[dev.garbage] Merge remote-tracking branch 'origin/master' into HEAD

Change-Id: I282fd9ce9db435dfd35e882a9502ab1abc185297

5 files changed, 334 insertions, 158 deletions

diff --git a/src/debug/pe/file.go b/src/debug/pe/file.go
index 1acc368e1b..1cd84d5727 100644
--- a/src/debug/pe/file.go
+++ b/src/debug/pe/file.go
@@ -8,11 +8,9 @@ package pe
 import (
 	"debug/dwarf"
 	"encoding/binary"
-	"errors"
 	"fmt"
 	"io"
 	"os"
-	"strconv"
 )
 
 // A File represents an open PE file.
@@ -20,83 +18,13 @@ type File struct {
 	FileHeader
 	OptionalHeader interface{} // of type *OptionalHeader32 or *OptionalHeader64
 	Sections       []*Section
-	Symbols        []*Symbol
+	Symbols        []*Symbol    // COFF symbols with auxiliary symbol records removed
+	COFFSymbols    []COFFSymbol // all COFF symbols (including auxiliary symbol records)
+	StringTable    StringTable
 
 	closer io.Closer
 }
 
-type SectionHeader struct {
-	Name                 string
-	VirtualSize          uint32
-	VirtualAddress       uint32
-	Size                 uint32
-	Offset               uint32
-	PointerToRelocations uint32
-	PointerToLineNumbers uint32
-	NumberOfRelocations  uint16
-	NumberOfLineNumbers  uint16
-	Characteristics      uint32
-}
-
-type Section struct {
-	SectionHeader
-
-	// Embed ReaderAt for ReadAt method.
-	// Do not embed SectionReader directly
-	// to avoid having Read and Seek.
-	// If a client wants Read and Seek it must use
-	// Open() to avoid fighting over the seek offset
-	// with other clients.
-	io.ReaderAt
-	sr *io.SectionReader
-}
-
-type Symbol struct {
-	Name          string
-	Value         uint32
-	SectionNumber int16
-	Type          uint16
-	StorageClass  uint8
-}
-
-type ImportDirectory struct {
-	OriginalFirstThunk uint32
-	TimeDateStamp      uint32
-	ForwarderChain     uint32
-	Name               uint32
-	FirstThunk         uint32
-
-	dll string
-}
-
-// Data reads and returns the contents of the PE section.
-func (s *Section) Data() ([]byte, error) {
-	dat := make([]byte, s.sr.Size())
-	n, err := s.sr.ReadAt(dat, 0)
-	if n == len(dat) {
-		err = nil
-	}
-	return dat[0:n], err
-}
-
-// Open returns a new ReadSeeker reading the PE section.
-func (s *Section) Open() io.ReadSeeker { return io.NewSectionReader(s.sr, 0, 1<<63-1) }
-
-type FormatError struct {
-	off int64
-	msg string
-	val interface{}
-}
-
-func (e *FormatError) Error() string {
-	msg := e.msg
-	if e.val != nil {
-		msg += fmt.Sprintf(" '%v'", e.val)
-	}
-	msg += fmt.Sprintf(" in record at byte %#x", e.off)
-	return msg
-}
-
 // Open opens the named file using os.Open and prepares it for use as a PE binary.
 func Open(name string) (*File, error) {
 	f, err := os.Open(name)
@@ -129,6 +57,8 @@ var (
 	sizeofOptionalHeader64 = uint16(binary.Size(OptionalHeader64{}))
 )
 
