xenia/kakoune
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Restore regex optimization pass by introducing basic block analysis

Browse Source 
Run the peephole optimizer on each basic block, avoiding the
previous issue that some instructions could move across their
boundaries.
This commit is contained in:
Maxime Coste 2019-12-05 21:10:14 +11:00
parent d539e8fb89
commit eb5af59d55
1 changed files with 39 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

39
src/regex_impl.cc
View File

@ -695,6 +695,7 @@ struct RegexCompiler
{
m_program.forward_start_desc = compute_start_desc<RegexMode::Forward>();
compile_node<RegexMode::Forward>(0);
optimize(0, m_program.instructions.size());
push_inst(CompiledRegex::Match);
}
@ -703,6 +704,7 @@ struct RegexCompiler
m_program.first_backward_inst = m_program.instructions.size();
m_program.backward_start_desc = compute_start_desc<RegexMode::Backward>();
compile_node<RegexMode::Backward>(0);
optimize(m_program.first_backward_inst, m_program.instructions.size());
push_inst(CompiledRegex::Match);
}
else
@ -1030,6 +1032,43 @@ private:
return std::make_unique<CompiledRegex::StartDesc>(start_desc);
}
void optimize(size_t begin, size_t end)
{
if (not (m_flags & RegexCompileFlags::Optimize))
return;
auto is_jump = [](CompiledRegex::Op op) { return op >= CompiledRegex::Op::Jump and op <= CompiledRegex::Op::Split_PrioritizeChild; };
for (auto i = begin; i < end; ++i)
{
auto& inst = m_program.instructions[i];
if (is_jump(inst.op))
m_program.instructions[inst.param].last_step = 0xffff; // tag as jump target
}
for (auto block_begin = begin; block_begin < end; )
{
auto block_end = block_begin + 1;
while (block_end < end and
not is_jump(m_program.instructions[block_end].op) and
m_program.instructions[block_end].last_step != 0xffff)
++block_end;
peephole_optimize(block_begin, block_end);
block_begin = block_end;
}
}
void peephole_optimize(size_t begin, size_t end)
{
// Move saves after all assertions on the same character
auto is_assertion = [](CompiledRegex::Op op) { return op >= CompiledRegex::LineStart; };
for (auto i = begin, j = begin + 1; j < end; ++i, ++j)
{
if (m_program.instructions[i].op == CompiledRegex::Save and
is_assertion(m_program.instructions[j].op))
std::swap(m_program.instructions[i], m_program.instructions[j]);
}
}
const ParsedRegex::Node& get_node(ParsedRegex::NodeIndex index) const
{
return m_parsed_regex.nodes[index];