cmd/cgo: bug fixes

* disallow embedding of C type (Fixes issue 2552)
* detect 0-length array (Fixes issue 2806)
* use typedefs when possible, to avoid attribute((unavailable)) (Fixes issue 2888)
* print Go types constructed from C types using original C types (Fixes issue 2612)

This fix changes _cgo_export.h to repeat the preamble from import "C".
Otherwise the fix to issue 2612 is impossible, since it cannot refer to
types that have not been defined.  If people are using //export and
putting non-header information in the preamble, they will need to
refactor their code.

R=golang-dev, r, r
CC=golang-dev
https://golang.org/cl/5672080

Author: rsc

doc/go1.html

@@ -1853,7 +1853,24 @@ <h3 id="url">The url package</h3>
 The semantic changes make it difficult for the fix tool to update automatically.
 </p>
 
-<h2 id="go_command">The go command</h2>
+<h2 id="cmd_go">The go command</h2>
+
+<p>
+TODO: Write this.
+</p>
+
+<h2 id="cmd_cgo">The cgo command</h2>
+
+<p>
+In Go 1, the <a href="/cmd/cgo">cgo command</a>
+uses a different <code>_cgo_export.h</code>
+file, which is generated for packages containing <code>//export</code> lines.
+The <code>_cgo_export.h</code> file now begins with the C preamble comment,
+so that exported function definitions can use types defined there.
+This has the effect of compiling the preamble multiple times, so a
+package using <code>//export</code> must not put function definitions
+or variable initializations in the C preamble.
+</p
 
 <h2 id="releases">Packaged releases</h2>
 

doc/go1.tmpl

@@ -1725,7 +1725,24 @@ Code that uses the old fields will fail to compile and must be updated by hand.
 The semantic changes make it difficult for the fix tool to update automatically.
 </p>
 
-<h2 id="go_command">The go command</h2>
+<h2 id="cmd_go">The go command</h2>
+
+<p>
+TODO: Write this.
+</p>
+
+<h2 id="cmd_cgo">The cgo command</h2>
+
+<p>
+In Go 1, the <a href="/cmd/cgo">cgo command</a>
+uses a different <code>_cgo_export.h</code>
+file, which is generated for packages containing <code>//export</code> lines.
+The <code>_cgo_export.h</code> file now begins with the C preamble comment,
+so that exported function definitions can use types defined there.
+This has the effect of compiling the preamble multiple times, so a
+package using <code>//export</code> must not put function definitions
+or variable initializations in the C preamble.
+</p
 
 <h2 id="releases">Packaged releases</h2>
 

src/cmd/cgo/ast.go

@@ -147,6 +147,9 @@ func (f *File) saveRef(x interface{}, context string) {
 			if context == "as2" {
 				context = "expr"
 			}
+			if context == "embed-type" {
+				error_(sel.Pos(), "cannot embed C type")
+			}
 			goname := sel.Sel.Name
 			if goname == "errno" {
 				error_(sel.Pos(), "cannot refer to errno directly; see documentation")
@@ -232,7 +235,11 @@ func (f *File) walk(x interface{}, context string, visit func(*File, interface{}
 
 	// These are ordered and grouped to match ../../pkg/go/ast/ast.go
 	case *ast.Field:
-		f.walk(&n.Type, "type", visit)
+		if len(n.Names) == 0 && context == "field" {
+			f.walk(&n.Type, "embed-type", visit)
+		} else {
+			f.walk(&n.Type, "type", visit)
+		}
 	case *ast.FieldList:
 		for _, field := range n.List {
 			f.walk(field, context, visit)
@@ -289,9 +296,9 @@ func (f *File) walk(x interface{}, context string, visit func(*File, interface{}
 	case *ast.StructType:
 		f.walk(n.Fields, "field", visit)
 	case *ast.FuncType:
-		f.walk(n.Params, "field", visit)
+		f.walk(n.Params, "param", visit)
 		if n.Results != nil {
-			f.walk(n.Results, "field", visit)
+			f.walk(n.Results, "param", visit)
 		}
 	case *ast.InterfaceType:
 		f.walk(n.Methods, "field", visit)
@@ -379,7 +386,7 @@ func (f *File) walk(x interface{}, context string, visit func(*File, interface{}
 		f.walk(n.Specs, "spec", visit)
 	case *ast.FuncDecl:
 		if n.Recv != nil {
-			f.walk(n.Recv, "field", visit)
+			f.walk(n.Recv, "param", visit)
 		}
 		f.walk(n.Type, "type", visit)
 		if n.Body != nil {

src/cmd/cgo/doc.go

@@ -16,8 +16,8 @@ the pseudo-package "C" and then refers to types such as C.size_t,
 variables such as C.stdout, or functions such as C.putchar.
 
 If the import of "C" is immediately preceded by a comment, that
-comment is used as a header when compiling the C parts of
-the package.  For example:
+comment, called the preamble, is used as a header when compiling
+the C parts of the package.  For example:
 
 	// #include <stdio.h>
 	// #include <errno.h>
@@ -57,6 +57,8 @@ The C type void* is represented by Go's unsafe.Pointer.
 To access a struct, union, or enum type directly, prefix it with
 struct_, union_, or enum_, as in C.struct_stat.
 
