mirror of
				https://github.com/ethereum/solidity
				synced 2023-10-03 13:03:40 +00:00 
			
		
		
		
	Also make sure that `i32.drop/i64.drop` won't silently drop its arguments, as they may have side-effects.
		
			
				
	
	
		
			724 lines
		
	
	
		
			19 KiB
		
	
	
	
		
			C++
		
	
	
	
	
	
			
		
		
	
	
			724 lines
		
	
	
		
			19 KiB
		
	
	
	
		
			C++
		
	
	
	
	
	
| /*
 | |
| 	This file is part of solidity.
 | |
| 
 | |
| 	solidity is free software: you can redistribute it and/or modify
 | |
| 	it under the terms of the GNU General Public License as published by
 | |
| 	the Free Software Foundation, either version 3 of the License, or
 | |
| 	(at your option) any later version.
 | |
| 
 | |
| 	solidity is distributed in the hope that it will be useful,
 | |
| 	but WITHOUT ANY WARRANTY; without even the implied warranty of
 | |
| 	MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
 | |
| 	GNU General Public License for more details.
 | |
| 
 | |
| 	You should have received a copy of the GNU General Public License
 | |
| 	along with solidity.  If not, see <http://www.gnu.org/licenses/>.
 | |
| */
 | |
| // SPDX-License-Identifier: GPL-3.0
 | |
| /**
 | |
|  * Component that transforms internal Wasm representation to binary.
 | |
|  */
 | |
| 
 | |
| #include <libyul/backends/wasm/BinaryTransform.h>
 | |
| 
 | |
| #include <libyul/Exceptions.h>
 | |
| #include <libsolutil/CommonData.h>
 | |
| #include <libsolutil/Visitor.h>
 | |
| #include <libsolutil/LEB128.h>
 | |
| 
 | |
| #include <boost/range/adaptor/reversed.hpp>
 | |
| #include <boost/range/adaptor/map.hpp>
 | |
| #include <boost/range/adaptor/transformed.hpp>
 | |
| 
 | |
| using namespace std;
 | |
| using namespace solidity;
 | |
| using namespace solidity::yul;
 | |
| using namespace solidity::yul::wasm;
 | |
| using namespace solidity::util;
 | |
| 
 | |
| namespace
 | |
| {
 | |
| 
 | |
| bytes toBytes(uint8_t _b)
 | |
| {
 | |
| 	return bytes(1, _b);
 | |
| }
 | |
| 
 | |
| enum class LimitsKind: uint8_t
 | |
| {
 | |
| 	Min = 0x00,
 | |
| 	MinMax = 0x01,
 | |
| };
 | |
| 
 | |
| enum class Mutability: uint8_t
 | |
| {
 | |
| 	Const = 0x00,
 | |
| 	Var = 0x01,
 | |
| };
 | |
| 
 | |
| enum class Section: uint8_t
 | |
| {
 | |
| 	CUSTOM = 0x00,
 | |
| 	TYPE = 0x01,
 | |
| 	IMPORT = 0x02,
 | |
| 	FUNCTION = 0x03,
 | |
| 	MEMORY = 0x05,
 | |
| 	GLOBAL = 0x06,
 | |
| 	EXPORT = 0x07,
 | |
| 	CODE = 0x0a
 | |
| };
 | |
| 
 | |
| bytes toBytes(Section _s)
 | |
| {
 | |
| 	return toBytes(uint8_t(_s));
 | |
| }
 | |
| 
 | |
| enum class ValueType: uint8_t
 | |
| {
 | |
| 	Void = 0x40,
 | |
| 	Function = 0x60,
 | |
| 	I64 = 0x7e,
 | |
| 	I32 = 0x7f
 | |
| };
 | |
| 
 | |
| bytes toBytes(ValueType _vt)
 | |
| {
 | |
| 	return toBytes(uint8_t(_vt));
 | |
| }
 | |
| 
 | |
| ValueType toValueType(wasm::Type _type)
 | |
| {
 | |
| 	if (_type == wasm::Type::i32)
 | |
| 		return ValueType::I32;
 | |
| 	else if (_type == wasm::Type::i64)
 | |
| 		return ValueType::I64;
 | |
| 	else
 | |
| 		yulAssert(false, "Invalid wasm variable type");
 | |
| }
 | |
| 
 | |
| enum class Export: uint8_t