+// TODO(brainman): add Load function, as a replacement for NewFile, that does not call removeAuxSymbols (for performance)
+
 // NewFile creates a new File for accessing a PE binary in an underlying reader.
 func NewFile(r io.ReaderAt) (*File, error) {
 	f := new(File)
@@ -144,66 +74,42 @@ func NewFile(r io.ReaderAt) (*File, error) {
 		var sign [4]byte
 		r.ReadAt(sign[:], signoff)
 		if !(sign[0] == 'P' && sign[1] == 'E' && sign[2] == 0 && sign[3] == 0) {
-			return nil, errors.New("Invalid PE File Format.")
+			return nil, fmt.Errorf("Invalid PE COFF file signature of %v.", sign)
 		}
 		base = signoff + 4
 	} else {
 		base = int64(0)
 	}
-	sr.Seek(base, os.SEEK_SET)
+	sr.Seek(base, io.SeekStart)
 	if err := binary.Read(sr, binary.LittleEndian, &f.FileHeader); err != nil {
 		return nil, err
 	}
-	if f.FileHeader.Machine != IMAGE_FILE_MACHINE_UNKNOWN && f.FileHeader.Machine != IMAGE_FILE_MACHINE_AMD64 && f.FileHeader.Machine != IMAGE_FILE_MACHINE_I386 {
-		return nil, errors.New("Invalid PE File Format.")
+	switch f.FileHeader.Machine {
+	case IMAGE_FILE_MACHINE_UNKNOWN, IMAGE_FILE_MACHINE_AMD64, IMAGE_FILE_MACHINE_I386:
+	default:
+		return nil, fmt.Errorf("Unrecognised COFF file header machine value of 0x%x.", f.FileHeader.Machine)
 	}
 
-	var ss []byte
-	if f.FileHeader.NumberOfSymbols > 0 {
-		// Get COFF string table, which is located at the end of the COFF symbol table.
-		sr.Seek(int64(f.FileHeader.PointerToSymbolTable+COFFSymbolSize*f.FileHeader.NumberOfSymbols), os.SEEK_SET)
-		var l uint32
-		if err := binary.Read(sr, binary.LittleEndian, &l); err != nil {
-			return nil, err
-		}
-		ss = make([]byte, l)
-		if _, err := r.ReadAt(ss, int64(f.FileHeader.PointerToSymbolTable+COFFSymbolSize*f.FileHeader.NumberOfSymbols)); err != nil {
-			return nil, err
-		}
+	var err error
 
-		// Process COFF symbol table.
-		sr.Seek(int64(f.FileHeader.PointerToSymbolTable), os.SEEK_SET)
-		aux := uint8(0)
-		for i := 0; i < int(f.FileHeader.NumberOfSymbols); i++ {
-			cs := new(COFFSymbol)
-			if err := binary.Read(sr, binary.LittleEndian, cs); err != nil {
-				return nil, err
-			}
-			if aux > 0 {
-				aux--
-				continue
-			}
-			var name string
-			if cs.Name[0] == 0 && cs.Name[1] == 0 && cs.Name[2] == 0 && cs.Name[3] == 0 {
-				si := int(binary.LittleEndian.Uint32(cs.Name[4:]))
-				name, _ = getString(ss, si)
-			} else {
-				name = cstring(cs.Name[:])
-			}
-			aux = cs.NumberOfAuxSymbols
-			s := &Symbol{
-				Name:          name,
-				Value:         cs.Value,
-				SectionNumber: cs.SectionNumber,
-				Type:          cs.Type,
-				StorageClass:  cs.StorageClass,
-			}
-			f.Symbols = append(f.Symbols, s)
-		}
+	// Read string table.
+	f.StringTable, err = readStringTable(&f.FileHeader, sr)
+	if err != nil {
+		return nil, err
+	}
+
+	// Read symbol table.
+	f.COFFSymbols, err = readCOFFSymbols(&f.FileHeader, sr)
+	if err != nil {
+		return nil, err
+	}
+	f.Symbols, err = removeAuxSymbols(f.COFFSymbols, f.StringTable)
+	if err != nil {
+		return nil, err
 	}
 
 	// Read optional header.
-	sr.Seek(base, os.SEEK_SET)
+	sr.Seek(base, io.SeekStart)
 	if err := binary.Read(sr, binary.LittleEndian, &f.FileHeader); err != nil {
 		return nil, err
 	}
@@ -235,12 +141,9 @@ func NewFile(r io.ReaderAt) (*File, error) {
 		if err := binary.Read(sr, binary.LittleEndian, sh); err != nil {
 			return nil, err
 		}
-		var name string
-		if sh.Name[0] == '\x2F' {
-			si, _ := strconv.Atoi(cstring(sh.Name[1:]))
-			name, _ = getString(ss, si)
-		} else {
-			name = cstring(sh.Name[0:])
+		name, err := sh.fullName(f.StringTable)
+		if err != nil {
+			return nil, err
 		}
 		s := new(Section)
 		s.SectionHeader = SectionHeader{
@@ -259,14 +162,15 @@ func NewFile(r io.ReaderAt) (*File, error) {
 		s.ReaderAt = s.sr
 		f.Sections[i] = s
 	}
-	return f, nil
-}
-
-func cstring(b []byte) string {
-	var i int
-	for i = 0; i < len(b) && b[i] != 0; i++ {
+	for i := range f.Sections {
+		var err error
+		f.Sections[i].Relocs, err = readRelocs(&f.Sections[i].SectionHeader, sr)
+		if err != nil {
+			return nil, err
+		}
 	}
-	return string(b[0:i])
+
+	return f, nil
 }
 