+Go structs cannot embed fields with C types.
+
 Any C function that returns a value may be called in a multiple
 assignment context to retrieve both the return value and the
 C errno variable as an error.  For example:
@@ -100,7 +102,8 @@ They will be available in the C code as:
 	extern int64 MyFunction(int arg1, int arg2, GoString arg3);
 	extern struct MyFunction2_return MyFunction2(int arg1, int arg2, GoString arg3);
 
-found in _cgo_export.h generated header. Functions with multiple
+found in _cgo_export.h generated header, after any preambles
+copied from the cgo input files. Functions with multiple
 return values are mapped to functions returning a struct.
 Not all Go types can be mapped to C types in a useful way.
 

src/cmd/cgo/gcc.go

@@ -709,7 +709,7 @@ func (p *Package) rewriteRef(f *File) {
 			// Substitute definition for mangled type name.
 			if id, ok := expr.(*ast.Ident); ok {
 				if t := typedef[id.Name]; t != nil {
-					expr = t
+					expr = t.Go
 				}
 				if id.Name == r.Name.Mangle && r.Name.Const != "" {
 					expr = ast.NewIdent(r.Name.Const)
@@ -894,7 +894,7 @@ type typeConv struct {
 }
 
 var tagGen int
-var typedef = make(map[string]ast.Expr)
+var typedef = make(map[string]*Type)
 var goIdent = make(map[string]*ast.Ident)
 
 func (c *typeConv) Init(ptrSize int64) {
@@ -1164,17 +1164,22 @@ func (c *typeConv) Type(dtype dwarf.Type, pos token.Pos) *Type {
 		goIdent[name.Name] = name
 		switch dt.Kind {
 		case "union", "class":
-			typedef[name.Name] = c.Opaque(t.Size)
 			if t.C.Empty() {
 				t.C.Set("typeof(unsigned char[%d])", t.Size)
 			}
+			typedef[name.Name] = t
 		case "struct":
 			g, csyntax, align := c.Struct(dt, pos)
 			if t.C.Empty() {
 				t.C.Set(csyntax)
 			}
 			t.Align = align
-			typedef[name.Name] = g
+			tt := *t
+			if tag != "" {
+				tt.C = &TypeRepr{"struct %s", []interface{}{tag}}
+			}
+			tt.Go = g
+			typedef[name.Name] = &tt
 		}
 
 	case *dwarf.TypedefType:
@@ -1203,7 +1208,9 @@ func (c *typeConv) Type(dtype dwarf.Type, pos token.Pos) *Type {
 		t.Size = sub.Size
 		t.Align = sub.Align
 		if _, ok := typedef[name.Name]; !ok {
-			typedef[name.Name] = sub.Go
+			tt := *t
+			tt.Go = sub.Go
+			typedef[name.Name] = &tt
 		}
 		if *godefs || *cdefs {
 			t.Go = sub.Go
@@ -1250,7 +1257,8 @@ func (c *typeConv) Type(dtype dwarf.Type, pos token.Pos) *Type {
 			}
 			s = strings.Join(strings.Split(s, " "), "") // strip spaces
 			name := c.Ident("_Ctype_" + s)
-			typedef[name.Name] = t.Go
+			tt := *t
+			typedef[name.Name] = &tt
 			if !*godefs && !*cdefs {
 				t.Go = name
 			}
@@ -1288,9 +1296,18 @@ func (c *typeConv) FuncArg(dtype dwarf.Type, pos token.Pos) *Type {
 		if ptr, ok := base(dt.Type).(*dwarf.PtrType); ok {
 			// Unless the typedef happens to point to void* since
 			// Go has special rules around using unsafe.Pointer.
-			if _, void := base(ptr.Type).(*dwarf.VoidType); !void {
-				return c.Type(ptr, pos)
+			if _, void := base(ptr.Type).(*dwarf.VoidType); void {
+				break
 			}
+
+			t = c.Type(ptr, pos)
+			if t == nil {
+				return nil
+			}
+
+			// Remember the C spelling, in case the struct
+			// has __attribute__((unavailable)) on it.  See issue 2888.
+			t.Typedef = dt.Name
 		}
 	}
 	return t
@@ -1443,7 +1460,7 @@ func (c *typeConv) Struct(dt *dwarf.StructType, pos token.Pos) (expr *ast.Struct
 		off = dt.ByteSize
 	}
 	if off != dt.ByteSize {
-		fatalf("%s: struct size calculation error", lineno(pos))
+		fatalf("%s: struct size calculation error off=%d bytesize=%d", lineno(pos), off, dt.ByteSize)
 	}
 	buf.WriteString("}")
 	csyntax = buf.String()

src/cmd/cgo/godefs.go

@@ -80,7 +80,7 @@ func (p *Package) godefs(f *File, srcfile string) string {
 	// and xxx is a typedef for yyy, make C.yyy format as T.
 	for typ, def := range typedef {
 		if new := override[typ]; new != "" {
-			if id, ok := def.(*ast.Ident); ok {
+			if id, ok := def.Go.(*ast.Ident); ok {
 				override[id.Name] = new
 			}
 		}

src/cmd/cgo/main.go

@@ -39,6 +39,7 @@ type Package struct {
 	Decl        []ast.Decl
 	GoFiles     []string // list of Go files
 	GccFiles    []string // list of gcc output files
+	Preamble    string   // collected preamble for _cgo_export.h
 }
 
