debug/dwarf: support for DWARF 3

R=rsc
CC=golang-dev
https://golang.org/cl/7662045

src/pkg/debug/dwarf/buf.go

@@ -13,17 +13,45 @@ import (
 
 // Data buffer being decoded.
 type buf struct {
-	dwarf    *Data
-	order    binary.ByteOrder
-	name     string
-	off      Offset
-	data     []byte
-	addrsize int
-	err      error
+	dwarf  *Data
+	order  binary.ByteOrder
+	format dataFormat
+	name   string
+	off    Offset
+	data   []byte
+	err    error
 }
 
-func makeBuf(d *Data, name string, off Offset, data []byte, addrsize int) buf {
-	return buf{d, d.order, name, off, data, addrsize, nil}
+// Data format, other than byte order.  This affects the handling of
+// certain field formats.
+type dataFormat interface {
+	// DWARF version number.  Zero means unknown.
+	version() int
+
+	// 64-bit DWARF format?
+	dwarf64() (dwarf64 bool, isKnown bool)
+
+	// Size of an address, in bytes.  Zero means unknown.
+	addrsize() int
+}
+
+// Some parts of DWARF have no data format, e.g., abbrevs.
+type unknownFormat struct{}
+
+func (u unknownFormat) version() int {
+	return 0
+}
+
+func (u unknownFormat) dwarf64() (bool, bool) {
+	return false, false
+}
+
+func (u unknownFormat) addrsize() int {
+	return 0
+}
+
+func makeBuf(d *Data, format dataFormat, name string, off Offset, data []byte) buf {
+	return buf{d, d.order, format, name, off, data, nil}
 }
 
 func (b *buf) uint8() uint8 {
@@ -121,7 +149,7 @@ func (b *buf) int() int64 {
 
 // Address-sized uint.
 func (b *buf) addr() uint64 {
-	switch b.addrsize {
+	switch b.format.addrsize() {
 	case 1:
 		return uint64(b.uint8())
 	case 2:

src/pkg/debug/dwarf/entry.go

@@ -40,7 +40,7 @@ func (d *Data) parseAbbrev(off uint32) (abbrevTable, error) {
 	} else {
 		data = data[off:]
 	}
-	b := makeBuf(d, "abbrev", 0, data, 0)
+	b := makeBuf(d, unknownFormat{}, "abbrev", 0, data)
 
 	// Error handling is simplified by the buf getters
 	// returning an endless stream of 0s after an error.
@@ -192,7 +192,21 @@ func (b *buf) entry(atab abbrevTable, ubase Offset) *Entry {
 
 		// reference to other entry
 		case formRefAddr:
-			val = Offset(b.addr())
+			vers := b.format.version()
+			if vers == 0 {
+				b.error("unknown version for DW_FORM_ref_addr")
+			} else if vers == 2 {
+				val = Offset(b.addr())
+			} else {
+				is64, known := b.format.dwarf64()
+				if !known {
+					b.error("unknown size for DW_FORM_ref_addr")
+				} else if is64 {
+					val = Offset(b.uint64())
+				} else {
+					val = Offset(b.uint32())
+				}
+			}
 		case formRef1:
 			val = Offset(b.uint8()) + ubase
 		case formRef2:
@@ -212,7 +226,7 @@ func (b *buf) entry(atab abbrevTable, ubase Offset) *Entry {
 			if b.err != nil {
 				return nil
 			}
-			b1 := makeBuf(b.dwarf, "str", 0, b.dwarf.str, 0)
+			b1 := makeBuf(b.dwarf, unknownFormat{}, "str", 0, b.dwarf.str)
 			b1.skip(int(off))
 			val = b1.string()
 			if b1.err != nil {
@@ -262,7 +276,7 @@ func (r *Reader) Seek(off Offset) {
 		}
 		u := &d.unit[0]
 		r.unit = 0
-		r.b = makeBuf(r.d, "info", u.off, u.data, u.addrsize)
+		r.b = makeBuf(r.d, u, "info", u.off, u.data)
 		return
 	}
 