 // getString extracts a string from symbol string table.
@@ -320,6 +224,18 @@ func (f *File) DWARF() (*dwarf.Data, error) {
 	return dwarf.New(abbrev, nil, nil, info, line, nil, ranges, str)
 }
 
+// TODO(brainman): document ImportDirectory once we decide what to do with it.
+
+type ImportDirectory struct {
+	OriginalFirstThunk uint32
+	TimeDateStamp      uint32
+	ForwarderChain     uint32
+	Name               uint32
+	FirstThunk         uint32
+
+	dll string
+}
+
 // ImportedSymbols returns the names of all symbols
 // referred to by the binary f that are expected to be
 // satisfied by other libraries at dynamic load time.
@@ -347,6 +263,12 @@ func (f *File) ImportedSymbols() ([]string, error) {
 		}
 		ida = append(ida, dt)
 	}
+	// TODO(brainman): this needs to be rewritten
+	//  ds.Data() return contets of .idata section. Why store in variable called "names"?
+	//  Why we are retrieving it second time? We already have it in "d", and it is not modified anywhere.
+	//  getString does not extracts a string from symbol string table (as getString doco says).
+	//  Why ds.Data() called again and again in the loop?
+	//  Needs test before rewrite.
 	names, _ := ds.Data()
 	var all []string
 	for _, dt := range ida {
@@ -395,3 +317,12 @@ func (f *File) ImportedLibraries() ([]string, error) {
 	// cgo -dynimport don't use this for windows PE, so just return.
 	return nil, nil
 }
+
+// FormatError is unused.
+// The type is retained for compatibility.
+type FormatError struct {
+}
+
+func (e *FormatError) Error() string {
+	return "unknown error"
+}
diff --git a/src/debug/pe/pe.go b/src/debug/pe/pe.go
index 6af40e2c78..8050d59c70 100644
--- a/src/debug/pe/pe.go
+++ b/src/debug/pe/pe.go
@@ -86,30 +86,6 @@ type OptionalHeader64 struct {
 	DataDirectory               [16]DataDirectory
 }
 
