dwarf/line: Support for parsing multiple file tables

Support multiple file / directory tables for multiple compilation units.

- added a type DebugLines that can hold number of DebugLineInfo
- added a supporting attribute to DebugLineInfo called 'Lookup' which is to be
used to quickly lookup if file exists in FileNames slice
- added supporting methods to lookup and return corresponding DebugLineInfo
- changed the debug_line parsing behavior to read all the available tables and
push them to DebugLines

- since Process.lineInfo is now a slice, it was breaking AllPCsBetween as well
- updated that function's definition to accept a new filename parameter to be
able to extract related DebugLineInfo
- updated calls to AllPCsBetween

- fixed tests that were broken due to attribute type change in Process
- updated _fixtures/cgotest program to include stdio.h, so that it updates
.debug_line header
- added a test to check 'next' in a cgo binary
- OSX - 1.4 does not support cgo, handle that in new testcase

_fixtures/cgotest.go

@@ -1,12 +1,15 @@
 package main
 
 /*
+#include <stdio.h>
 char* foo(void) { return "hello, world!"; }
 */
 import "C"
 
 import "fmt"
+import "runtime"
 
 func main() {
+	runtime.GOMAXPROCS(runtime.NumCPU())
 	fmt.Println(C.GoString(C.foo()))
 }

dwarf/line/line_parser.go

@@ -8,9 +8,9 @@ import (
 )
 
 type DebugLinePrologue struct {
-	Length         uint32
+	UnitLength     uint32
 	Version        uint16
-	PrologueLength uint32
+	Length         uint32
 	MinInstrLength uint8
 	InitialIsStmt  uint8
 	LineBase       int8
@@ -24,6 +24,7 @@ type DebugLineInfo struct {
 	IncludeDirs  []string
 	FileNames    []*FileEntry
 	Instructions []byte
+	Lookup       map[string]*FileEntry
 }
 
 type FileEntry struct {
@@ -33,26 +34,51 @@ type FileEntry struct {
 	Length      uint64
 }
 
-func Parse(data []byte) *DebugLineInfo {
+type DebugLines []*DebugLineInfo
+
+func (d *DebugLines) GetLineInfo(name string) *DebugLineInfo {
+	// Find in which table file exists and return it.
+	for _, l := range *d {
+		if _, ok := l.Lookup[name]; ok {
+			return l
+		}
+	}
+	return nil
+}
+
+func Parse(data []byte) DebugLines {
 	var (
-		dbl = new(DebugLineInfo)
-		buf = bytes.NewBuffer(data)
+		lines = make(DebugLines, 0)
+		buf   = bytes.NewBuffer(data)
 	)
 
-	parseDebugLinePrologue(dbl, buf)
-	parseIncludeDirs(dbl, buf)
-	parseFileEntries(dbl, buf)
-	dbl.Instructions = buf.Bytes()
+	// We have to parse multiple file name tables here.
+	for buf.Len() > 0 {
+		dbl := new(DebugLineInfo)
+		dbl.Lookup = make(map[string]*FileEntry)
 
-	return dbl
+		parseDebugLinePrologue(dbl, buf)
+		parseIncludeDirs(dbl, buf)
+		parseFileEntries(dbl, buf)
+
+		// Instructions size calculation breakdown:
+		//   - dbl.Prologue.UnitLength is the length of the entire unit, not including the 4 bytes to represent that length.
+		//   - dbl.Prologue.Length is the length of the prologue not including unit length, version or prologue length itself.
+		//   - So you have UnitLength - PrologueLength - (version_length_bytes(2) + prologue_length_bytes(4)).
+		dbl.Instructions = buf.Next(int(dbl.Prologue.UnitLength - dbl.Prologue.Length - 6))
+
+		lines = append(lines, dbl)
+	}
+
+	return lines
 }
 
 func parseDebugLinePrologue(dbl *DebugLineInfo, buf *bytes.Buffer) {
 	p := new(DebugLinePrologue)
 
-	p.Length = binary.LittleEndian.Uint32(buf.Next(4))
+	p.UnitLength = binary.LittleEndian.Uint32(buf.Next(4))
 	p.Version = binary.LittleEndian.Uint16(buf.Next(2))
-	p.PrologueLength = binary.LittleEndian.Uint32(buf.Next(4))
+	p.Length = binary.LittleEndian.Uint32(buf.Next(4))
 	p.MinInstrLength = uint8(buf.Next(1)[0])
 	p.InitialIsStmt = uint8(buf.Next(1)[0])
 	p.LineBase = int8(buf.Next(1)[0])
@@ -91,5 +117,6 @@ func parseFileEntries(info *DebugLineInfo, buf *bytes.Buffer) {
 		entry.Length, _ = util.DecodeULEB128(buf)
 
 		info.FileNames = append(info.FileNames, entry)
+		info.Lookup[name] = entry
 	}
 }