 | |
| {
 | |
| 	Function = 0x0,
 | |
| 	Memory = 0x2
 | |
| };
 | |
| 
 | |
| bytes toBytes(Export _export)
 | |
| {
 | |
| 	return toBytes(uint8_t(_export));
 | |
| }
 | |
| 
 | |
| // NOTE: This is a subset of WebAssembly opcodes.
 | |
| //       Those available as a builtin are listed further down.
 | |
| enum class Opcode: uint8_t
 | |
| {
 | |
| 	Block = 0x02,
 | |
| 	Loop = 0x03,
 | |
| 	If = 0x04,
 | |
| 	Else = 0x05,
 | |
| 	End = 0x0b,
 | |
| 	Br = 0x0c,
 | |
| 	BrIf = 0x0d,
 | |
| 	BrTable = 0x0e, // Not used yet.
 | |
| 	Return = 0x0f,
 | |
| 	Call = 0x10,
 | |
| 	CallIndirect = 0x11, // Not used yet.
 | |
| 	LocalGet = 0x20,
 | |
| 	LocalSet = 0x21,
 | |
| 	LocalTee = 0x22, // Not used yet.
 | |
| 	GlobalGet = 0x23,
 | |
| 	GlobalSet = 0x24,
 | |
| 	I32Const = 0x41,
 | |
| 	I64Const = 0x42,
 | |
| };
 | |
| 
 | |
| bytes toBytes(Opcode _o)
 | |
| {
 | |
| 	return toBytes(uint8_t(_o));
 | |
| }
 | |
| 
 | |
| Opcode constOpcodeFor(ValueType _type)
 | |
| {
 | |
| 	if (_type == ValueType::I32)
 | |
| 		return Opcode::I32Const;
 | |
| 	else if (_type == ValueType::I64)
 | |
| 		return Opcode::I64Const;
 | |
| 	else
 | |
| 		yulAssert(false, "Values of this type cannot be used with const opcode");
 | |
| }
 | |
| 
 | |
| static map<string, uint8_t> const builtins = {
 | |
| 	{"unreachable", 0x00},
 | |
| 	{"nop", 0x01},
 | |
| 	{"i32.drop", 0x1a},
 | |
| 	{"i64.drop", 0x1a},
 | |
| 	{"i32.select", 0x1b},
 | |
| 	{"i64.select", 0x1b},
 | |
| 	{"i32.load", 0x28},
 | |
| 	{"i64.load", 0x29},
 | |
| 	{"i32.load8_s", 0x2c},
 | |
| 	{"i32.load8_u", 0x2d},
 | |
| 	{"i32.load16_s", 0x2e},
 | |
| 	{"i32.load16_u", 0x2f},
 | |
| 	{"i64.load8_s", 0x30},
 | |
| 	{"i64.load8_u", 0x31},
 | |
| 	{"i64.load16_s", 0x32},
 | |
| 	{"i64.load16_u", 0x33},
 | |
| 	{"i64.load32_s", 0x34},
 | |
| 	{"i64.load32_u", 0x35},
 | |
| 	{"i32.store", 0x36},
 | |
| 	{"i64.store", 0x37},
 | |
| 	{"i32.store8", 0x3a},
 | |
| 	{"i32.store16", 0x3b},
 | |
| 	{"i64.store8", 0x3c},
 | |
| 	{"i64.store16", 0x3d},
 | |
| 	{"i64.store32", 0x3e},
 | |
| 	{"memory.size", 0x3f},
 | |
| 	{"memory.grow", 0x40},
 | |
| 	{"i32.eqz", 0x45},
 | |
| 	{"i32.eq", 0x46},
 | |
| 	{"i32.ne", 0x47},
 | |
| 	{"i32.lt_s", 0x48},
 | |
| 	{"i32.lt_u", 0x49},
 | |
| 	{"i32.gt_s", 0x4a},
 | |
| 	{"i32.gt_u", 0x4b},
 | |
| 	{"i32.le_s", 0x4c},
 | |
| 	{"i32.le_u", 0x4d},
 | |
| 	{"i32.ge_s", 0x4e},
 | |
| 	{"i32.ge_u", 0x4f},
 | |
| 	{"i64.eqz", 0x50},
 | |
| 	{"i64.eq", 0x51},
 | |
| 	{"i64.ne", 0x52},
 | |
| 	{"i64.lt_s", 0x53},
 | |
| 	{"i64.lt_u", 0x54},
 | |
| 	{"i64.gt_s", 0x55},
 | |
| 	{"i64.gt_u", 0x56},
 | |
| 	{"i64.le_s", 0x57},
 | |
| 	{"i64.le_u", 0x58},
 | |
| 	{"i64.ge_s", 0x59},
 | |
| 	{"i64.ge_u", 0x5a},
 | |
| 	{"i32.clz", 0x67},
 | |
| 	{"i32.ctz", 0x68},
 | |
| 	{"i32.popcnt", 0x69},
 | |
| 	{"i32.add", 0x6a},
 | |
| 	{"i32.sub", 0x6b},
 | |
| 	{"i32.mul", 0x6c},
 | |
| 	{"i32.div_s", 0x6d},
 | |
| 	{"i32.div_u", 0x6e},
 | |
| 	{"i32.rem_s", 0x6f},
 | |
| 	{"i32.rem_u", 0x70},
 | |
| 	{"i32.and", 0x71},
 | |
| 	{"i32.or", 0x72},
 | |
| 	{"i32.xor", 0x73},
 | |
| 	{"i32.shl", 0x74},