 // A File collects information about a single Go input file.
@@ -98,6 +99,7 @@ type Type struct {
 	C          *TypeRepr
 	Go         ast.Expr
 	EnumValues map[string]int64
+	Typedef    string
 }
 
 // A FuncType collects information about a function type in both the C and Go worlds.
@@ -312,6 +314,9 @@ func (p *Package) Record(f *File) {
 		}
 	}
 
-	p.ExpFunc = append(p.ExpFunc, f.ExpFunc...)
+	if f.ExpFunc != nil {
+		p.ExpFunc = append(p.ExpFunc, f.ExpFunc...)
+		p.Preamble += "\n" + f.Preamble
+	}
 	p.Decl = append(p.Decl, f.AST.Decls...)
 }

src/cmd/cgo/out.go

@@ -59,7 +59,7 @@ func (p *Package) writeDefs() {
 
 	for name, def := range typedef {
 		fmt.Fprintf(fgo2, "type %s ", name)
-		conf.Fprint(fgo2, fset, def)
+		conf.Fprint(fgo2, fset, def.Go)
 		fmt.Fprintf(fgo2, "\n\n")
 	}
 	fmt.Fprintf(fgo2, "type _Ctype_void [0]byte\n")
@@ -196,7 +196,11 @@ func (p *Package) structType(n *Name) (string, int64) {
 			fmt.Fprintf(&buf, "\t\tchar __pad%d[%d];\n", off, pad)
 			off += pad
 		}
-		fmt.Fprintf(&buf, "\t\t%s p%d;\n", t.C, i)
+		c := t.Typedef
+		if c == "" {
+			c = t.C.String()
+		}
+		fmt.Fprintf(&buf, "\t\t%s p%d;\n", c, i)
 		off += t.Size
 	}
 	if off%p.PtrSize != 0 {
@@ -428,6 +432,7 @@ func (p *Package) writeExports(fgo2, fc, fm *os.File) {
 	fgcch := creat(*objDir + "_cgo_export.h")
 
 	fmt.Fprintf(fgcch, "/* Created by cgo - DO NOT EDIT. */\n")
+	fmt.Fprintf(fgcch, "%s\n", p.Preamble)
 	fmt.Fprintf(fgcch, "%s\n", gccExportHeaderProlog)
 
 	fmt.Fprintf(fgcc, "/* Created by cgo - DO NOT EDIT. */\n")
@@ -693,10 +698,8 @@ func (p *Package) cgoType(e ast.Expr) *Type {
 				}
 			}
 		}
-		for name, def := range typedef {
-			if name == t.Name {
-				return p.cgoType(def)
-			}
+		if def := typedef[t.Name]; def != nil {
+			return def
 		}
 		if t.Name == "uintptr" {
 			return &Type{Size: p.PtrSize, Align: p.PtrSize, C: c("uintptr")}
@@ -721,7 +724,7 @@ func (p *Package) cgoType(e ast.Expr) *Type {
 			return &Type{Size: p.PtrSize, Align: p.PtrSize, C: c("void*")}
 		}
 	}
-	error_(e.Pos(), "unrecognized Go type %T", e)
+	error_(e.Pos(), "Go type not supported in export: %s", gofmt(e))
 	return &Type{Size: 4, Align: 4, C: c("int")}
 }
 

src/cmd/cgo/util.go

@@ -70,7 +70,11 @@ func lineno(pos token.Pos) string {
 
 // Die with an error message.
 func fatalf(msg string, args ...interface{}) {
-	fmt.Fprintf(os.Stderr, msg+"\n", args...)
+	// If we've already printed other errors, they might have
+	// caused the fatal condition.  Assume they're enough.
+	if nerrors == 0 {
+		fmt.Fprintf(os.Stderr, msg+"\n", args...)
+	}
 	os.Exit(2)
 }
 

src/pkg/debug/dwarf/testdata/typedef.c

@@ -28,8 +28,13 @@ typedef struct my_struct {
 	volatile int vi;
 	char x : 1;
 	int y : 4;
+	int z[0];
 	long long array[40];
+	int zz[0];
 } t_my_struct;
+typedef struct my_struct1 {
+	int zz [1];
+} t_my_struct1;
 typedef union my_union {
 	volatile int vi;
 	char x : 1;
@@ -65,7 +70,8 @@ t_func_void_of_char *a9;
 t_func_void_of_void *a10;
 t_func_void_of_ptr_char_dots *a11;
 t_my_struct *a12;
-t_my_union *a12a;
+t_my_struct1 *a12a;
+t_my_union *a12b;
 t_my_enum *a13;
 t_my_list *a14;
 t_my_tree *a15;