cmd/internal/goobj: parse native objects in the archive

Also add HasCGO() to internal/testenv for tests.

Updates #21706

Change-Id: I938188047024052bdb42b3ac1a77708f3c2a6dbb
Reviewed-on: https://go-review.googlesource.com/62591
Run-TryBot: Ian Lance Taylor <iant@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Ian Lance Taylor <iant@golang.org>

src/cmd/internal/goobj/goobj_test.go

@@ -5,8 +5,12 @@
 package goobj
 
 import (
+	"debug/elf"
+	"debug/macho"
+	"debug/pe"
 	"fmt"
 	"internal/testenv"
+	"io"
 	"io/ioutil"
 	"os"
 	"os/exec"
@@ -16,10 +20,11 @@ import (
 )
 
 var (
-	buildDir  string
-	go1obj    string
-	go2obj    string
-	goarchive string
+	buildDir   string
+	go1obj     string
+	go2obj     string
+	goarchive  string
+	cgoarchive string
 )
 
 func TestMain(m *testing.M) {
@@ -47,6 +52,48 @@ func TestMain(m *testing.M) {
 	os.Exit(exit)
 }
 
+func copyDir(dst, src string) error {
+	err := os.MkdirAll(dst, 0777)
+	if err != nil {
+		return err
+	}
+	fis, err := ioutil.ReadDir(src)
+	if err != nil {
+		return err
+	}
+	for _, fi := range fis {
+		err = copyFile(filepath.Join(dst, fi.Name()), filepath.Join(src, fi.Name()))
+		if err != nil {
+			return err
+		}
+	}
+	return nil
+}
+
+func copyFile(dst, src string) (err error) {
+	var s, d *os.File
+	s, err = os.Open(src)
+	if err != nil {
+		return err
+	}
+	defer s.Close()
+	d, err = os.Create(dst)
+	if err != nil {
+		return err
+	}
+	defer func() {
+		e := d.Close()
+		if err == nil {
+			err = e
+		}
+	}()
+	_, err = io.Copy(d, s)
+	if err != nil {
+		return err
+	}
+	return nil
+}
+
 func buildGoobj() error {
 	var err error
 
@@ -80,6 +127,29 @@ func buildGoobj() error {
 		return fmt.Errorf("go tool pack c %s %s %s: %v\n%s", goarchive, go1obj, go2obj, err, out)
 	}
 
+	if testenv.HasCGO() {
+		gopath := filepath.Join(buildDir, "gopath")
+		err = copyDir(filepath.Join(gopath, "src", "mycgo"), filepath.Join("testdata", "mycgo"))
+		if err != nil {
+			return err
+		}
+		cmd := exec.Command(gotool, "install", "mycgo")
+		cmd.Env = append(os.Environ(), "GOPATH="+gopath)
+		out, err = cmd.CombinedOutput()
+		if err != nil {
+			return fmt.Errorf("go install mycgo: %v\n%s", err, out)
+		}
+		pat := filepath.Join(gopath, "pkg", "*", "mycgo.a")
+		ms, err := filepath.Glob(pat)
+		if err != nil {
+			return err
+		}
+		if len(ms) == 0 {
+			return fmt.Errorf("cannot found paths for pattern %s", pat)
+		}
+		cgoarchive = ms[0]
+	}
+
 	return nil
 }
 