 | |
| 	{"i32.shr_s", 0x75},
 | |
| 	{"i32.shr_u", 0x76},
 | |
| 	{"i32.rotl", 0x77},
 | |
| 	{"i32.rotr", 0x78},
 | |
| 	{"i64.clz", 0x79},
 | |
| 	{"i64.ctz", 0x7a},
 | |
| 	{"i64.popcnt", 0x7b},
 | |
| 	{"i64.add", 0x7c},
 | |
| 	{"i64.sub", 0x7d},
 | |
| 	{"i64.mul", 0x7e},
 | |
| 	{"i64.div_s", 0x7f},
 | |
| 	{"i64.div_u", 0x80},
 | |
| 	{"i64.rem_s", 0x81},
 | |
| 	{"i64.rem_u", 0x82},
 | |
| 	{"i64.and", 0x83},
 | |
| 	{"i64.or", 0x84},
 | |
| 	{"i64.xor", 0x85},
 | |
| 	{"i64.shl", 0x86},
 | |
| 	{"i64.shr_s", 0x87},
 | |
| 	{"i64.shr_u", 0x88},
 | |
| 	{"i64.rotl", 0x89},
 | |
| 	{"i64.rotr", 0x8a},
 | |
| 	{"i32.wrap_i64", 0xa7},
 | |
| 	{"i64.extend_i32_s", 0xac},
 | |
| 	{"i64.extend_i32_u", 0xad},
 | |
| };
 | |
| 
 | |
| bytes prefixSize(bytes _data)
 | |
| {
 | |
| 	size_t size = _data.size();
 | |
| 	return lebEncode(size) + move(_data);
 | |
| }
 | |
| 
 | |
| bytes makeSection(Section _section, bytes _data)
 | |
| {
 | |
| 	return toBytes(_section) + prefixSize(move(_data));
 | |
| }
 | |
| 
 | |
| /// This is a kind of run-length-encoding of local types.
 | |
| vector<pair<size_t, ValueType>> groupLocalVariables(vector<VariableDeclaration> _localVariables)
 | |
| {
 | |
| 	vector<pair<size_t, ValueType>> localEntries;
 | |
| 
 | |
| 	size_t entrySize = 0;
 | |
| 	ValueType entryType = ValueType::I32; // Any type would work here
 | |
| 	for (VariableDeclaration const& localVariable: _localVariables)
 | |
| 	{
 | |
| 		ValueType variableType = toValueType(localVariable.type);
 | |
| 
 | |
| 		if (variableType != entryType)
 | |
| 		{
 | |
| 			if (entrySize > 0)
 | |
| 				localEntries.emplace_back(entrySize, entryType);
 | |
| 
 | |
| 			entryType = variableType;
 | |
| 			entrySize = 0;
 | |
| 		}
 | |
| 
 | |
| 		++entrySize;
 | |
| 	}
 | |
| 	if (entrySize > 0)
 | |
| 		localEntries.emplace_back(entrySize, entryType);
 | |
| 
 | |
| 	return localEntries;
 | |
| }
 | |
| 
 | |
| }
 | |
| 
 | |
| bytes BinaryTransform::run(Module const& _module)
 | |
| {
 | |
| 	map<Type, vector<string>> const types = typeToFunctionMap(_module.imports, _module.functions);
 | |
| 
 | |
| 	map<string, size_t> const globalIDs = enumerateGlobals(_module);
 | |
| 	map<string, size_t> const functionIDs = enumerateFunctions(_module);
 | |
| 	map<string, size_t> const functionTypes = enumerateFunctionTypes(types);
 | |
| 
 | |
| 	yulAssert(globalIDs.size() == _module.globals.size(), "");
 | |
| 	yulAssert(functionIDs.size() == _module.imports.size() + _module.functions.size(), "");
 | |
| 	yulAssert(functionTypes.size() == functionIDs.size(), "");
 | |
| 	yulAssert(functionTypes.size() >= types.size(), "");
 | |
| 
 | |
| 	bytes ret{0, 'a', 's', 'm'};
 | |
| 	// version
 | |
| 	ret += bytes{1, 0, 0, 0};
 | |
| 	ret += typeSection(types);
 | |
| 	ret += importSection(_module.imports, functionTypes);
 | |
| 	ret += functionSection(_module.functions, functionTypes);
 | |
| 	ret += memorySection();
 | |
| 	ret += globalSection(_module.globals);
 | |
| 	ret += exportSection(functionIDs);
 | |
| 
 | |
| 	map<string, pair<size_t, size_t>> subModulePosAndSize;
