llgo/internal/abi/type.go

/*
 * Copyright (c) 2024 The GoPlus Authors (goplus.org). All rights reserved.
 *
 * Licensed under the Apache License, Version 2.0 (the "License");
 * you may not use this file except in compliance with the License.
 * You may obtain a copy of the License at
 *
 *     http://www.apache.org/licenses/LICENSE-2.0
 *
 * Unless required by applicable law or agreed to in writing, software
 * distributed under the License is distributed on an "AS IS" BASIS,
 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 * See the License for the specific language governing permissions and
 * limitations under the License.
 */

package abi

import (
	"unsafe"
)

// -----------------------------------------------------------------------------

// Type is the runtime representation of a Go type.
//
// Type is also referenced implicitly
// (in the form of expressions involving constants and arch.PtrSize)
// in cmd/compile/internal/reflectdata/reflect.go
// and cmd/link/internal/ld/decodesym.go
// (e.g. data[2*arch.PtrSize+4] references the TFlag field)
// unsafe.OffsetOf(Type{}.TFlag) cannot be used directly in those
// places because it varies with cross compilation and experiments.
type Type struct {
	Size_       uintptr
	PtrBytes    uintptr // number of (prefix) bytes in the type that can contain pointers
	Hash        uint32  // hash of type; avoids computation in hash tables
	TFlag       TFlag   // extra type information flags
	Align_      uint8   // alignment of variable with this type
	FieldAlign_ uint8   // alignment of struct field with this type
	Kind_       uint8   // enumeration for C
	// function for comparing objects of this type
	// (ptr to object A, ptr to object B) -> ==?
	Equal func(unsafe.Pointer, unsafe.Pointer) bool
	// GCData stores the GC type data for the garbage collector.
	// If the KindGCProg bit is set in kind, GCData is a GC program.
	// Otherwise it is a ptrmask bitmap. See mbitmap.go for details.
	GCData    *byte
	Str       NameOff // string form
	PtrToThis TypeOff // type for pointer to this type, may be zero
}

// A Kind represents the specific kind of type that a Type represents.
// The zero Kind is not a valid kind.
type Kind uint

const (
	Invalid Kind = iota
	Bool
	Int
	Int8
	Int16
	Int32
	Int64
	Uint
	Uint8
	Uint16
	Uint32
	Uint64
	Uintptr
	Float32
	Float64
	Complex64
	Complex128
	Array
	Chan
	Func
	Interface
	Map
	Pointer
	Slice
	String
	Struct
	UnsafePointer
)

/*
const (
	// TODO (khr, drchase) why aren't these in TFlag?  Investigate, fix if possible.
	KindDirectIface = 1 << 5
	KindGCProg      = 1 << 6 // Type.gc points to GC program
	KindMask        = (1 << 5) - 1
)
*/

// TFlag is used by a Type to signal what extra type information is
// available in the memory directly following the Type value.
type TFlag uint8

const (
	// TFlagUncommon means that there is a data with a type, UncommonType,
	// just beyond the shared-per-type common data.  That is, the data
	// for struct types will store their UncommonType at one offset, the
	// data for interface types will store their UncommonType at a different
	// offset.  UncommonType is always accessed via a pointer that is computed
	// using trust-us-we-are-the-implementors pointer arithmetic.
	//
	// For example, if t.Kind() == Struct and t.tflag&TFlagUncommon != 0,
	// then t has UncommonType data and it can be accessed as:
	//
	//	type structTypeUncommon struct {
	//		structType
	//		u UncommonType
	//	}
	//	u := &(*structTypeUncommon)(unsafe.Pointer(t)).u
	TFlagUncommon TFlag = 1 << 0

	// TFlagExtraStar means the name in the str field has an
	// extraneous '*' prefix. This is because for most types T in
	// a program, the type *T also exists and reusing the str data
	// saves binary size.
	TFlagExtraStar TFlag = 1 << 1

	// TFlagNamed means the type has a name.
	TFlagNamed TFlag = 1 << 2

	// TFlagRegularMemory means that equal and hash functions can treat
	// this type as a single region of t.size bytes.
	TFlagRegularMemory TFlag = 1 << 3
)

// NameOff is the offset to a name from moduledata.types.  See resolveNameOff in runtime.
type NameOff int32

// TypeOff is the offset to a type from moduledata.types.  See resolveTypeOff in runtime.
type TypeOff int32

