name
stringlengths
1
473k
code
stringlengths
7
647k
asm
stringlengths
4
3.39M
file
stringlengths
8
196
bool baryonyx::itm::solver_inequalities_01coeff<long double, baryonyx::itm::maximize_tag, baryonyx::itm::default_cost_type<long double>, false>::push_and_compute_update_row<baryonyx::bit_array, __gnu_cxx::__normal_iterator<std::pair<int, int> const*, std::vector<std::pair<int, int>, std::allocator<std::pair<int, int>>>>>(baryonyx::bit_array&, __gnu_cxx::__normal_iterator<std::pair<int, int> const*, std::vector<std::pair<int, int>, std::allocator<std::pair<int, int>>>>, __gnu_cxx::__normal_iterator<std::pair<int, int> const*, std::vector<std::pair<int, int>, std::allocator<std::pair<int, int>>>>, long double, long double, long double, long double)
bool push_and_compute_update_row(Xtype& x, Iterator first, Iterator last, Float kappa, Float delta, Float theta, Float objective_amplifier) { auto at_least_one_pi_changed{ false }; logger::log("push-update-row {} {} {}\n", kappa, delta, theta); for (; first != last; ++first) { auto k = constraint(first); bx_expects(k < m); const auto it = ap.row(k); decrease_preference(std::get<0>(it), std::get<1>(it), theta); const auto r_size = compute_reduced_costs(std::get<0>(it), std::get<1>(it), x); // Before sort and select variables, we apply the push method: for // each reduces cost, we had the cost multiply with an objective // amplifier. for (int i = 0; i != r_size; ++i) R[i].value += objective_amplifier * c(R[i].id, x); calculator_sort<Mode>(R.get(), R.get() + r_size, rng); int selected = select_variables(r_size, b[k].min, b[k].max); logger::log("constraints {}: {} <= ", k, b[k].min); for (int i = 0; i < r_size; ++i) logger::log("({} {}) ", R[i].value, R[i].id); logger::log("<= {} => Selected: {}\n", b[k].max, selected); auto pi_change = affect( *this, x, std::get<0>(it), k, selected, r_size, kappa, delta); at_least_one_pi_changed = at_least_one_pi_changed || pi_change; } return at_least_one_pi_changed; }
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x78, %rsp movq %rcx, 0x48(%rsp) leaq 0x10(%rdi), %rax movq %rdx, %r14 movq %rsi, %r15 movq %rdi, %r12 xorl %r13d, %r13d fldt 0xe0(%rsp) fldt 0xd0(%rsp) fstpt 0x6c(%rsp) fldt 0xc0(%rsp) fstpt 0x60(%rsp) fldt 0xb0(%rsp) fstpt 0x54(%rsp) movq %rax, 0x40(%rsp) fstpt 0x34(%rsp) cmpq 0x48(%rsp), %r14 je 0x287cd7 movslq (%r14), %rbx cmpl 0x78(%r12), %ebx jge 0x287ced movq 0x40(%rsp), %rsi leaq 0x20(%rsp), %rdi movl %ebx, %edx callq 0x3de66 movq 0x20(%rsp), %rdx movq 0x28(%rsp), %rsi fldt 0x6c(%rsp) movq %r12, %rdi fstpt (%rsp) callq 0x2859c8 movq 0x20(%rsp), %rdx movq 0x28(%rsp), %rsi movq %r12, %rdi movq %r15, %rcx callq 0x2859ee fldt 0x34(%rsp) movl %eax, %ebp movq 0x58(%r12), %rdi movq 0x70(%r12), %rax movl %ebp, %ecx shlq $0x5, %rcx xorl %edx, %edx cmpq %rdx, %rcx je 0x287c72 movslq 0x10(%rdi,%rdx), %rsi movq 0x8(%rax), %r8 shlq $0x4, %rsi fldt (%r8,%rsi) fmul %st(1), %st fldt (%rdi,%rdx) faddp %st, %st(1) fstpt (%rdi,%rdx) addq $0x20, %rdx jmp 0x287c4c fstp %st(0) movq 0x8(%r12), %rdx movslq %ebp, %rsi shlq $0x5, %rsi addq %rdi, %rsi callq 0x285ab4 movq 0x60(%r12), %rax movq %r12, %rdi movl %ebp, %esi movl (%rax,%rbx,8), %edx movl 0x4(%rax,%rbx,8), %ecx callq 0x285ae8 movq 0x28(%rsp), %rdx fldt 0x60(%rsp) fstpt 0x10(%rsp) movq %r12, %rdi movq %r15, %rsi movl %ebx, %ecx movl %eax, %r8d movl %ebp, %r9d fldt 0x54(%rsp) fstpt (%rsp) callq 0x285b59 fldt 0x34(%rsp) fstp %st(0) orb %al, %r13b addq $0x8, %r14 jmp 0x287bdc andb $0x1, %r13b movl %r13d, %eax addq $0x78, %rsp popq %rbx popq %r12 popq %r13 popq %r14 popq %r15 popq %rbp retq leaq 0x1d04ea(%rip), %rdi # 0x4581de leaq 0x1d277f(%rip), %rsi # 0x45a47a leaq 0x1d92d3(%rip), %rdx # 0x460fd5 leaq 0x1d936a(%rip), %rcx # 0x461073 callq 0x2813f
/quesnel[P]baryonyx/lib/src/itm-solver-inequalities-01.cpp
bool baryonyx::itm::solver_inequalities_01coeff<long double, baryonyx::itm::maximize_tag, baryonyx::itm::default_cost_type<long double>, false>::push_and_compute_update_row<baryonyx::bit_array, __gnu_cxx::__normal_iterator<int const*, std::vector<int, std::allocator<int>>>>(baryonyx::bit_array&, __gnu_cxx::__normal_iterator<int const*, std::vector<int, std::allocator<int>>>, __gnu_cxx::__normal_iterator<int const*, std::vector<int, std::allocator<int>>>, long double, long double, long double, long double)
bool push_and_compute_update_row(Xtype& x, Iterator first, Iterator last, Float kappa, Float delta, Float theta, Float objective_amplifier) { auto at_least_one_pi_changed{ false }; logger::log("push-update-row {} {} {}\n", kappa, delta, theta); for (; first != last; ++first) { auto k = constraint(first); bx_expects(k < m); const auto it = ap.row(k); decrease_preference(std::get<0>(it), std::get<1>(it), theta); const auto r_size = compute_reduced_costs(std::get<0>(it), std::get<1>(it), x); // Before sort and select variables, we apply the push method: for // each reduces cost, we had the cost multiply with an objective // amplifier. for (int i = 0; i != r_size; ++i) R[i].value += objective_amplifier * c(R[i].id, x); calculator_sort<Mode>(R.get(), R.get() + r_size, rng); int selected = select_variables(r_size, b[k].min, b[k].max); logger::log("constraints {}: {} <= ", k, b[k].min); for (int i = 0; i < r_size; ++i) logger::log("({} {}) ", R[i].value, R[i].id); logger::log("<= {} => Selected: {}\n", b[k].max, selected); auto pi_change = affect( *this, x, std::get<0>(it), k, selected, r_size, kappa, delta); at_least_one_pi_changed = at_least_one_pi_changed || pi_change; } return at_least_one_pi_changed; }
