mirrors/z3
3
0
Fork 0
mirror of https://github.com/Z3Prover/z3 synced 2025-11-21 21:26:40 +00:00
Code Activity

Merge branch 'parallel' of https://github.com/Z3Prover/z3 into param-tuning

Browse source
This commit is contained in:
Ilana Shapiro 2025-11-05 09:15:15 -08:00
commit 78844a05e9
5 changed files with 22 additions and 25 deletions
Download patch file Download diff file Expand all files Collapse all files

10
src/smt/smt_context.cpp
View file

@ -3617,7 +3617,7 @@ namespace smt {
\remark A logical context can only be configured at scope level 0,
and before internalizing any formulas.
*/
lbool context::setup_and_check(bool reset_cancel, bool enable_parallel_param_tuning) {
lbool context::setup_and_check(bool reset_cancel) {
if (!check_preamble(reset_cancel)) return l_undef;
SASSERT(m_scope_lvl == 0);
SASSERT(!m_setup.already_configured());
@ -3625,7 +3625,7 @@ namespace smt {
if (m_fparams.m_threads > 1 && !m.has_trace_stream()) {
expr_ref_vector asms(m);
parallel p(*this, enable_parallel_param_tuning);
parallel p(*this);
return p(asms);
}
@ -3685,15 +3685,15 @@ namespace smt {
}
}
lbool context::check(unsigned num_assumptions, expr * const * assumptions, bool reset_cancel, bool enable_parallel_param_tuning) {
lbool context::check(unsigned num_assumptions, expr * const * assumptions, bool reset_cancel) {
if (!check_preamble(reset_cancel)) return l_undef;
SASSERT(at_base_level());
setup_context(false);
search_completion sc(*this);
if (m_fparams.m_threads > 1 && !m.has_trace_stream()) {
expr_ref_vector asms(m, num_assumptions, assumptions);
IF_VERBOSE(1, verbose_stream() << "Starting parallel check with " << asms.size() << " assumptions and param tuning enabled: " << enable_parallel_param_tuning << "\n");
parallel p(*this, enable_parallel_param_tuning);
IF_VERBOSE(1, verbose_stream() << "Starting parallel check with " << asms.size() << " assumptions\n");
parallel p(*this);
return p(asms);
}
lbool r = l_undef;

4
src/smt/smt_context.h
View file

@ -1690,7 +1690,7 @@ namespace smt {
void pop(unsigned num_scopes);
lbool check(unsigned num_assumptions = 0, expr * const * assumptions = nullptr, bool reset_cancel = true, bool enable_parallel_param_tuning = true);
lbool check(unsigned num_assumptions = 0, expr * const * assumptions = nullptr, bool reset_cancel = true);
lbool check(expr_ref_vector const& cube, vector<expr_ref_vector> const& clauses);
@ -1700,7 +1700,7 @@ namespace smt {
lbool preferred_sat(expr_ref_vector const& asms, vector<expr_ref_vector>& cores);
lbool setup_and_check(bool reset_cancel = true, bool enable_parallel_param_tuning = true);
lbool setup_and_check(bool reset_cancel = true);
void reduce_assertions();

3
src/smt/smt_internalizer.cpp
View file

@ -990,8 +990,7 @@ namespace smt {
void context::undo_mk_bool_var() {
SASSERT(!m_b_internalized_stack.empty(ue key per literal
m_lit_scores[lit.sign()][v] += 1.));
SASSERT(!m_b_internalized_stack.empty());
m_stats.m_num_del_bool_var++;
expr * n = m_b_internalized_stack.back();
unsigned n_id = n->get_id();

