzo/ozzo-validation/v4 v4.3.0
 	github.com/goburrow/cache v0.1.4
-	github.com/google/pprof v0.0.0-20211214055906-6f57359322fd
+	github.com/google/pprof v0.0.0-20220218203455-0368bd9e19a7
 	github.com/google/uuid v1.3.0
 	github.com/grpc-ecosystem/go-grpc-middleware v1.3.0
 	github.com/grpc-ecosystem/go-grpc-middleware/providers/kit/v2 v2.0.0-20201002093600-73cf2ae9d891
 	github.com/grpc-ecosystem/go-grpc-middleware/v2 v2.0.0-rc.2.0.20201207153454-9f6bf00c00a7
 	github.com/grpc-ecosystem/go-grpc-prometheus v1.2.0
 	github.com/grpc-ecosystem/grpc-gateway/v2 v2.7.3
 	github.com/hashicorp/go-multierror v1.1.1
-	github.com/ianlancetaylor/demangle v0.0.0-20211126204342-3ad08eb09c01
+	github.com/ianlancetaylor/demangle v0.0.0-20220203202831-b7f99f1dbc96
 	github.com/improbable-eng/grpc-web v0.15.0
 	github.com/oklog/run v1.1.0
 	github.com/prometheus/client_golang v1.12.1
@@ -41,6 +41,7 @@ require (
 	go.uber.org/atomic v1.9.0
 	golang.org/x/net v0.0.0-20220105145211-5b0dc2dfae98
 	golang.org/x/sync v0.0.0-20210220032951-036812b2e83c
+	golang.org/x/sys v0.0.0-20220209214540-3681064d5158 // indirect
 	google.golang.org/genproto v0.0.0-20220118154757-00ab72f36ad5
 	google.golang.org/grpc v1.44.0
 	google.golang.org/grpc/cmd/protoc-gen-go-grpc v1.2.0