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x78, %rsp movq %rcx, 0x48(%rsp) leaq 0x10(%rdi), %rax movq %rdx, %r14 movq %rsi, %r15 movq %rdi, %r12 xorl %r13d, %r13d fldt 0xe0(%rsp) fldt 0xd0(%rsp) fstpt 0x6c(%rsp) fldt 0xc0(%rsp) fstpt 0x60(%rsp) fldt 0xb0(%rsp) fstpt 0x54(%rsp) movq %rax, 0x40(%rsp) fstpt 0x34(%rsp) cmpq 0x48(%rsp), %r14 je 0x287e5d movslq (%r14), %rbx cmpl 0x78(%r12), %ebx jge 0x287e73 movq 0x40(%rsp), %rsi leaq 0x20(%rsp), %rdi movl %ebx, %edx callq 0x3de66 movq 0x20(%rsp), %rdx movq 0x28(%rsp), %rsi fldt 0x6c(%rsp) movq %r12, %rdi fstpt (%rsp) callq 0x2859c8 movq 0x20(%rsp), %rdx movq 0x28(%rsp), %rsi movq %r12, %rdi movq %r15, %rcx callq 0x2859ee fldt 0x34(%rsp) movl %eax, %ebp movq 0x58(%r12), %rdi movq 0x70(%r12), %rax movl %ebp, %ecx shlq $0x5, %rcx xorl %edx, %edx cmpq %rdx, %rcx je 0x287df8 movslq 0x10(%rdi,%rdx), %rsi movq 0x8(%rax), %r8 shlq $0x4, %rsi fldt (%r8,%rsi) fmul %st(1), %st fldt (%rdi,%rdx) faddp %st, %st(1) fstpt (%rdi,%rdx) addq $0x20, %rdx jmp 0x287dd2 fstp %st(0) movq 0x8(%r12), %rdx movslq %ebp, %rsi shlq $0x5, %rsi addq %rdi, %rsi callq 0x285ab4 movq 0x60(%r12), %rax movq %r12, %rdi movl %ebp, %esi movl (%rax,%rbx,8), %edx movl 0x4(%rax,%rbx,8), %ecx callq 0x285ae8 movq 0x28(%rsp), %rdx fldt 0x60(%rsp) fstpt 0x10(%rsp) movq %r12, %rdi movq %r15, %rsi movl %ebx, %ecx movl %eax, %r8d movl %ebp, %r9d fldt 0x54(%rsp) fstpt (%rsp) callq 0x285b59 fldt 0x34(%rsp) fstp %st(0) orb %al, %r13b addq $0x4, %r14 jmp 0x287d62 andb $0x1, %r13b movl %r13d, %eax addq $0x78, %rsp popq %rbx popq %r12 popq %r13 popq %r14 popq %r15 popq %rbp retq leaq 0x1d0364(%rip), %rdi # 0x4581de leaq 0x1d25f9(%rip), %rsi # 0x45a47a leaq 0x1d914d(%rip), %rdx # 0x460fd5 leaq 0x1d91e4(%rip), %rcx # 0x461073 callq 0x2813f
/quesnel[P]baryonyx/lib/src/itm-solver-inequalities-01.cpp
baryonyx::itm::solver_functor<baryonyx::itm::solver_inequalities_01coeff<long double, baryonyx::itm::maximize_tag, baryonyx::itm::default_cost_type<long double>, false>, long double, baryonyx::itm::maximize_tag, baryonyx::itm::default_cost_type<long double>, baryonyx::file_observer>::operator()(std::vector<baryonyx::itm::merged_constraint, std::allocator<baryonyx::itm::merged_constraint>> const&, int, baryonyx::itm::default_cost_type<long double> const&, double)
result operator()(const std::vector<merged_constraint>& constraints, int variables, const Cost& original_costs, double cost_constant) { result r; bit_array x(variables); int best_remaining = INT_MAX; auto& p = m_ctx.parameters; auto norm_costs = normalize_costs<Float, Cost>( m_ctx, original_costs, m_rng, variables); const auto kappa_step = static_cast<Float>(p.kappa_step); const auto kappa_max = static_cast<Float>(p.kappa_max); const auto alpha = static_cast<Float>(p.alpha); const auto theta = static_cast<Float>(p.theta); const auto delta = p.delta < 0 ? compute_delta<Float, Cost>(m_ctx, norm_costs, theta, variables) : static_cast<Float>(p.delta); const auto pushing_k_factor = static_cast<Float>(p.pushing_k_factor); const auto pushing_objective_amplifier = static_cast<Float>(p.pushing_objective_amplifier); const long int w_limit = static_cast<long int>(p.w); Solver slv( m_rng, length(constraints), variables, norm_costs, constraints); compute_order compute(p.order, variables); { std::bernoulli_distribution choose_mutation( m_ctx.parameters.init_policy_random); bit_array empty_x; switch (p.init_policy) { case solver_parameters::init_policy_type::pessimistic_solve: init_with_pre_solve<Cost, Mode>( x, empty_x, m_rng, original_costs, constraints, 1.0); break; case solver_parameters::init_policy_type::optimistic_solve: init_with_pre_solve<Cost, Mode>( empty_x, x, m_rng, original_costs, constraints, 1.0); break; case solver_parameters::init_policy_type::bastert: init_with_bastert<Cost, Mode>(x, original_costs, variables, 0); break; } for (int i = 0, e = x.size(); i != e; ++i) if (choose_mutation(m_rng)) x.invert(i); } bool start_push = false; auto kappa = static_cast<Float>(p.kappa_min); Observer obs("img", slv.m, slv.n, p.limit); m_begin = std::chrono::steady_clock::now(); m_end = std::chrono::steady_clock::now(); compute.init(slv, x); for (long int i = 0; i != p.limit; ++i) { auto remaining = compute.run(slv, x, m_rng, kappa, delta, theta); obs.make_observation(slv.ap, slv.P.get(), slv.pi.get()); if (remaining == 0) { store_if_better( x, original_costs.results(x, cost_constant), i); best_remaining = remaining; start_push = true; break; } if (remaining < best_remaining) { store_if_better(x, remaining, i); best_remaining = remaining; } if (i > w_limit) kappa += kappa_step * std::pow(static_cast<Float>(remaining) / static_cast<Float>(slv.m), alpha); if (kappa > kappa_max) { r.status = result_status::kappa_max_reached; break; } if (is_timelimit_reached()) { r.status = result_status::time_limit_reached; break; } } if (!start_push) { r.status = result_status::limit_reached; } else { for (int push = 0; push < p.pushes_limit; ++push) { auto remaining = compute.push_and_run(slv, x, m_rng, pushing_k_factor * kappa, delta, theta, pushing_objective_amplifier); if (remaining == 0) store_if_better(x, original_costs.results(x, cost_constant), -push * p.pushing_iteration_limit - 1); if (is_timelimit_reached()) break; for (int iter = 0; iter < p.pushing_iteration_limit; ++iter) { remaining = compute.run(slv, x, m_rng, kappa, delta, theta); if (remaining == 0) { store_if_better( x, original_costs.results(x, cost_constant), -push * p.pushing_iteration_limit - iter - 1); break; } if (iter > p.w) kappa += kappa_step * std::pow(static_cast<Float>(remaining) / static_cast<Float>(slv.m), alpha); if (kappa > kappa_max) break; if (is_timelimit_reached()) break; } } } if (m_best.remaining_constraints == 0) r.status = result_status::success; if (!m_best.x.empty()) { r.solutions.resize(1); convert(m_best, r.solutions[0], variables); } return r; }