 | |
| 	for (auto const& [name, module]: _module.subModules)
 | |
| 	{
 | |
| 		// TODO should we prefix and / or shorten the name?
 | |
| 		bytes data = BinaryTransform::run(module);
 | |
| 		size_t const length = data.size();
 | |
| 		ret += customSection(name, move(data));
 | |
| 		// Skip all the previous sections and the size field of this current custom section.
 | |
| 		size_t const offset = ret.size() - length;
 | |
| 		subModulePosAndSize[name] = {offset, length};
 | |
| 	}
 | |
| 	for (auto const& [name, data]: _module.customSections)
 | |
| 	{
 | |
| 		size_t const length = data.size();
 | |
| 		ret += customSection(name, data);
 | |
| 		// Skip all the previous sections and the size field of this current custom section.
 | |
| 		size_t const offset = ret.size() - length;
 | |
| 		subModulePosAndSize[name] = {offset, length};
 | |
| 	}
 | |
| 
 | |
| 	BinaryTransform bt(
 | |
| 		move(globalIDs),
 | |
| 		move(functionIDs),
 | |
| 		move(functionTypes),
 | |
| 		move(subModulePosAndSize)
 | |
| 	);
 | |
| 
 | |
| 	ret += bt.codeSection(_module.functions);
 | |
| 	return ret;
 | |
| }
 | |
| 
 | |
| bytes BinaryTransform::operator()(Literal const& _literal)
 | |
| {
 | |
| 	return std::visit(GenericVisitor{
 | |
| 		[&](uint32_t _value) -> bytes { return toBytes(Opcode::I32Const) + lebEncodeSigned(static_cast<int32_t>(_value)); },
 | |
| 		[&](uint64_t _value) -> bytes { return toBytes(Opcode::I64Const) + lebEncodeSigned(static_cast<int64_t>(_value)); },
 | |
| 	}, _literal.value);
 | |
| }
 | |
| 
 | |
| bytes BinaryTransform::operator()(StringLiteral const&)
 | |
| {
 | |
| 	// StringLiteral is a special AST element used for certain builtins.
 | |
| 	// It is not mapped to actual WebAssembly, and should be processed in visit(BuiltinCall).
 | |
| 	yulAssert(false, "");
 | |
| }
 | |
| 
 | |
| bytes BinaryTransform::operator()(LocalVariable const& _variable)
 | |
| {
 | |
| 	return toBytes(Opcode::LocalGet) + lebEncode(m_locals.at(_variable.name));
 | |
| }
 | |
| 
 | |
| bytes BinaryTransform::operator()(GlobalVariable const& _variable)
 | |
| {
 | |
| 	return toBytes(Opcode::GlobalGet) + lebEncode(m_globalIDs.at(_variable.name));
 | |
| }
 | |
| 
 | |
| bytes BinaryTransform::operator()(BuiltinCall const& _call)
 | |
| {
 | |
| 	// We need to avoid visiting the arguments of `dataoffset` and `datasize` because
 | |
| 	// they are references to object names that should not end up in the code.
 | |
| 	if (_call.functionName == "dataoffset")
 | |
| 	{
 | |
| 		string name = get<StringLiteral>(_call.arguments.at(0)).value;
 | |
| 		// TODO: support the case where name refers to the current object
 | |
| 		yulAssert(m_subModulePosAndSize.count(name), "");
 | |
| 		return toBytes(Opcode::I64Const) + lebEncodeSigned(static_cast<int64_t>(m_subModulePosAndSize.at(name).first));
 | |
| 	}
 | |
| 	else if (_call.functionName == "datasize")
 | |
| 	{
 | |
| 		string name = get<StringLiteral>(_call.arguments.at(0)).value;
 | |
| 		// TODO: support the case where name refers to the current object
 | |
| 		yulAssert(m_subModulePosAndSize.count(name), "");