go.sum

@@ -961,8 +961,9 @@ github.com/google/pprof v0.0.0-20210601050228-01bbb1931b22/go.mod h1:kpwsk12EmLe
 github.com/google/pprof v0.0.0-20210609004039-a478d1d731e9/go.mod h1:kpwsk12EmLew5upagYY7GY0pfYCcupk39gWOCRROcvE=
 github.com/google/pprof v0.0.0-20210720184732-4bb14d4b1be1/go.mod h1:kpwsk12EmLew5upagYY7GY0pfYCcupk39gWOCRROcvE=
 github.com/google/pprof v0.0.0-20211122183932-1daafda22083/go.mod h1:KgnwoLYCZ8IQu3XUZ8Nc/bM9CCZFOyjUNOSygVozoDg=
-github.com/google/pprof v0.0.0-20211214055906-6f57359322fd h1:1FjCyPC+syAzJ5/2S8fqdZK1R22vvA0J7JZKcuOIQ7Y=
 github.com/google/pprof v0.0.0-20211214055906-6f57359322fd/go.mod h1:KgnwoLYCZ8IQu3XUZ8Nc/bM9CCZFOyjUNOSygVozoDg=
+github.com/google/pprof v0.0.0-20220218203455-0368bd9e19a7 h1:IR7vKogOtQRR3JhSl1UftKpFLnuJ2o+woqCFXqJ7wXw=
+github.com/google/pprof v0.0.0-20220218203455-0368bd9e19a7/go.mod h1:KgnwoLYCZ8IQu3XUZ8Nc/bM9CCZFOyjUNOSygVozoDg=
 github.com/google/renameio v0.1.0/go.mod h1:KWCgfxg9yswjAJkECMjeO8J8rahYeXnNhOm40UhjYkI=
 github.com/google/uuid v1.0.0/go.mod h1:TIyPZe4MgqvfeYDBFedMoGGpEw/LqOeaOT+nhxU+yHo=
 github.com/google/uuid v1.1.1/go.mod h1:TIyPZe4MgqvfeYDBFedMoGGpEw/LqOeaOT+nhxU+yHo=
@@ -1117,8 +1118,8 @@ github.com/iancoleman/strcase v0.2.0/go.mod h1:iwCmte+B7n89clKwxIoIXy/HfoL7AsD47
 github.com/ianlancetaylor/demangle v0.0.0-20181102032728-5e5cf60278f6/go.mod h1:aSSvb/t6k1mPoxDqO4vJh6VOCGPwU4O0C2/Eqndh1Sc=
 github.com/ianlancetaylor/demangle v0.0.0-20200824232613-28f6c0f3b639/go.mod h1:aSSvb/t6k1mPoxDqO4vJh6VOCGPwU4O0C2/Eqndh1Sc=
 github.com/ianlancetaylor/demangle v0.0.0-20210905161508-09a460cdf81d/go.mod h1:aYm2/VgdVmcIU8iMfdMvDMsRAQjcfZSKFby6HOFvi/w=
-github.com/ianlancetaylor/demangle v0.0.0-20211126204342-3ad08eb09c01 h1:+0qIm4/XbPn2PYkj6QM6CX/FJN5DGvFOaMkSyB1xuh8=
-github.com/ianlancetaylor/demangle v0.0.0-20211126204342-3ad08eb09c01/go.mod h1:aYm2/VgdVmcIU8iMfdMvDMsRAQjcfZSKFby6HOFvi/w=
+github.com/ianlancetaylor/demangle v0.0.0-20220203202831-b7f99f1dbc96 h1:OOB/uHFWWV4Ek3aRFgRaVwSw5IuQw4csg4mKF6fCbN4=
+github.com/ianlancetaylor/demangle v0.0.0-20220203202831-b7f99f1dbc96/go.mod h1:aYm2/VgdVmcIU8iMfdMvDMsRAQjcfZSKFby6HOFvi/w=
 github.com/imdario/mergo v0.3.4/go.mod h1:2EnlNZ0deacrJVfApfmtdGgDfMuh/nq6Ok1EcJh5FfA=
 github.com/imdario/mergo v0.3.5/go.mod h1:2EnlNZ0deacrJVfApfmtdGgDfMuh/nq6Ok1EcJh5FfA=
 github.com/imdario/mergo v0.3.8/go.mod h1:2EnlNZ0deacrJVfApfmtdGgDfMuh/nq6Ok1EcJh5FfA=
@@ -2276,8 +2277,9 @@ golang.org/x/sys v0.0.0-20211019181941-9d821ace8654/go.mod h1:oPkhp1MJrh7nUepCBc
 golang.org/x/sys v0.0.0-20211025201205-69cdffdb9359/go.mod h1:oPkhp1MJrh7nUepCBck5+mAzfO9JrbApNNgaTdGDITg=
 golang.org/x/sys v0.0.0-20211117180635-dee7805ff2e1/go.mod h1:oPkhp1MJrh7nUepCBck5+mAzfO9JrbApNNgaTdGDITg=
 golang.org/x/sys v0.0.0-20211124211545-fe61309f8881/go.mod h1:oPkhp1MJrh7nUepCBck5+mAzfO9JrbApNNgaTdGDITg=
-golang.org/x/sys v0.0.0-20220114195835-da31bd327af9 h1:XfKQ4OlFl8okEOr5UvAqFRVj8pY/4yfcXrddB8qAbU0=
 golang.org/x/sys v0.0.0-20220114195835-da31bd327af9/go.mod h1:oPkhp1MJrh7nUepCBck5+mAzfO9JrbApNNgaTdGDITg=
+golang.org/x/sys v0.0.0-20220209214540-3681064d5158 h1:rm+CHSpPEEW2IsXUib1ThaHIjuBVZjxNgSKmBLFfD4c=
+golang.org/x/sys v0.0.0-20220209214540-3681064d5158/go.mod h1:oPkhp1MJrh7nUepCBck5+mAzfO9JrbApNNgaTdGDITg=
 golang.org/x/term v0.0.0-20201117132131-f5c789dd3221/go.mod h1:Nr5EML6q2oocZ2LXRh80K7BxOlk5/8JxuGnuhpl+muw=
 golang.org/x/term v0.0.0-20201126162022-7de9c90e9dd1/go.mod h1:bj7SfCRtBDWHUb9snDiAeCFNEtKQo2Wmx5Cou7ajbmo=
 golang.org/x/term v0.0.0-20210220032956-6a3ed077a48d h1:SZxvLBoTP5yHO3Frd4z4vrF+DBX9vMVanchswa69toE=

pkg/symbol/addr2line/dwarf.go

@@ -14,149 +14,47 @@
 package addr2line
 
 import (
-	"debug/dwarf"
 	"errors"
 	"fmt"
-	"io"
-	"sort"
 
-	"github.com/go-delve/delve/pkg/dwarf/godwarf"
-	"github.com/go-delve/delve/pkg/dwarf/reader"
 	"github.com/go-kit/log"
 	"github.com/go-kit/log/level"
 
 	pb "github.com/parca-dev/parca/gen/proto/go/parca/metastore/v1alpha1"
-	"github.com/parca-dev/parca/internal/go/debug/elf"
 	"github.com/parca-dev/parca/pkg/metastore"
 	"github.com/parca-dev/parca/pkg/symbol/demangle"
+	"github.com/parca-dev/parca/pkg/symbol/elfutils"
 )
 
 var ErrLocationFailedBefore = errors.New("failed to symbolize location, attempts are exhausted")
 
 type DwarfLiner struct {
-	logger    log.Logger
-	demangler *demangle.Demangler
+	logger log.Logger
 