// -----------------------------------------------------------------------------

// ArrayType represents a fixed array type.
type ArrayType struct {
	Type
	Elem  *Type // array element type
	Slice *Type // slice type
	Len   uintptr
}

type SliceType struct {
	Type
	Elem *Type // slice element type
}

type MapType struct {
	Type
	Key    *Type
	Elem   *Type
	Bucket *Type // internal type representing a hash bucket
	// function for hashing keys (ptr to key, seed) -> hash
	Hasher     func(unsafe.Pointer, uintptr) uintptr
	KeySize    uint8  // size of key slot
	ValueSize  uint8  // size of elem slot
	BucketSize uint16 // size of bucket
	Flags      uint32
}

type PtrType struct {
	Type
	Elem *Type // pointer element (pointed at) type
}

type ChanDir int

const (
	RecvDir    ChanDir = 1 << iota         // <-chan
	SendDir                                // chan<-
	BothDir            = RecvDir | SendDir // chan
	InvalidDir ChanDir = 0
)

// ChanType represents a channel type
type ChanType struct {
	Type
	Elem *Type
	Dir  ChanDir
}

// funcType represents a function type.
//
// A *Type for each in and out parameter is stored in an array that
// directly follows the funcType (and possibly its uncommonType). So
// a function type with one method, one input, and one output is:
//
//	struct {
//		funcType
//		uncommonType
//		[2]*rtype    // [0] is in, [1] is out
//	}
type FuncType struct {
	Type
	InCount  uint16
	OutCount uint16 // top bit is set if last input parameter is ...
}

type StructField struct {
	Name   Name    // name is always non-empty
	Typ    *Type   // type of field
	Offset uintptr // byte offset of field
}

func (f *StructField) Embedded() bool {
	return f.Name.IsEmbedded()
}

type StructType struct {
	Type
	PkgPath Name
	Fields  []StructField
}

type InterfaceType struct {
	Type
	PkgPath Name      // import path
	Methods []Imethod // sorted by hash
}

// Imethod represents a method on an interface type
type Imethod struct {
	Name NameOff // name of method
	Typ  TypeOff // .(*FuncType) underneath
}

func (t *Type) Kind() Kind { return Kind(t.Kind_) }

// Size returns the size of data with type t.
func (t *Type) Size() uintptr { return t.Size_ }

// Align returns the alignment of data with type t.
func (t *Type) Align() int { return int(t.Align_) }

func (t *Type) FieldAlign() int { return int(t.FieldAlign_) }

func (t *Type) Common() *Type {
	return t
}

// Len returns the length of t if t is an array type, otherwise 0
func (t *Type) Len() int {
	if t.Kind() == Array {
		return int((*ArrayType)(unsafe.Pointer(t)).Len)
	}
	return 0
}

// Elem returns the element type for t if t is an array, channel, map, pointer, or slice, otherwise nil.
func (t *Type) Elem() *Type {
	switch t.Kind() {
	case Array:
		tt := (*ArrayType)(unsafe.Pointer(t))
		return tt.Elem
	case Chan:
		tt := (*ChanType)(unsafe.Pointer(t))
		return tt.Elem
	case Map:
		tt := (*MapType)(unsafe.Pointer(t))
		return tt.Elem
	case Pointer:
		tt := (*PtrType)(unsafe.Pointer(t))
		return tt.Elem
	case Slice:
		tt := (*SliceType)(unsafe.Pointer(t))
		return tt.Elem
	}
	return nil
}

// StructType returns t cast to a *StructType, or nil if its tag does not match.
func (t *Type) StructType() *StructType {
	if t.Kind() != Struct {
		return nil
	}
	return (*StructType)(unsafe.Pointer(t))
}

// MapType returns t cast to a *MapType, or nil if its tag does not match.
func (t *Type) MapType() *MapType {
	if t.Kind() != Map {
		return nil
	}
	return (*MapType)(unsafe.Pointer(t))
}

// ArrayType returns t cast to a *ArrayType, or nil if its tag does not match.
func (t *Type) ArrayType() *ArrayType {
	if t.Kind() != Array {
		return nil
	}
	return (*ArrayType)(unsafe.Pointer(t))
}

// FuncType returns t cast to a *FuncType, or nil if its tag does not match.
func (t *Type) FuncType() *FuncType {
	if t.Kind() != Func {
		return nil
	}
	return (*FuncType)(unsafe.Pointer(t))
}

// InterfaceType returns t cast to a *InterfaceType, or nil if its tag does not match.
func (t *Type) InterfaceType() *InterfaceType {
	if t.Kind() != Interface {
		return nil
	}
	return (*InterfaceType)(unsafe.Pointer(t))
}

// -----------------------------------------------------------------------------

// addChecked returns p+x.
//
// The whySafe string is ignored, so that the function still inlines
// as efficiently as p+x, but all call sites should use the string to
// record why the addition is safe, which is to say why the addition
// does not cause x to advance to the very end of p's allocation
// and therefore point incorrectly at the next block in memory.
func addChecked(p unsafe.Pointer, x uintptr, whySafe string) unsafe.Pointer {
	_ = whySafe
	return unsafe.Pointer(uintptr(p) + x)
}