 | |
| 		return toBytes(Opcode::I64Const) + lebEncodeSigned(static_cast<int64_t>(m_subModulePosAndSize.at(name).second));
 | |
| 	}
 | |
| 
 | |
| 	yulAssert(builtins.count(_call.functionName), "Builtin " + _call.functionName + " not found");
 | |
| 	// NOTE: the dialect ensures we have the right amount of arguments
 | |
| 	bytes args = visit(_call.arguments);
 | |
| 	bytes ret = move(args) + toBytes(builtins.at(_call.functionName));
 | |
| 	if (
 | |
| 		_call.functionName.find(".load") != string::npos ||
 | |
| 		_call.functionName.find(".store") != string::npos
 | |
| 	)
 | |
| 		// Alignment hint and offset. Interpreters ignore the alignment. JITs/AOTs can take it
 | |
| 		// into account to generate more efficient code but if the hint is invalid it could
 | |
| 		// actually be more expensive. It's best to hint at 1-byte alignment if we don't plan
 | |
| 		// to control the memory layout accordingly.
 | |
| 		ret += bytes{{0, 0}}; // 2^0 == 1-byte alignment
 | |
| 
 | |
| 	return ret;
 | |
| }
 | |
| 
 | |
| bytes BinaryTransform::operator()(FunctionCall const& _call)
 | |
| {
 | |
| 	return visit(_call.arguments) + toBytes(Opcode::Call) + lebEncode(m_functionIDs.at(_call.functionName));
 | |
| }
 | |
| 
 | |
| bytes BinaryTransform::operator()(LocalAssignment const& _assignment)
 | |
| {
 | |
| 	return
 | |
| 		std::visit(*this, *_assignment.value) +
 | |
| 		toBytes(Opcode::LocalSet) +
 | |
| 		lebEncode(m_locals.at(_assignment.variableName));
 | |
| }
 | |
| 
 | |
| bytes BinaryTransform::operator()(GlobalAssignment const& _assignment)
 | |
| {
 | |
| 	return
 | |
| 		std::visit(*this, *_assignment.value) +
 | |
| 		toBytes(Opcode::GlobalSet) +
 | |
| 		lebEncode(m_globalIDs.at(_assignment.variableName));
 | |
| }
 | |
| 
 | |
| bytes BinaryTransform::operator()(If const& _if)
 | |
| {
 | |
| 	bytes result =
 | |
| 		std::visit(*this, *_if.condition) +
 | |
| 		toBytes(Opcode::If) +
 | |
| 		toBytes(ValueType::Void);
 | |
| 
 | |
| 	m_labels.emplace_back();
 | |
| 
 | |
| 	result += visit(_if.statements);
 | |
| 	if (_if.elseStatements)
 | |
| 		result += toBytes(Opcode::Else) + visit(*_if.elseStatements);
 | |
| 
 | |
| 	m_labels.pop_back();
 | |
| 
 | |
| 	result += toBytes(Opcode::End);
 | |
| 	return result;
 | |
| }
 | |
| 
 | |
| bytes BinaryTransform::operator()(Loop const& _loop)
 | |
| {
 | |
| 	bytes result = toBytes(Opcode::Loop) + toBytes(ValueType::Void);
 | |
| 
 | |
| 	m_labels.emplace_back(_loop.labelName);
 | |
| 	result += visit(_loop.statements);
 | |
| 	m_labels.pop_back();
 | |
| 
 | |
| 	result += toBytes(Opcode::End);
 | |
| 	return result;
 | |
| }
 | |
| 
 | |
| bytes BinaryTransform::operator()(Branch const& _branch)
 | |
| {
 | |
| 	return toBytes(Opcode::Br) + encodeLabelIdx(_branch.label.name);
 | |
| }
 | |
| 
 | |
| bytes BinaryTransform::operator()(BranchIf const& _branchIf)
 | |
| {
 | |
| 	bytes result = std::visit(*this, *_branchIf.condition);
 | |
| 	result += toBytes(Opcode::BrIf) + encodeLabelIdx(_branchIf.label.name);
 | |
| 	return result;
 | |
| }
 | |
| 
 | |