-type SectionHeader32 struct {
-	Name                 [8]uint8
-	VirtualSize          uint32
-	VirtualAddress       uint32
-	SizeOfRawData        uint32
-	PointerToRawData     uint32
-	PointerToRelocations uint32
-	PointerToLineNumbers uint32
-	NumberOfRelocations  uint16
-	NumberOfLineNumbers  uint16
-	Characteristics      uint32
-}
-
-const COFFSymbolSize = 18
-
-type COFFSymbol struct {
-	Name               [8]uint8
-	Value              uint32
-	SectionNumber      int16
-	Type               uint16
-	StorageClass       uint8
-	NumberOfAuxSymbols uint8
-}
-
 const (
 	IMAGE_FILE_MACHINE_UNKNOWN   = 0x0
 	IMAGE_FILE_MACHINE_AM33      = 0x1d3
diff --git a/src/debug/pe/section.go b/src/debug/pe/section.go
new file mode 100644
index 0000000000..5d881577d3
--- /dev/null
+++ b/src/debug/pe/section.go
@@ -0,0 +1,111 @@
+// Copyright 2016 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 pe
+
+import (
+	"encoding/binary"
+	"fmt"
+	"io"
+	"strconv"
+)
+
+// SectionHeader32 represents real PE COFF section header.
+type SectionHeader32 struct {
+	Name                 [8]uint8
+	VirtualSize          uint32
+	VirtualAddress       uint32
+	SizeOfRawData        uint32
+	PointerToRawData     uint32
+	PointerToRelocations uint32
+	PointerToLineNumbers uint32
+	NumberOfRelocations  uint16
+	NumberOfLineNumbers  uint16
+	Characteristics      uint32
+}
+
+// fullName finds real name of section sh. Normally name is stored
+// in sh.Name, but if it is longer then 8 characters, it is stored
+// in COFF string table st instead.
+func (sh *SectionHeader32) fullName(st StringTable) (string, error) {
+	if sh.Name[0] != '/' {
+		return cstring(sh.Name[:]), nil
+	}
+	i, err := strconv.Atoi(cstring(sh.Name[1:]))
+	if err != nil {
+		return "", err
+	}
+	return st.String(uint32(i))
+}
+
+// TODO(brainman): copy all IMAGE_REL_* consts from ldpe.go here
+
+// Reloc represents a PE COFF relocation.
+// Each section contains its own relocation list.
+type Reloc struct {
+	VirtualAddress   uint32
+	SymbolTableIndex uint32
+	Type             uint16
+}
+
+func readRelocs(sh *SectionHeader, r io.ReadSeeker) ([]Reloc, error) {
+	if sh.NumberOfRelocations <= 0 {
+		return nil, nil
+	}
+	_, err := r.Seek(int64(sh.PointerToRelocations), io.SeekStart)
+	if err != nil {
+		return nil, fmt.Errorf("fail to seek to %q section relocations: %v", sh.Name, err)
+	}
+	relocs := make([]Reloc, sh.NumberOfRelocations)
+	err = binary.Read(r, binary.LittleEndian, relocs)
+	if err != nil {
+		return nil, fmt.Errorf("fail to read section relocations: %v", err)
+	}
+	return relocs, nil
+}
+
+// SectionHeader is similar to SectionHeader32 with Name
+// field replaced by Go string.
+type SectionHeader struct {
+	Name                 string
+	VirtualSize          uint32
+	VirtualAddress       uint32
+	Size                 uint32
+	Offset               uint32
+	PointerToRelocations uint32
+	PointerToLineNumbers uint32
+	NumberOfRelocations  uint16
+	NumberOfLineNumbers  uint16
+	Characteristics      uint32
+}
+
+// Section provides access to PE COFF section.
+type Section struct {
+	SectionHeader
+	Relocs []Reloc
+
+	// Embed ReaderAt for ReadAt method.
+	// Do not embed SectionReader directly
+	// to avoid having Read and Seek.
+	// If a client wants Read and Seek it must use
+	// Open() to avoid fighting over the seek offset
+	// with other clients.
+	io.ReaderAt
+	sr *io.SectionReader
+}
+
+// Data reads and returns the contents of the PE section s.
+func (s *Section) Data() ([]byte, error) {
+	dat := make([]byte, s.sr.Size())
+	n, err := s.sr.ReadAt(dat, 0)
+	if n == len(dat) {
+		err = nil
+	}
+	return dat[0:n], err
+}
+
+// Open returns a new ReadSeeker reading the PE section s.
+func (s *Section) Open() io.ReadSeeker {
+	return io.NewSectionReader(s.sr, 0, 1<<63-1)
+}
diff --git a/src/debug/pe/string.go b/src/debug/pe/string.go
new file mode 100644
index 0000000000..e00bd97dd4
--- /dev/null
+++ b/src/debug/pe/string.go
@@ -0,0 +1,63 @@
+// Copyright 2016 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 pe
+
+import (
+	"encoding/binary"
+	"fmt"
+	"io"
+)
+
+// cstring converts ASCII byte sequence b to string.