@@ -144,3 +214,110 @@ func TestParseArchive(t *testing.T) {
 		t.Errorf(`%s: symbol "mypkg.go2" not found`, path)
 	}
 }
+
+func TestParseCGOArchive(t *testing.T) {
+	testenv.MustHaveCGO(t)
+
+	path := cgoarchive
+
+	f, err := os.Open(path)
+	if err != nil {
+		t.Fatal(err)
+	}
+	defer f.Close()
+
+	p, err := Parse(f, "mycgo")
+	if err != nil {
+		t.Fatal(err)
+	}
+	if p.Arch != runtime.GOARCH {
+		t.Errorf("%s: got %v, want %v", path, p.Arch, runtime.GOARCH)
+	}
+	var found1 bool
+	var found2 bool
+	for _, s := range p.Syms {
+		if s.Name == "mycgo.go1" {
+			found1 = true
+		}
+		if s.Name == "mycgo.go2" {
+			found2 = true
+		}
+	}
+	if !found1 {
+		t.Errorf(`%s: symbol "mycgo.go1" not found`, path)
+	}
+	if !found2 {
+		t.Errorf(`%s: symbol "mycgo.go2" not found`, path)
+	}
+
+	c1 := "c1"
+	c2 := "c2"
+
+	found1 = false
+	found2 = false
+
+	switch runtime.GOOS {
+	case "darwin":
+		c1 = "_" + c1
+		c2 = "_" + c2
+		for _, obj := range p.Native {
+			mf, err := macho.NewFile(obj)
+			if err != nil {
+				t.Fatal(err)
+			}
+			for _, s := range mf.Symtab.Syms {
+				switch s.Name {
+				case c1:
+					found1 = true
+				case c2:
+					found2 = true
+				}
+			}
+		}
+	case "windows":
+		if runtime.GOARCH == "386" {
+			c1 = "_" + c1
+			c2 = "_" + c2
+		}
+		for _, obj := range p.Native {
+			pf, err := pe.NewFile(obj)
+			if err != nil {
+				t.Fatal(err)
+			}
+			for _, s := range pf.Symbols {
+				switch s.Name {
+				case c1:
+					found1 = true
+				case c2:
+					found2 = true
+				}
+			}
+		}
+	default:
+		for _, obj := range p.Native {
+			ef, err := elf.NewFile(obj)
+			if err != nil {
+				t.Fatal(err)
+			}
+			syms, err := ef.Symbols()
+			if err != nil {
+				t.Fatal(err)
+			}
+			for _, s := range syms {
+				switch s.Name {
+				case c1:
+					found1 = true
+				case c2:
+					found2 = true
+				}
+			}
+		}
+	}
+
+	if !found1 {
+		t.Errorf(`%s: symbol %q not found`, path, c1)
+	}
+	if !found2 {
+		t.Errorf(`%s: symbol %q not found`, path, c2)
+	}
+}

src/cmd/internal/goobj/read.go

@@ -6,7 +6,6 @@
 //
 // TODO(rsc): Decide where this package should live. (golang.org/issue/6932)
 // TODO(rsc): Decide the appropriate integer types for various fields.
-// TODO(rsc): Write tests. (File format still up in the air a little.)
 package goobj
 
 import (
@@ -16,6 +15,7 @@ import (
 	"errors"
 	"fmt"
 	"io"
+	"os"
 	"strconv"
 	"strings"
 )
@@ -127,12 +127,13 @@ type InlinedCall struct {
 
 // A Package is a parsed Go object file or archive defining a Go package.
 type Package struct {
-	ImportPath string   // import path denoting this package
-	Imports    []string // packages imported by this package
-	SymRefs    []SymID  // list of symbol names and versions referred to by this pack
-	Syms       []*Sym   // symbols defined by this package
-	MaxVersion int      // maximum Version in any SymID in Syms
-	Arch       string   // architecture
+	ImportPath string        // import path denoting this package
+	Imports    []string      // packages imported by this package
+	SymRefs    []SymID       // list of symbol names and versions referred to by this pack
+	Syms       []*Sym        // symbols defined by this package
+	MaxVersion int           // maximum Version in any SymID in Syms
+	Arch       string        // architecture
+	Native     []io.ReaderAt // native object data (e.g. ELF)
 }
 
 var (
@@ -150,7 +151,7 @@ var (
 type objReader struct {
 	p          *Package
 	b          *bufio.Reader
-	f          io.ReadSeeker
+	f          *os.File
 	err        error
 	offset     int64
 	dataOffset int64
@@ -160,7 +161,7 @@ type objReader struct {
 }
 
 // init initializes r to read package p from f.
-func (r *objReader) init(f io.ReadSeeker, p *Package) {
+func (r *objReader) init(f *os.File, p *Package) {
 	r.f = f
 	r.p = p
 	r.offset, _ = f.Seek(0, io.SeekCurrent)
@@ -185,6 +186,24 @@ func (r *objReader) error(err error) error {
 	return r.err
 }
 
+// peek returns the next n bytes without advancing the reader.
+func (r *objReader) peek(n int) ([]byte, error) {
+	if r.err != nil {
+		return nil, r.err
+	}
+	if r.offset >= r.limit {
+		r.error(io.ErrUnexpectedEOF)
+		return nil, r.err
+	}
+	b, err := r.b.Peek(n)
+	if err != nil {
+		if err != bufio.ErrBufferFull {
+			r.error(err)
+		}
+	}
+	return b, err
+}
+
 // readByte reads and returns a byte from the input file.
 // On I/O error or EOF, it records the error but returns byte 0.
 // A sequence of 0 bytes will eventually terminate any
@@ -322,9 +341,9 @@ func (r *objReader) skip(n int64) {
 	}
 }
 