| bytes BinaryTransform::operator()(Return const&)
 | |
| {
 | |
| 	// Note that this does not work if the function returns a value.
 | |
| 	return toBytes(Opcode::Return);
 | |
| }
 | |
| 
 | |
| bytes BinaryTransform::operator()(Block const& _block)
 | |
| {
 | |
| 	m_labels.emplace_back(_block.labelName);
 | |
| 	bytes result =
 | |
| 		toBytes(Opcode::Block) +
 | |
| 		toBytes(ValueType::Void) +
 | |
| 		visit(_block.statements) +
 | |
| 		toBytes(Opcode::End);
 | |
| 	m_labels.pop_back();
 | |
| 	return result;
 | |
| }
 | |
| 
 | |
| bytes BinaryTransform::operator()(FunctionDefinition const& _function)
 | |
| {
 | |
| 	bytes ret;
 | |
| 
 | |
| 	vector<pair<size_t, ValueType>> localEntries = groupLocalVariables(_function.locals);
 | |
| 	ret += lebEncode(localEntries.size());
 | |
| 	for (pair<size_t, ValueType> const& entry: localEntries)
 | |
| 	{
 | |
| 		ret += lebEncode(entry.first);
 | |
| 		ret += toBytes(entry.second);
 | |
| 	}
 | |
| 
 | |
| 	m_locals.clear();
 | |
| 	size_t varIdx = 0;
 | |
| 	for (size_t i = 0; i < _function.parameters.size(); ++i)
 | |
| 		m_locals[_function.parameters[i].name] = varIdx++;
 | |
| 	for (size_t i = 0; i < _function.locals.size(); ++i)
 | |
| 		m_locals[_function.locals[i].variableName] = varIdx++;
 | |
| 
 | |
| 	yulAssert(m_labels.empty(), "Stray labels.");
 | |
| 
 | |
| 	ret += visit(_function.body);
 | |
| 	ret += toBytes(Opcode::End);
 | |
| 
 | |
| 	yulAssert(m_labels.empty(), "Stray labels.");
 | |
| 
 | |
| 	return prefixSize(move(ret));
 | |
| }
 | |
| 
 | |
| BinaryTransform::Type BinaryTransform::typeOf(FunctionImport const& _import)
 | |
| {
 | |
| 	return {
 | |
| 		encodeTypes(_import.paramTypes),
 | |
| 		encodeTypes(_import.returnType ? vector<wasm::Type>(1, *_import.returnType) : vector<wasm::Type>())
 | |
| 	};
 | |
| }
 | |
| 
 | |
| BinaryTransform::Type BinaryTransform::typeOf(FunctionDefinition const& _funDef)
 | |
| {
 | |
| 	return {
 | |
| 		encodeTypes(_funDef.parameters),
 | |
| 		encodeTypes(_funDef.returnType ? vector<wasm::Type>(1, *_funDef.returnType) : vector<wasm::Type>())
 | |
| 	};
 | |
| }
 | |
| 
 | |
| uint8_t BinaryTransform::encodeType(wasm::Type _type)
 | |
| {
 | |
| 	return uint8_t(toValueType(_type));
 | |
| }
 | |
| 
 | |
| vector<uint8_t> BinaryTransform::encodeTypes(vector<wasm::Type> const& _types)
 | |
| {
 | |
| 	vector<uint8_t> result;
 | |
| 	for (wasm::Type t: _types)
 | |
| 		result.emplace_back(encodeType(t));
 | |
| 	return result;
 | |
| }
 | |
| 
 | |
| vector<uint8_t> BinaryTransform::encodeTypes(wasm::TypedNameList const& _typedNameList)
 | |
| {
 | |
| 	vector<uint8_t> result;
 | |
| 	for (TypedName const& typedName: _typedNameList)
 | |
| 		result.emplace_back(encodeType(typedName.type));
 | |
| 	return result;
 | |
| }
 | |
| 
 | |
| map<BinaryTransform::Type, vector<string>> BinaryTransform::typeToFunctionMap(
 | |
| 	vector<wasm::FunctionImport> const& _imports,
 | |
| 	vector<wasm::FunctionDefinition> const& _functions
 | |
| )
 | |
| {
 | |
| 	map<Type, vector<string>> types;
 | |
| 	for (auto const& import: _imports)
 | |