-	mapping             *pb.Mapping
-	data                *dwarf.Data
-	lineEntries         map[dwarf.Offset][]dwarf.LineEntry
-	subprograms         map[dwarf.Offset][]*godwarf.Tree
-	abstractSubprograms map[dwarf.Offset]*dwarf.Entry
+	dbgFile elfutils.DebugInfoFile
 
 	attemptThreshold int
 	attempts         map[uint64]int
 	failed           map[uint64]struct{}
 }
 
-func DWARF(logger log.Logger, demangler *demangle.Demangler, attemptThreshold int, m *pb.Mapping, path string) (*DwarfLiner, error) {
-	// TODO(kakkoyun): Handle offset, start and limit for dynamically linked libraries.
-	//f, err := s.bu.Open(file, m.Start, m.Limit, m.Offset)
-	//if err != nil {
-	//	return nil, fmt.Errorf("open object file: %w", err)
-	//}
-	f, err := elf.Open(path)
+// DWARF is a symbolizer that uses DWARF debug info to symbolize addresses.
+func DWARF(logger log.Logger, path string, m *pb.Mapping, demangler *demangle.Demangler, attemptThreshold int) (*DwarfLiner, error) {
+	dbgFile, err := elfutils.NewDebugInfoFile(path, m, demangler)
 	if err != nil {
-		return nil, fmt.Errorf("failed to open elf: %w", err)
-	}
-	defer f.Close()
-
-	data, err := f.DWARF()
-	if err != nil {
-		return nil, fmt.Errorf("failed to read DWARF data: %w", err)
+		return nil, err
 	}
 
 	return &DwarfLiner{
-		logger:    logger,
-		demangler: demangler,
-
-		mapping: m,
-		data:    data,
-
-		lineEntries:         map[dwarf.Offset][]dwarf.LineEntry{},
-		subprograms:         map[dwarf.Offset][]*godwarf.Tree{},
-		abstractSubprograms: map[dwarf.Offset]*dwarf.Entry{},
+		logger:  logger,
+		dbgFile: dbgFile,
 
 		attemptThreshold: attemptThreshold,
 		attempts:         map[uint64]int{},
 		failed:           map[uint64]struct{}{},
 	}, nil
 }
 
-func (dl *DwarfLiner) ensureLookUpTablesBuilt(cu *dwarf.Entry) error {
-	if _, ok := dl.lineEntries[cu.Offset]; ok {
-		// Already created.
-		return nil
-	}
-
-	// The reader is positioned at byte offset 0 in the DWARF “line” section.
-	lr, err := dl.data.LineReader(cu)
-	if err != nil {
-		return err
-	}
-	if lr == nil {
-		return errors.New("failed to initialize line reader")
-	}
-
-	for {
-		le := dwarf.LineEntry{}
-		err := lr.Next(&le)
-		if err != nil {
-			break
-		}
-		if le.IsStmt {
-			dl.lineEntries[cu.Offset] = append(dl.lineEntries[cu.Offset], le)
-		}
-	}
-
-	er := dl.data.Reader()
-	// The reader is positioned at byte offset of compile unit in the DWARF “info” section.
-	er.Seek(cu.Offset)
-	entry, err := er.Next()
-	if err != nil || entry == nil {
-		return errors.New("failed to read entry for compile unit")
-	}
-
-	if entry.Tag != dwarf.TagCompileUnit {
-		return errors.New("failed to find entry for compile unit")
-	}
-
-outer:
-	for {
-		entry, err := er.Next()
-		if err != nil {
-			if err == io.EOF {
-				break
-			}
-			continue
-		}
-		if entry == nil {
-			break
-		}
-		if entry.Tag == dwarf.TagCompileUnit {
-			// Reached to another compile unit.
-			break
-		}
-
-		if entry.Tag == dwarf.TagSubprogram {
-			for _, field := range entry.Field {
-				if field.Attr == dwarf.AttrInline {
-					dl.abstractSubprograms[entry.Offset] = entry
-					continue outer
-				}
-			}
-
-			tr, err := godwarf.LoadTree(entry.Offset, dl.data, 0)
-			if err != nil {
-				return fmt.Errorf("failed to extract dwarf tree: %w", err)
-			}
-
-			dl.subprograms[cu.Offset] = append(dl.subprograms[cu.Offset], tr)
-		}
-	}
-
-	return nil
-}
-
 func (dl *DwarfLiner) PCToLines(addr uint64) (lines []metastore.LocationLine, err error) {
 	// Check if we already attempt to symbolize this location and failed.
 	if _, failedBefore := dl.failed[addr]; failedBefore {
@@ -170,7 +68,7 @@ func (dl *DwarfLiner) PCToLines(addr uint64) (lines []metastore.LocationLine, er
 		}
 	}()
 