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x428, %rsp # imm = 0x428 vmovsd %xmm0, 0x88(%rsp) vxorpd %xmm0, %xmm0, %xmm0 movq %r8, %r13 movl %ecx, %ebx movq %rdx, %r15 movq %rsi, %r14 movq %rdi, 0x60(%rsp) vmovupd %zmm0, 0x58(%rdi) vmovupd %zmm0, 0x40(%rdi) vmovupd %zmm0, (%rdi) movq $0x7fffffff, 0x98(%rdi) # imm = 0x7FFFFFFF movl $0x2, 0xa0(%rdi) leaq 0x70(%rsp), %rdi movl %ecx, %esi vzeroupper callq 0x368ae movq 0x10(%r14), %r12 movq 0x18(%r14), %rcx leaq 0xb8(%rsp), %rdi movq %r12, %rsi movq %r13, %rdx movl %ebx, %r8d callq 0x8d226 fldl 0x20(%r12) fstpt 0xd8(%rsp) vxorpd %xmm1, %xmm1, %xmm1 fldl 0x28(%r12) fstpt 0xe4(%rsp) fldl 0x30(%r12) fstpt 0xcc(%rsp) fldl 0x8(%r12) vmovsd 0x10(%r12), %xmm0 fld %st(0) fstpt 0x54(%rsp) vucomisd %xmm0, %xmm1 jbe 0x289a64 movq 0x10(%r14), %rdi leaq 0xb8(%rsp), %rsi movl %ebx, %edx fstpt (%rsp) callq 0x8d304 fstpt 0x9c(%rsp) jmp 0x289a7d fstp %st(0) vmovsd %xmm0, 0xf8(%rsp) fldl 0xf8(%rsp) fstpt 0x9c(%rsp) fldl 0x38(%r12) fstpt 0x10c(%rsp) fldl 0x40(%r12) fstpt 0x100(%rsp) vcvttsd2si 0xa0(%r12), %rax movq 0x18(%r14), %rsi movq %rax, 0x90(%rsp) movq 0x8(%r15), %rax subq (%r15), %rax pushq $0x28 popq %rcx cqto idivq %rcx leaq 0x150(%rsp), %rdi leaq 0xb8(%rsp), %r8 movl %eax, %edx movl %ebx, %ecx movq %r15, %r9 callq 0x284acc movl 0xcc(%r12), %esi leaq 0x118(%rsp), %rdi movl %ebx, %edx callq 0x38cd8 movq 0x10(%r14), %rax movl %ebx, 0x98(%rsp) vmovsd 0x48(%rax), %xmm0 vmovsd %xmm0, 0x48(%rsp) vxorpd %xmm0, %xmm0, %xmm0 vmovapd %xmm0, 0x1d0(%rsp) movl 0xd4(%r12), %eax testl %eax, %eax je 0x289b5f cmpl $0x1, %eax je 0x289b39 cmpl $0x2, %eax jne 0x289b70 movq 0x18(%r14), %rdx leaq 0x1d0(%rsp), %rdi leaq 0x70(%rsp), %rsi jmp 0x289b4a movq 0x18(%r14), %rdx leaq 0x70(%rsp), %rdi leaq 0x1d0(%rsp), %rsi vmovsd 0x1ca4e6(%rip), %xmm0 # 0x454038 movq %r13, %rcx movq %r15, %r8 callq 0x8b497 jmp 0x289b70 leaq 0x70(%rsp), %rdi movq %r13, %rsi movl %ebx, %edx xorl %ecx, %ecx callq 0x8b41f leaq 0x70(%rsp), %r15 movq %r13, 0x80(%rsp) xorl %ebx, %ebx movl (%r15), %ebp cmpl %ebx, %ebp je 0x289ba9 movq 0x18(%r14), %rdi callq 0x36ac3 vmovsd 0x48(%rsp), %xmm1 vucomisd %xmm0, %xmm1 jbe 0x289ba5 movq %r15, %rdi movl %ebx, %esi callq 0x3626e incl %ebx jmp 0x289b82 leaq 0x1d8(%rsp), %rdi callq 0x357fe vmovsd 0x18(%r12), %xmm0 movl 0x1c8(%rsp), %ecx movl 0x1cc(%rsp), %r8d movq 0xb0(%r12), %r9 vmovsd %xmm0, 0x48(%rsp) leaq 0x1d0996(%rip), %rdx # 0x45a577 leaq 0x1d0(%rsp), %rdi pushq $0x3 popq %rsi callq 0x4346e callq 0xb5a0 movq %rax, (%r14) callq 0xb5a0 movq %rax, 0x8(%r14) leaq 0x118(%rsp), %rdi leaq 0x150(%rsp), %rsi leaq 0x70(%rsp), %rdx callq 0x284c76 vmovsd 0x48(%rsp), %xmm0 movl $0x7fffffff, %ebp # imm = 0x7FFFFFFF leaq 0x70(%rsp), %r13 xorl %r15d, %r15d vmovsd %xmm0, 0xf0(%rsp) fldl 0xf0(%rsp) fldt 0x54(%rsp) cmpq 0xb0(%r12), %r15 je 0x289d52 movq 0x18(%r14), %rcx fstpt 0x20(%rsp) leaq 0x118(%rsp), %rdi leaq 0x150(%rsp), %rsi movq %r13, %rdx fldt 0x9c(%rsp) fstpt 0x10(%rsp) fld %st(0) fstpt 0x48(%rsp) fstpt (%rsp) callq 0x284cb8 movq 0x1a0(%rsp), %rdx movq 