| 		types[typeOf(import)].emplace_back(import.internalName);
 | |
| 	for (auto const& fun: _functions)
 | |
| 		types[typeOf(fun)].emplace_back(fun.name);
 | |
| 
 | |
| 	return types;
 | |
| }
 | |
| 
 | |
| map<string, size_t> BinaryTransform::enumerateGlobals(Module const& _module)
 | |
| {
 | |
| 	map<string, size_t> globals;
 | |
| 	for (size_t i = 0; i < _module.globals.size(); ++i)
 | |
| 		globals[_module.globals[i].variableName] = i;
 | |
| 
 | |
| 	return globals;
 | |
| }
 | |
| 
 | |
| map<string, size_t> BinaryTransform::enumerateFunctions(Module const& _module)
 | |
| {
 | |
| 	map<string, size_t> functions;
 | |
| 	size_t funID = 0;
 | |
| 	for (FunctionImport const& fun: _module.imports)
 | |
| 		functions[fun.internalName] = funID++;
 | |
| 	for (FunctionDefinition const& fun: _module.functions)
 | |
| 		functions[fun.name] = funID++;
 | |
| 
 | |
| 	return functions;
 | |
| }
 | |
| 
 | |
| map<string, size_t> BinaryTransform::enumerateFunctionTypes(map<Type, vector<string>> const& _typeToFunctionMap)
 | |
| {
 | |
| 	map<string, size_t> functionTypes;
 | |
| 	size_t typeID = 0;
 | |
| 	for (vector<string> const& funNames: _typeToFunctionMap | boost::adaptors::map_values)
 | |
| 	{
 | |
| 		for (string const& name: funNames)
 | |
| 			functionTypes[name] = typeID;
 | |
| 		++typeID;
 | |
| 	}
 | |
| 
 | |
| 	return functionTypes;
 | |
| }
 | |
| 
 | |
| bytes BinaryTransform::typeSection(map<BinaryTransform::Type, vector<string>> const& _typeToFunctionMap)
 | |
| {
 | |
| 	bytes result;
 | |
| 	size_t index = 0;
 | |
| 	for (Type const& type: _typeToFunctionMap | boost::adaptors::map_keys)
 | |
| 	{
 | |
| 		result += toBytes(ValueType::Function);
 | |
| 		result += lebEncode(type.first.size()) + type.first;
 | |
| 		result += lebEncode(type.second.size()) + type.second;
 | |
| 
 | |
| 		index++;
 | |
| 	}
 | |
| 
 | |
| 	return makeSection(Section::TYPE, lebEncode(index) + move(result));
 | |
| }
 | |
| 
 | |
| bytes BinaryTransform::importSection(
 | |
| 	vector<FunctionImport> const& _imports,
 | |
| 	map<string, size_t> const& _functionTypes
 | |
| )
 | |
| {
 | |
| 	bytes result = lebEncode(_imports.size());
 | |
| 	for (FunctionImport const& import: _imports)
 | |
| 	{
 | |
| 		uint8_t importKind = 0; // function
 | |
| 		result +=
 | |
| 			encodeName(import.module) +
 | |
| 			encodeName(import.externalName) +
 | |
| 			toBytes(importKind) +
 | |
| 			lebEncode(_functionTypes.at(import.internalName));
 | |
| 	}
 | |
| 	return makeSection(Section::IMPORT, move(result));
 | |
| }
 | |
| 
 | |
| bytes BinaryTransform::functionSection(
 | |
| 	vector<FunctionDefinition> const& _functions,
 | |
| 	map<string, size_t> const& _functionTypes
 | |
| )
 | |
| {
 | |
| 	bytes result = lebEncode(_functions.size());
 | |
| 	for (auto const& fun: _functions)
 | |
| 		result += lebEncode(_functionTypes.at(fun.name));
 | |
| 	return makeSection(Section::FUNCTION, move(result));
 | |
| }
 | |
| 
 | |
| bytes BinaryTransform::memorySection()
 | |
| {
 | |
| 	bytes result = lebEncode(1);
 | |
| 	result.push_back(static_cast<uint8_t>(LimitsKind::Min));
 | |