-	lines, err = dl.sourceLines(addr)
+	lines, err = dl.dbgFile.SourceLines(addr)
 	if err != nil {
 		return nil, dl.handleError(addr, err)
 	}
@@ -198,93 +96,3 @@ func (dl *DwarfLiner) handleError(addr uint64, err error) error {
 	dl.attempts[addr] = 1
 	return err
 }
-
-func (dl *DwarfLiner) sourceLines(addr uint64) ([]metastore.LocationLine, error) {
-	// The reader is positioned at byte offset 0 in the DWARF “info” section.
-	er := dl.data.Reader()
-	cu, err := er.SeekPC(addr)
-	if err != nil {
-		return nil, err
-	}
-	if cu == nil {
-		return nil, errors.New("failed to find a corresponding dwarf entry for given address")
-	}
-
-	if err := dl.ensureLookUpTablesBuilt(cu); err != nil {
-		return nil, err
-	}
-
-	lines := []metastore.LocationLine{}
-	var tr *godwarf.Tree
-	for _, t := range dl.subprograms[cu.Offset] {
-		if t.ContainsPC(addr) {
-			tr = t
-			break
-		}
-	}
-	if tr == nil {
-		return lines, nil
-	}
-
-	name := tr.Entry.Val(dwarf.AttrName).(string)
-	file, line := findLineInfo(dl.lineEntries[cu.Offset], tr.Ranges)
-	lines = append(lines, metastore.LocationLine{
-		Line: line,
-		Function: dl.demangler.Demangle(&pb.Function{
-			Name:     name,
-			Filename: file,
-		}),
-	})
-
-	// If pc is 0 then all inlined calls will be returned.
-	for _, ch := range reader.InlineStack(tr, addr) {
-		var name string
-		if ch.Tag == dwarf.TagSubprogram {
-			name = tr.Entry.Val(dwarf.AttrName).(string)
-		} else {
-			abstractOrigin := dl.abstractSubprograms[ch.Entry.Val(dwarf.AttrAbstractOrigin).(dwarf.Offset)]
-			name = getFunctionName(abstractOrigin)
-		}
-
-		file, line := findLineInfo(dl.lineEntries[cu.Offset], ch.Ranges)
-		lines = append(lines, metastore.LocationLine{
-			Line: line,
-			Function: dl.demangler.Demangle(&pb.Function{
-				Name:     name,
-				Filename: file,
-			}),
-		})
-	}
-
-	return lines, nil
-}
-
-func findLineInfo(entries []dwarf.LineEntry, rg [][2]uint64) (string, int64) {
-	file := "?"
-	var line int64
-	i := sort.Search(len(entries), func(i int) bool {
-		return entries[i].Address >= rg[0][0]
-	})
-	if i >= len(entries) {
-		return file, line
-	}
-
-	le := dwarf.LineEntry{}
-	pc := entries[i].Address
-	if rg[0][0] <= pc && pc < rg[0][1] {
-		le = entries[i]
-		return le.File.Name, int64(le.Line)
-	}
-
-	return file, line
-}
-
-func getFunctionName(entry *dwarf.Entry) string {
-	var name string
-	for _, field := range entry.Field {
-		if field.Attr == dwarf.AttrName {
-			name = field.Val.(string)
-		}
-	}
-	return name
-}

pkg/symbol/addr2line/go.go

@@ -52,6 +52,7 @@ func (gl *GoLiner) PCToLines(addr uint64) (lines []metastore.LocationLine, err e
 		}
 	}()
 
+	// TODO(kakkoyun): Do we need to consider the base address for any part of Go binaries?
 	file, line, fn := gl.symtab.PCToLine(addr)
 	name := "?"
 	if fn != nil {
@@ -61,7 +62,8 @@ func (gl *GoLiner) PCToLines(addr uint64) (lines []metastore.LocationLine, err e
 		line = 0
 	}
 
-	// TODO(kakkoyun): Find a way to symbolize inline functions.
+	// TODO(kakkoyun): These lines miss the inline functions.
+	// - Find a way to symbolize inline functions.
 	lines = append(lines, metastore.LocationLine{
 		Line: int64(line),
 		Function: &pb.Function{