-// Parse parses an object file or archive from r,
+// Parse parses an object file or archive from f,
 // assuming that its import path is pkgpath.
-func Parse(r io.ReadSeeker, pkgpath string) (*Package, error) {
+func Parse(f *os.File, pkgpath string) (*Package, error) {
 	if pkgpath == "" {
 		pkgpath = `""`
 	}
@@ -332,7 +351,7 @@ func Parse(r io.ReadSeeker, pkgpath string) (*Package, error) {
 	p.ImportPath = pkgpath
 
 	var rd objReader
-	rd.init(r, p)
+	rd.init(f, p)
 	err := rd.readFull(rd.tmp[:8])
 	if err != nil {
 		if err == io.EOF {
@@ -365,9 +384,6 @@ func trimSpace(b []byte) string {
 }
 
 // parseArchive parses a Unix archive of Go object files.
-// TODO(rsc): Need to skip non-Go object files.
-// TODO(rsc): Maybe record table of contents in r.p so that
-// linker can avoid having code to parse archives too.
 func (r *objReader) parseArchive() error {
 	for r.offset < r.limit {
 		if err := r.readFull(r.tmp[:60]); err != nil {
@@ -413,9 +429,19 @@ func (r *objReader) parseArchive() error {
 		default:
 			oldLimit := r.limit
 			r.limit = r.offset + size
-			if err := r.parseObject(nil); err != nil {
-				return fmt.Errorf("parsing archive member %q: %v", name, err)
+
+			p, err := r.peek(8)
+			if err != nil {
+				return err
 			}
+			if bytes.Equal(p, goobjHeader) {
+				if err := r.parseObject(nil); err != nil {
+					return fmt.Errorf("parsing archive member %q: %v", name, err)
+				}
+			} else {
+				r.p.Native = append(r.p.Native, io.NewSectionReader(r.f, r.offset, size))
+			}
+
 			r.skip(r.limit - r.offset)
 			r.limit = oldLimit
 		}

src/cmd/internal/goobj/testdata/mycgo/c1.c

@@ -0,0 +1,9 @@
+// Copyright 2017 The Go Authors. All rights reserved.
+// Use of this source code is governed by a BSD-style
+// license that can be found in the LICENSE file.
+
+#include <stdio.h>
+
+void c1(void) {
+	puts("c1");
+}

src/cmd/internal/goobj/testdata/mycgo/c2.c

@@ -0,0 +1,9 @@
+// Copyright 2017 The Go Authors. All rights reserved.
+// Use of this source code is governed by a BSD-style
+// license that can be found in the LICENSE file.
+
+#include <stdio.h>
+
+void c2(void) {
+	puts("c2");
+}

src/cmd/internal/goobj/testdata/mycgo/go.go

@@ -0,0 +1,5 @@
+package mycgo
+
+// void c1(void);
+// void c2(void);
+import "C"

src/cmd/internal/goobj/testdata/mycgo/go1.go

@@ -0,0 +1,11 @@
+// Copyright 2017 The Go Authors. All rights reserved.
+// Use of this source code is governed by a BSD-style
+// license that can be found in the LICENSE file.
+
+package mycgo
+
+import "fmt"
+
+func go1() {
+	fmt.Println("go1")
+}

src/cmd/internal/goobj/testdata/mycgo/go2.go

@@ -0,0 +1,11 @@
+// Copyright 2017 The Go Authors. All rights reserved.
+// Use of this source code is governed by a BSD-style
+// license that can be found in the LICENSE file.
+
+package mycgo
+
+import "fmt"
+
+func go2() {
+	fmt.Println("go2")
+}

src/internal/testenv/testenv.go

@@ -153,6 +153,11 @@ func MustHaveExternalNetwork(t *testing.T) {
 
 var haveCGO bool
 
+// HasCGO reports whether the current system can use cgo.
+func HasCGO() bool {
+	return haveCGO
+}
+
 // MustHaveCGO calls t.Skip if cgo is not available.
 func MustHaveCGO(t *testing.T) {
 	if !haveCGO {