// Name is an encoded type Name with optional extra data.
//
// The first byte is a bit field containing:
//
//	1<<0 the name is exported
//	1<<1 tag data follows the name
//	1<<2 pkgPath nameOff follows the name and tag
//	1<<3 the name is of an embedded (a.k.a. anonymous) field
//
// Following that, there is a varint-encoded length of the name,
// followed by the name itself.
//
// If tag data is present, it also has a varint-encoded length
// followed by the tag itself.
//
// If the import path follows, then 4 bytes at the end of
// the data form a nameOff. The import path is only set for concrete
// methods that are defined in a different package than their type.
//
// If a name starts with "*", then the exported bit represents
// whether the pointed to type is exported.
//
// Note: this encoding must match here and in:
//   cmd/compile/internal/reflectdata/reflect.go
//   cmd/link/internal/ld/decodesym.go

type Name struct {
	Bytes *byte
}

// DataChecked does pointer arithmetic on n's Bytes, and that arithmetic is asserted to
// be safe for the reason in whySafe (which can appear in a backtrace, etc.)
func (n Name) DataChecked(off int, whySafe string) *byte {
	return (*byte)(addChecked(unsafe.Pointer(n.Bytes), uintptr(off), whySafe))
}

// Data does pointer arithmetic on n's Bytes, and that arithmetic is asserted to
// be safe because the runtime made the call (other packages use DataChecked)
func (n Name) Data(off int) *byte {
	return (*byte)(addChecked(unsafe.Pointer(n.Bytes), uintptr(off), "the runtime doesn't need to give you a reason"))
}

// IsExported returns "is n exported?"
func (n Name) IsExported() bool {
	return (*n.Bytes)&(1<<0) != 0
}

// HasTag returns true iff there is tag data following this name
func (n Name) HasTag() bool {
	return (*n.Bytes)&(1<<1) != 0
}

// IsEmbedded returns true iff n is embedded (an anonymous field).
func (n Name) IsEmbedded() bool {
	return (*n.Bytes)&(1<<3) != 0
}

// ReadVarint parses a varint as encoded by encoding/binary.
// It returns the number of encoded bytes and the encoded value.
func (n Name) ReadVarint(off int) (int, int) {
	v := 0
	for i := 0; ; i++ {
		x := *n.DataChecked(off+i, "read varint")
		v += int(x&0x7f) << (7 * i)
		if x&0x80 == 0 {
			return i + 1, v
		}
	}
}

// IsBlank indicates whether n is "_".
func (n Name) IsBlank() bool {
	if n.Bytes == nil {
		return false
	}
	_, l := n.ReadVarint(1)
	return l == 1 && *n.Data(2) == '_'
}

// writeVarint writes n to buf in varint form. Returns the
// number of bytes written. n must be nonnegative.
// Writes at most 10 bytes.
func writeVarint(buf []byte, n int) int {
	for i := 0; ; i++ {
		b := byte(n & 0x7f)
		n >>= 7
		if n == 0 {
			buf[i] = b
			return i + 1
		}
		buf[i] = b | 0x80
	}
}

// Name returns the tag string for n, or empty if there is none.
func (n Name) Name() string {
	if n.Bytes == nil {
		return ""
	}
	i, l := n.ReadVarint(1)
	return unsafe.String(n.DataChecked(1+i, "non-empty string"), l)
}

// Tag returns the tag string for n, or empty if there is none.
func (n Name) Tag() string {
	if !n.HasTag() {
		return ""
	}
	i, l := n.ReadVarint(1)
	i2, l2 := n.ReadVarint(1 + i + l)
	return unsafe.String(n.DataChecked(1+i+l+i2, "non-empty string"), l2)
}

func NewName(n, tag string, exported, embedded bool) Name {
	if len(n) >= 1<<29 {
		panic("abi.NewName: name too long: " + n[:1024] + "...")
	}
	if len(tag) >= 1<<29 {
		panic("abi.NewName: tag too long: " + tag[:1024] + "...")
	}
	var nameLen [10]byte
	var tagLen [10]byte
	nameLenLen := writeVarint(nameLen[:], len(n))
	tagLenLen := writeVarint(tagLen[:], len(tag))

	var bits byte
	l := 1 + nameLenLen + len(n)
	if exported {
		bits |= 1 << 0
	}
	if len(tag) > 0 {
		l += tagLenLen + len(tag)
		bits |= 1 << 1
	}
	if embedded {
		bits |= 1 << 3
	}

	b := make([]byte, l)
	b[0] = bits
	copy(b[1:], nameLen[:nameLenLen])
	copy(b[1+nameLenLen:], n)
	if len(tag) > 0 {
		tb := b[1+nameLenLen+len(n):]
		copy(tb, tagLen[:tagLenLen])
		copy(tb[tagLenLen:], tag)
	}

	return Name{Bytes: &b[0]}
}

// -----------------------------------------------------------------------------