| 	result.push_back(1); // initial length
 | |
| 	return makeSection(Section::MEMORY, move(result));
 | |
| }
 | |
| 
 | |
| bytes BinaryTransform::globalSection(vector<wasm::GlobalVariableDeclaration> const& _globals)
 | |
| {
 | |
| 	bytes result = lebEncode(_globals.size());
 | |
| 	for (wasm::GlobalVariableDeclaration const& global: _globals)
 | |
| 	{
 | |
| 		ValueType globalType = toValueType(global.type);
 | |
| 		result +=
 | |
| 			toBytes(globalType) +
 | |
| 			lebEncode(static_cast<uint8_t>(Mutability::Var)) +
 | |
| 			toBytes(constOpcodeFor(globalType)) +
 | |
| 			lebEncodeSigned(0) +
 | |
| 			toBytes(Opcode::End);
 | |
| 	}
 | |
| 
 | |
| 	return makeSection(Section::GLOBAL, move(result));
 | |
| }
 | |
| 
 | |
| bytes BinaryTransform::exportSection(map<string, size_t> const& _functionIDs)
 | |
| {
 | |
| 	bool hasMain = _functionIDs.count("main");
 | |
| 	bytes result = lebEncode(hasMain ? 2 : 1);
 | |
| 	result += encodeName("memory") + toBytes(Export::Memory) + lebEncode(0);
 | |
| 	if (hasMain)
 | |
| 		result += encodeName("main") + toBytes(Export::Function) + lebEncode(_functionIDs.at("main"));
 | |
| 	return makeSection(Section::EXPORT, move(result));
 | |
| }
 | |
| 
 | |
| bytes BinaryTransform::customSection(string const& _name, bytes _data)
 | |
| {
 | |
| 	bytes result = encodeName(_name) + move(_data);
 | |
| 	return makeSection(Section::CUSTOM, move(result));
 | |
| }
 | |
| 
 | |
| bytes BinaryTransform::codeSection(vector<wasm::FunctionDefinition> const& _functions)
 | |
| {
 | |
| 	bytes result = lebEncode(_functions.size());
 | |
| 	for (FunctionDefinition const& fun: _functions)
 | |
| 		result += (*this)(fun);
 | |
| 	return makeSection(Section::CODE, move(result));
 | |
| }
 | |
| 
 | |
| bytes BinaryTransform::visit(vector<Expression> const& _expressions)
 | |
| {
 | |
| 	bytes result;
 | |
| 	for (auto const& expr: _expressions)
 | |
| 		result += std::visit(*this, expr);
 | |
| 	return result;
 | |
| }
 | |
| 
 | |
| bytes BinaryTransform::visitReversed(vector<Expression> const& _expressions)
 | |
| {
 | |
| 	bytes result;
 | |
| 	for (auto const& expr: _expressions | boost::adaptors::reversed)
 | |
| 		result += std::visit(*this, expr);
 | |
| 	return result;
 | |
| }
 | |
| 
 | |
| bytes BinaryTransform::encodeLabelIdx(string const& _label) const
 | |
| {
 | |
| 	yulAssert(!_label.empty(), "Empty label.");
 | |
| 	size_t depth = 0;
 | |
| 	for (string const& label: m_labels | boost::adaptors::reversed)
 | |
| 		if (label == _label)
 | |
| 			return lebEncode(depth);
 | |
| 		else
 | |
| 			++depth;
 | |
| 	yulAssert(false, "Label not found.");
 | |
| }
 | |
| 
 | |
| bytes BinaryTransform::encodeName(string const& _name)
 | |
| {
 | |
| 	// UTF-8 is allowed here by the Wasm spec, but since all names here should stem from
 | |
| 	// Solidity or Yul identifiers or similar, non-ascii characters ending up here
 | |
| 	// is a very bad sign.
 | |
| 	for (char c: _name)
 | |
| 		yulAssert(uint8_t(c) <= 0x7f, "Non-ascii character found.");
 | |
| 	return lebEncode(_name.size()) + asBytes(_name);
 | |
| }
 |