21
src/smt/smt_parallel.cpp
View file

@ -68,7 +68,7 @@ namespace smt {
lbool parallel::param_generator::run_prefix_step() {
if (m.limit().is_canceled())
return l_undef;
IF_VERBOSE(1, verbose_stream() << " PARAM TUNER running prefix step\n");
IF_VERBOSE(1, verbose_stream() << " PARAM TUNER running prefix step with conflicts " << m_max_prefix_conflicts <<"\n");
ctx->get_fparams().m_max_conflicts = m_max_prefix_conflicts;
ctx->get_fparams().m_threads = 1;
@ -76,7 +76,7 @@ namespace smt {
ctx->m_recorded_cubes = &m_recorded_cubes;
lbool r = l_undef;
try {
r = ctx->check(0, nullptr, true, false);
r = ctx->check(0, nullptr, true);
IF_VERBOSE(1, verbose_stream() << " PARAM TUNER: prefix step result " << r << "\n");
}
catch (z3_error &err) {
@ -126,7 +126,7 @@ namespace smt {
if (m.limit().is_canceled())
return;
// the conflicts and decisions are cumulative over all cube replays inside the probe_ctx
lbool r = probe_ctx->check(cube.size(), cube.data(), true, false);
lbool r = probe_ctx->check(cube.size(), cube.data(), true);
IF_VERBOSE(2, verbose_stream() << " PARAM TUNER " << i << ": cube replay result " << r << "\n");
}
unsigned conflicts = probe_ctx->m_stats.m_num_conflicts;
@ -340,10 +340,9 @@ namespace smt {
parallel::param_generator::param_generator(parallel& p)
: b(p.m_batch_manager), m_p(p.ctx.get_params()), m_l2g(m, p.ctx.m) {
// patch fix so that ctx = alloc(context, m, p.ctx.get_fparams(), m_p); doesn't crash due to some issue with default construction of m
ast_translation m_g2l(p.ctx.m, m);
m_g2l(p.ctx.m.mk_true());
m.copy_families_plugins(p.ctx.m);
SASSERT(p.ctx.get_fparams().m_threads == 1);
ctx = alloc(context, m, p.ctx.get_fparams(), m_p);
context::copy(p.ctx, *ctx, true);
// don't share initial units
@ -539,7 +538,7 @@ namespace smt {
<< bounded_pp_exprs(cube)
<< "with max_conflicts: " << ctx->get_fparams().m_max_conflicts << "\n";);
try {
r = ctx->check(asms.size(), asms.data(), true, false);
r = ctx->check(asms.size(), asms.data(), true);
} catch (z3_error &err) {
b.set_exception(err.error_code());
} catch (z3_exception &ex) {
@ -692,6 +691,7 @@ namespace smt {
scoped_clear(parallel &p) : p(p) {}
~scoped_clear() {
p.m_workers.reset();
p.m_param_generator = nullptr;
}
};
scoped_clear clear(*this);
@ -701,6 +701,8 @@ namespace smt {
scoped_limits sl(m.limit());
flet<unsigned> _nt(ctx.m_fparams.m_threads, 1);
m_param_generator = alloc(param_generator, *this);
SASSERT(num_threads > 1);
for (unsigned i = 0; i < num_threads; ++i)
m_workers.push_back(alloc(worker, i, *this, asms));
@ -711,13 +713,12 @@ namespace smt {
sl.push_child(&(m_param_generator->limit()));
// Launch threads
vector<std::thread> threads(m_enable_param_tuner ? num_threads + 1 : num_threads); // +1 for param generator
vector<std::thread> threads(num_threads + 1); // +1 for param generator
for (unsigned i = 0; i < num_threads; ++i) {
threads[i] = std::thread([&, i]() { m_workers[i]->run(); });
}
// the final thread runs the parameter generator
if (m_enable_param_tuner)
threads[num_threads] = std::thread([&]() { m_param_generator->protocol_iteration(); });
threads[num_threads] = std::thread([&]() { m_param_generator->protocol_iteration(); });
// Wait for all threads to finish
for (auto &th : threads)

9
src/smt/smt_parallel.h
View file

@ -36,7 +36,6 @@ namespace smt {
class parallel {
context& ctx;
unsigned num_threads;
bool m_enable_param_tuner;
struct shared_clause {
unsigned source_worker_id;
@ -136,7 +135,7 @@ namespace smt {
scoped_ptr<context> ctx;
unsigned N = 4; // number of prefix permutations to test (including current)
unsigned m_max_prefix_conflicts = 1000;
unsigned m_max_prefix_conflicts = 1000; // todo- maybe make this adaptive to grow up to 1000 but start with a smaller base
scoped_ptr<context> m_prefix_solver;
vector<expr_ref_vector> m_recorded_cubes;
@ -222,14 +221,12 @@ namespace smt {
scoped_ptr<param_generator> m_param_generator;
public:
parallel(context& ctx, bool enable_param_tuner = true) :
parallel(context& ctx) :
ctx(ctx),
num_threads(std::min(
(unsigned)std::thread::hardware_concurrency(),
ctx.get_fparams().m_threads)),
m_enable_param_tuner(enable_param_tuner),
m_batch_manager(ctx.m, *this),
m_param_generator(enable_param_tuner ? alloc(param_generator, *this) : nullptr) {}
m_batch_manager(ctx.m, *this) {}
lbool operator()(expr_ref_vector const& asms);
};