mirrors/z3
3
0
Fork 0
mirror of https://github.com/Z3Prover/z3 synced 2025-10-08 00:41:56 +00:00
Code Activity

introduce shuffle

Browse source 
Signed-off-by: Nikolaj Bjorner <nbjorner@microsoft.com>
This commit is contained in:
Nikolaj Bjorner 2025-09-23 11:24:50 +03:00
parent 3c9ab28731
commit 08ef4de4a6
2 changed files with 17 additions and 7 deletions
Download patch file Download diff file Expand all files Collapse all files

20
src/math/lp/nla_core.cpp
View file

@ -636,12 +636,20 @@ void core::erase_from_to_refine(lpvar j) {
void core::init_to_refine() {
TRACE(nla_solver_details, tout << "emons:" << pp_emons(*this, m_emons););
m_to_refine.reset();
// Traversal order of monomials uses randomization to avoid bias.
// We can cache a shuffle of m_emons indices instead of using cyclic shifts
// The shuffle can survive backtracking points to enforce bias within a scope.
unsigned r = random(), sz = m_emons.number_of_monics();
for (unsigned k = 0; k < sz; k++) {
auto const & m = *(m_emons.begin() + (k + r)% sz);
unsigned sz = m_emons.number_of_monics();
// shuffle is sticky within a backtracking level
if (!m_emon_shuffle_valid || sz != m_emon_shuffle.size()) {
m_emon_shuffle.reset();
for (unsigned i = 0; i < sz; i++)
m_emon_shuffle.push_back(i);
random_gen rand(random());
shuffle(m_emon_shuffle.size(), m_emon_shuffle.data(), rand);
trail().push(value_trail(m_emon_shuffle_valid));
m_emon_shuffle_valid = true;
}
for (auto k : m_emon_shuffle) {
auto const & m = m_emons[k];
if (!check_monic(m))
insert_to_refine(m.var());
}

2
src/math/lp/nla_core.h
View file

@ -78,6 +78,8 @@ class core {
vector<ineq> m_literals;
vector<lp::equality> m_equalities;
vector<lp::fixed_equality> m_fixed_equalities;
unsigned_vector m_emon_shuffle;
bool m_emon_shuffle_valid = false;
indexed_uint_set m_to_refine;
indexed_uint_set m_monics_with_changed_bounds;
tangents m_tangents;