From c4684d0d849798bf7543131ce51158cd90cce82e Mon Sep 17 00:00:00 2001 From: Maxime Coste Date: Sat, 15 Jun 2024 14:03:10 +1000 Subject: Store instruction pointers directly in ThreadedRegexVM::Thread The previous tradeoff of having a very small Thread struct is not necessary anymore as we do not memcpy Threads on swap_next since d708b77186c1685dcbd2298246ada7d204acec2f. This requires offsets to be used instead of indices for jump/split ops. --- src/regex_impl.cc | 47 +++++++++++++++++++++++++++++------------------ 1 file changed, 29 insertions(+), 18 deletions(-) (limited to 'src/regex_impl.cc') diff --git a/src/regex_impl.cc b/src/regex_impl.cc index 3388cde5..34cd104a 100644 --- a/src/regex_impl.cc +++ b/src/regex_impl.cc @@ -706,7 +706,7 @@ private: { auto& node = get_node(index); - const uint32_t start_pos = (uint32_t)m_program.instructions.size(); + const OpIndex start_pos = op_count(); const bool ignore_case = node.ignore_case; const bool save = (node.op == ParsedRegex::Alternation or node.op == ParsedRegex::Sequence) and @@ -746,14 +746,14 @@ private: if (child != index+1) push_inst(CompiledRegex::Split); } - auto split_pos = m_program.instructions.size(); + auto split_pos = op_count(); const auto end = node.children_end; for (auto child : Children<>{m_parsed_regex, index}) { auto node = compile_node(child); if (child != index+1) - m_program.instructions[--split_pos].param.split = CompiledRegex::Param::Split{.target = node, .prioritize_parent = true}; + m_program.instructions[--split_pos].param.split = CompiledRegex::Param::Split{.offset = offset(node, split_pos), .prioritize_parent = true}; if (get_node(child).children_end != end) { auto jump = push_inst(CompiledRegex::Jump); @@ -801,8 +801,8 @@ private: break; } - for (auto& offset : goto_inner_end_offsets) - m_program.instructions[offset].param.jump_target = m_program.instructions.size(); + for (auto& index : goto_inner_end_offsets) + m_program.instructions[index].param.jump_offset = offset(op_count(), index); if (save) push_inst(CompiledRegex::Save, {.save_index=int16_t(node.value * 2 + (forward ? 1 : 0))}); @@ -810,19 +810,29 @@ private: return start_pos; } + OpIndex op_count() const + { + return static_cast(m_program.instructions.size()); + } + + static OpIndex offset(OpIndex to, OpIndex from) + { + return static_cast(to - from); + } + template OpIndex compile_node(ParsedRegex::NodeIndex index) { auto& node = get_node(index); - const OpIndex start_pos = (OpIndex)m_program.instructions.size(); + const OpIndex start_pos = op_count(); Vector goto_ends; auto& quantifier = node.quantifier; if (quantifier.allows_none()) { - auto split_pos = push_inst(CompiledRegex::Split, {.split={.target=0, .prioritize_parent=quantifier.greedy}}); + auto split_pos = push_inst(CompiledRegex::Split, {.split={.offset=0, .prioritize_parent=quantifier.greedy}}); goto_ends.push_back(split_pos); } @@ -832,18 +842,18 @@ private: inner_pos = compile_node_inner(index); if (quantifier.allows_infinite_repeat()) - push_inst(CompiledRegex::Split, {.split = {.target=inner_pos, .prioritize_parent=not quantifier.greedy}}); + push_inst(CompiledRegex::Split, {.split = {.offset=offset(inner_pos, op_count()), .prioritize_parent=not quantifier.greedy}}); // Write the node as an optional match for the min -> max counts else for (int i = std::max((int16_t)1, quantifier.min); // STILL UGLY ! i < quantifier.max; ++i) { - auto split_pos = push_inst(CompiledRegex::Split, {.split={.target=0, .prioritize_parent=quantifier.greedy}}); + auto split_pos = push_inst(CompiledRegex::Split, {.split={.offset=0, .prioritize_parent=quantifier.greedy}}); goto_ends.push_back(split_pos); compile_node_inner(index); } - for (auto offset : goto_ends) - m_program.instructions[offset].param.split.target = m_program.instructions.size(); + for (auto index : goto_ends) + m_program.instructions[index].param.split.offset = offset(op_count(), index); return start_pos; } @@ -851,7 +861,7 @@ private: OpIndex push_inst(CompiledRegex::Op op, CompiledRegex::Param param = {}) { constexpr auto max_instructions = std::numeric_limits::max(); - const auto res = m_program.instructions.size(); + const auto res = op_count(); if (res >= max_instructions) throw regex_error(format("regex compiled to more than {} instructions", max_instructions)); m_program.instructions.push_back({ op, 0, param }); @@ -1031,7 +1041,7 @@ private: { auto& inst = m_program.instructions[i]; if (is_jump(inst.op)) - m_program.instructions[inst.param.jump_target].last_step = 0xffff; // tag as jump target + m_program.instructions[i + inst.param.jump_offset].last_step = 0xffff; // tag as jump target } for (auto block_begin = begin; block_begin < end; ) @@ -1071,11 +1081,11 @@ private: String dump_regex(const CompiledRegex& program) { String res; - int count = 0; + int index = 0; for (auto& inst : program.instructions) { char buf[20]; - format_to(buf, " {:03} ", count++); + format_to(buf, " {:03} ", index); res += buf; switch (inst.op) { @@ -1095,13 +1105,13 @@ String dump_regex(const CompiledRegex& program) res += format("character type {}\n", to_underlying(inst.param.character_type)); break; case CompiledRegex::Jump: - res += format("jump {}\n", inst.param.jump_target); + res += format("jump {} ({:03})\n", inst.param.jump_offset, index + inst.param.jump_offset); break; case CompiledRegex::Split: { - res += format("split (prioritize {}) {}\n", + res += format("split (prioritize {}) {} ({:03})\n", (inst.param.split.prioritize_parent) ? "parent" : "child", - inst.param.split.target); + inst.param.split.offset, index + inst.param.split.offset); break; } case CompiledRegex::Save: @@ -1135,6 +1145,7 @@ String dump_regex(const CompiledRegex& program) case CompiledRegex::Match: res += "match\n"; } + ++index; } auto dump_start_desc = [&](const CompiledRegex::StartDesc& desc, StringView name) { res += name + " start desc: ["; -- cgit v1.2.3