+// It stops once it finds 0 or reaches end of b.
+func cstring(b []byte) string {
+	var i int
+	for i = 0; i < len(b) && b[i] != 0; i++ {
+	}
+	return string(b[:i])
+}
+
+// StringTable is a COFF string table.
+type StringTable []byte
+
+func readStringTable(fh *FileHeader, r io.ReadSeeker) (StringTable, error) {
+	// COFF string table is located right after COFF symbol table.
+	offset := fh.PointerToSymbolTable + COFFSymbolSize*fh.NumberOfSymbols
+	_, err := r.Seek(int64(offset), io.SeekStart)
+	if err != nil {
+		return nil, fmt.Errorf("fail to seek to string table: %v", err)
+	}
+	var l uint32
+	err = binary.Read(r, binary.LittleEndian, &l)
+	if err != nil {
+		return nil, fmt.Errorf("fail to read string table length: %v", err)
+	}
+	// string table length includes itself
+	if l <= 4 {
+		return nil, nil
+	}
+	l -= 4
+	buf := make([]byte, l)
+	_, err = io.ReadFull(r, buf)
+	if err != nil {
+		return nil, fmt.Errorf("fail to read string table: %v", err)
+	}
+	return StringTable(buf), nil
+}
+
+// TODO(brainman): decide if start parameter should be int instead of uint32
+
+// String extracts string from COFF string table st at offset start.
+func (st StringTable) String(start uint32) (string, error) {
+	// start includes 4 bytes of string table length
+	if start < 4 {
+		return "", fmt.Errorf("offset %d is before the start of string table", start)
+	}
+	start -= 4
+	if int(start) > len(st) {
+		return "", fmt.Errorf("offset %d is beyond the end of string table", start)
+	}
+	return cstring(st[start:]), nil
+}
diff --git a/src/debug/pe/symbol.go b/src/debug/pe/symbol.go
new file mode 100644
index 0000000000..3cf5a101e7
--- /dev/null
+++ b/src/debug/pe/symbol.go
@@ -0,0 +1,95 @@
+// Copyright 2016 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 pe
+
+import (
+	"encoding/binary"
+	"fmt"
+	"io"
+)
+
+const COFFSymbolSize = 18
+
+// COFFSymbol represents single COFF symbol table record.
+type COFFSymbol struct {
+	Name               [8]uint8
+	Value              uint32
+	SectionNumber      int16
+	Type               uint16
+	StorageClass       uint8
+	NumberOfAuxSymbols uint8
+}
+
+func readCOFFSymbols(fh *FileHeader, r io.ReadSeeker) ([]COFFSymbol, error) {
+	if fh.NumberOfSymbols <= 0 {
+		return nil, nil
+	}
+	_, err := r.Seek(int64(fh.PointerToSymbolTable), io.SeekStart)
+	if err != nil {
+		return nil, fmt.Errorf("fail to seek to symbol table: %v", err)
+	}
+	syms := make([]COFFSymbol, fh.NumberOfSymbols)
+	err = binary.Read(r, binary.LittleEndian, syms)
+	if err != nil {
+		return nil, fmt.Errorf("fail to read symbol table: %v", err)
+	}
+	return syms, nil
+}
+
+// isSymNameOffset checks symbol name if it is encoded as offset into string table.
+func isSymNameOffset(name [8]byte) (bool, uint32) {
+	if name[0] == 0 && name[1] == 0 && name[2] == 0 && name[3] == 0 {
+		return true, binary.LittleEndian.Uint32(name[4:])
+	}
+	return false, 0
+}
+
+// FullName finds real name of symbol sym. Normally name is stored
+// in sym.Name, but if it is longer then 8 characters, it is stored
+// in COFF string table st instead.
+func (sym *COFFSymbol) FullName(st StringTable) (string, error) {
+	if ok, offset := isSymNameOffset(sym.Name); ok {
+		return st.String(offset)
+	}
+	return cstring(sym.Name[:]), nil
+}
+
+func removeAuxSymbols(allsyms []COFFSymbol, st StringTable) ([]*Symbol, error) {
+	if len(allsyms) == 0 {
+		return nil, nil
+	}
+	syms := make([]*Symbol, 0)
+	aux := uint8(0)
+	for _, sym := range allsyms {
+		if aux > 0 {
+			aux--
+			continue
+		}
+		name, err := sym.FullName(st)
+		if err != nil {
+			return nil, err
+		}
+		aux = sym.NumberOfAuxSymbols
+		s := &Symbol{
+			Name:          name,
+			Value:         sym.Value,
+			SectionNumber: sym.SectionNumber,
+			Type:          sym.Type,
+			StorageClass:  sym.StorageClass,
+		}
+		syms = append(syms, s)
+	}
+	return syms, nil
+}
+
+// Symbol is similar to COFFSymbol with Name field replaced
+// by Go string. Symbol also does not have NumberOfAuxSymbols.
+type Symbol struct {
+	Name          string
+	Value         uint32
+	SectionNumber int16
+	Type          uint16
+	StorageClass  uint8
+}