dwarf/line/line_parser_test.go

@@ -44,7 +44,8 @@ func TestDebugLinePrologueParser(t *testing.T) {
 	}
 	defer os.Remove(p)
 	data := grabDebugLineSection(p, t)
-	dbl := Parse(data)
+	debugLines := Parse(data)
+	dbl := debugLines[0]
 	prologue := dbl.Prologue
 
 	if prologue.Version != uint16(2) {

dwarf/line/state_machine.go

@@ -74,12 +74,13 @@ func newStateMachine(dbl *DebugLineInfo) *StateMachine {
 
 // Returns all PCs for a given file/line. Useful for loops where the 'for' line
 // could be split amongst 2 PCs.
-func (dbl *DebugLineInfo) AllPCsForFileLine(f string, l int) (pcs []uint64) {
+func (dbl *DebugLines) AllPCsForFileLine(f string, l int) (pcs []uint64) {
 	var (
 		foundFile bool
 		lastAddr  uint64
-		sm        = newStateMachine(dbl)
-		buf       = bytes.NewBuffer(dbl.Instructions)
+		lineInfo  = dbl.GetLineInfo(f)
+		sm        = newStateMachine(lineInfo)
+		buf       = bytes.NewBuffer(lineInfo.Instructions)
 	)
 
 	for b, err := buf.ReadByte(); err == nil; b, err = buf.ReadByte() {
@@ -105,11 +106,12 @@ func (dbl *DebugLineInfo) AllPCsForFileLine(f string, l int) (pcs []uint64) {
 	return
 }
 
-func (dbl *DebugLineInfo) AllPCsBetween(begin, end uint64) []uint64 {
+func (dbl *DebugLines) AllPCsBetween(begin, end uint64, filename string) []uint64 {
+	lineInfo := dbl.GetLineInfo(filename)
 	var (
 		pcs []uint64
-		sm  = newStateMachine(dbl)
-		buf = bytes.NewBuffer(dbl.Instructions)
+		sm  = newStateMachine(lineInfo)
+		buf = bytes.NewBuffer(lineInfo.Instructions)
 	)
 
 	for b, err := buf.ReadByte(); err == nil; b, err = buf.ReadByte() {

proc/proc.go

@@ -38,7 +38,7 @@ type Process struct {
 	dwarf                   *dwarf.Data
 	goSymTable              *gosym.Table
 	frameEntries            frame.FrameDescriptionEntries
-	lineInfo                *line.DebugLineInfo
+	lineInfo                line.DebugLines
 	firstStart              bool
 	os                      *OSProcessDetails
 	arch                    Arch

proc/proc_test.go

@@ -495,6 +495,33 @@ func TestSwitchThread(t *testing.T) {
 	})
 }
 
+func TestCGONext(t *testing.T) {
+	// Test if one can do 'next' in a cgo binary
+	// On OSX with Go < 1.5 CGO is not supported due to: https://github.com/golang/go/issues/8973
+	if runtime.GOOS == "darwin" && strings.Contains(runtime.Version(), "1.4") {
+		return
+	}
+
+	withTestProcess("cgotest", t, func(p *Process, fixture protest.Fixture) {
+		pc, err := p.FindFunctionLocation("main.main", true, 0)
+		if err != nil {
+			t.Fatal(err)
+		}
+		_, err = p.SetBreakpoint(pc)
+		if err != nil {
+			t.Fatal(err)
+		}
+		err = p.Continue()
+		if err != nil {
+			t.Fatal(err)
+		}
+		err = p.Next()
+		if err != nil {
+			t.Fatal(err)
+		}
+	})
+}
+
 type loc struct {
 	line int
 	fn   string

proc/threads.go

@@ -168,7 +168,7 @@ func (thread *Thread) setNextBreakpoints() (err error) {
 	if filepath.Ext(loc.File) == ".go" {
 		err = thread.next(curpc, fde, loc.File, loc.Line)
 	} else {
-		err = thread.cnext(curpc, fde)
+		err = thread.cnext(curpc, fde, loc.File)
 	}
 	return err
 }
@@ -226,8 +226,8 @@ func (thread *Thread) next(curpc uint64, fde *frame.FrameDescriptionEntry, file
 // Set a breakpoint at every reachable location, as well as the return address. Without
 // the benefit of an AST we can't be sure we're not at a branching statement and thus
 // cannot accurately predict where we may end up.
-func (thread *Thread) cnext(curpc uint64, fde *frame.FrameDescriptionEntry) error {
-	pcs := thread.dbp.lineInfo.AllPCsBetween(fde.Begin(), fde.End())
+func (thread *Thread) cnext(curpc uint64, fde *frame.FrameDescriptionEntry, file string) error {
+	pcs := thread.dbp.lineInfo.AllPCsBetween(fde.Begin(), fde.End(), file)
 	ret, err := thread.ReturnAddress()
 	if err != nil {
 		return err