@@ -273,7 +287,7 @@ func (r *Reader) Seek(off Offset) {
 		u = &d.unit[i]
 		if u.off <= off && off < u.off+Offset(len(u.data)) {
 			r.unit = i
-			r.b = makeBuf(r.d, "info", off, u.data[off-u.off:], u.addrsize)
+			r.b = makeBuf(r.d, u, "info", off, u.data[off-u.off:])
 			return
 		}
 	}
@@ -285,7 +299,7 @@ func (r *Reader) maybeNextUnit() {
 	for len(r.b.data) == 0 && r.unit+1 < len(r.d.unit) {
 		r.unit++
 		u := &r.d.unit[r.unit]
-		r.b = makeBuf(r.d, "info", u.off, u.data, u.addrsize)
+		r.b = makeBuf(r.d, u, "info", u.off, u.data)
 	}
 }
 

src/pkg/debug/dwarf/type.go

@@ -435,7 +435,9 @@ func (d *Data) Type(off Offset) (Type, error) {
 					goto Error
 				}
 				if loc, ok := kid.Val(AttrDataMemberLoc).([]byte); ok {
-					b := makeBuf(d, "location", 0, loc, d.addrsize)
+					// TODO: Should have original compilation
+					// unit here, not unknownFormat.
+					b := makeBuf(d, unknownFormat{}, "location", 0, loc)
 					if b.uint8() != opPlusUconst {
 						err = DecodeError{"info", kid.Offset, "unexpected opcode"}
 						goto Error

src/pkg/debug/dwarf/unit.go

@@ -10,19 +10,44 @@ import "strconv"
 // Each unit has its own abbreviation table and address size.
 
 type unit struct {
-	base     Offset // byte offset of header within the aggregate info
-	off      Offset // byte offset of data within the aggregate info
-	data     []byte
-	atable   abbrevTable
-	addrsize int
+	base   Offset // byte offset of header within the aggregate info
+	off    Offset // byte offset of data within the aggregate info
+	data   []byte
+	atable abbrevTable
+	asize  int
+	vers   int
+	is64   bool // True for 64-bit DWARF format
+}
+
+// Implement the dataFormat interface.
+
+func (u *unit) version() int {
+	return u.vers
+}
+
+func (u *unit) dwarf64() (bool, bool) {
+	return u.is64, true
+}
+
+func (u *unit) addrsize() int {
+	return u.asize
 }
 
 func (d *Data) parseUnits() ([]unit, error) {
 	// Count units.
 	nunit := 0
-	b := makeBuf(d, "info", 0, d.info, 0)
+	b := makeBuf(d, unknownFormat{}, "info", 0, d.info)
 	for len(b.data) > 0 {
-		b.skip(int(b.uint32()))
+		len := b.uint32()
+		if len == 0xffffffff {
+			len64 := b.uint64()
+			if len64 != uint64(uint32(len64)) {
+				b.error("unit length overflow")
+				break
+			}
+			len = uint32(len64)
+		}
+		b.skip(int(len))
 		nunit++
 	}
 	if b.err != nil {
@@ -30,16 +55,22 @@ func (d *Data) parseUnits() ([]unit, error) {
 	}
 
 	// Again, this time writing them down.
-	b = makeBuf(d, "info", 0, d.info, 0)
+	b = makeBuf(d, unknownFormat{}, "info", 0, d.info)
 	units := make([]unit, nunit)
 	for i := range units {
 		u := &units[i]
 		u.base = b.off
 		n := b.uint32()
-		if vers := b.uint16(); vers != 2 {
+		if n == 0xffffffff {
+			u.is64 = true
+			n = uint32(b.uint64())
+		}
+		vers := b.uint16()
+		if vers != 2 && vers != 3 {
 			b.error("unsupported DWARF version " + strconv.Itoa(int(vers)))
 			break
 		}
+		u.vers = int(vers)
 		atable, err := d.parseAbbrev(b.uint32())
 		if err != nil {
 			if b.err == nil {
@@ -48,7 +79,7 @@ func (d *Data) parseUnits() ([]unit, error) {
 			break
 		}
 		u.atable = atable
-		u.addrsize = int(b.uint8())
+		u.asize = int(b.uint8())
 		u.off = b.off
 		u.data = b.bytes(int(n - (2 + 4 + 1)))
 	}