0x1b8(%rsp), %rcx movl %eax, %ebx leaq 0x1d0(%rsp), %rdi leaq 0x160(%rsp), %rsi callq 0x95692 testl %ebx, %ebx je 0x289e06 cmpl %ebp, %ebx jge 0x289cca movq %r14, %rdi movq %r13, %rsi movl %ebx, %edx movq %r15, %rcx callq 0x28a8c0 movl %ebx, %ebp cmpq 0x90(%rsp), %r15 jle 0x289d1b movl %ebx, 0xb0(%rsp) fildl 0xb0(%rsp) movl 0x1c8(%rsp), %eax movl %eax, 0xb4(%rsp) fidivl 0xb4(%rsp) fldt 0xcc(%rsp) fstpt 0x10(%rsp) fstpt (%rsp) callq 0xb110 fldt 0xd8(%rsp) fmulp %st, %st(1) fldt 0x48(%rsp) faddp %st, %st(1) jmp 0x289d1f fldt 0x48(%rsp) fldt 0xe4(%rsp) fxch %st(1) fucomi %st(1), %st fstp %st(1) fldz ja 0x289d52 fstp %st(0) movq %r14, %rdi fstpt 0x48(%rsp) callq 0x28a8f8 fldt 0x48(%rsp) fldt 0x54(%rsp) testb %al, %al jne 0x289d52 incq %r15 jmp 0x289c43 fstp %st(0) fstp %st(0) movq 0x60(%rsp), %rax movl $0x5, 0xa0(%rax) fldz fldz fstp %st(1) fstp %st(0) cmpl $0x0, 0x50(%r14) jne 0x289d80 movq 0x60(%rsp), %rax andl $0x0, 0xa0(%rax) cmpq $0x0, 0x28(%r14) je 0x289db1 movq 0x60(%rsp), %rax leaq 0x68(%rax), %rbx pushq $0x1 popq %rsi movq %rbx, %rdi callq 0x35292 movq (%rbx), %rsi addq $0x20, %r14 movl 0x98(%rsp), %edx movq %r14, %rdi callq 0x35781 leaq 0x1d0(%rsp), %rdi callq 0x435d0 leaq 0x118(%rsp), %rdi callq 0x393ac leaq 0x150(%rsp), %rdi callq 0x28518c leaq 0xc0(%rsp), %rdi callq 0x8affa leaq 0x78(%rsp), %rdi callq 0x357fe movq 0x60(%rsp), %rax addq $0x428, %rsp # imm = 0x428 popq %rbx popq %r12 popq %r13 popq %r14 popq %r15 popq %rbp retq vmovsd 0x88(%rsp), %xmm0 movq 0x80(%rsp), %rdi leaq 0x70(%rsp), %rbx movq %rbx, %rsi callq 0x8b74a movq %r14, %rdi movq %rbx, %rsi movq %r15, %rdx callq 0x28a876 movl $0x0, 0x6c(%rsp) leaq 0x150(%rsp), %r15 leaq 0x70(%rsp), %rbp xorl %ebx, %ebx fldt 0x54(%rsp) fldt 0x48(%rsp) cmpl 0xbc(%r12), %ebx jge 0x289d69 movq 0x18(%r14), %rcx fldt 0x10c(%rsp) fld %st(1) fstpt 0x48(%rsp) fmulp %st, %st(1) fldt 0x100(%rsp) fstpt 0x30(%rsp) fxch %st(1) fstpt 0x20(%rsp) leaq 0x118(%rsp), %rdi movq %r15, %rsi movq %rbp, %rdx fldt 0x9c(%rsp) fstpt 0x10(%rsp) fstpt (%rsp) callq 0x284ee8 testl %eax, %eax jne 0x289edc vmovsd 0x88(%rsp), %xmm0 movq 0x80(%rsp), %rdi movq %rbp, %rsi callq 0x8b74a movl 0xc0(%r12), %eax imull %ebx, %eax notl %eax movslq %eax, %rdx movq %r14, %rdi movq %rbp, %rsi callq 0x28a876 movq %r14, %rdi callq 0x28a8f8 fldt 0x54(%rsp) fldt 0x48(%rsp) testb %al, %al jne 0x289d69 movl %ebx, 0x90(%rsp) pushq $0x1 popq %rax movl %eax, %ebx leal -0x1(%rbx), %r13d cmpl 0xc0(%r12), %r13d jge 0x28a025 movq 0x18(%r14), %rcx fxch %st(1) fstpt 0x20(%rsp) leaq 0x118(%rsp), %rdi movq %r15, %rsi movq %rbp, %rdx fldt 0x9c(%rsp) fstpt 0x10(%rsp) fld %st(0) fstpt 0x48(%rsp) fstpt (%rsp) callq 0x284cb8 testl %eax, %eax je 0x289fd5 vcvtsi2sd %r13d, %xmm1, %xmm0 vucomisd 0xa0(%r12), %xmm0 jbe 0x289fa3 movl %eax, 0xa8(%rsp) fildl 0xa8(%rsp) movl 0x1c8(%rsp), %eax movl %eax, 0xac(%rsp) fidivl 0xac(%rsp) fldt 0xcc(%rsp) fstpt 0x10(%rsp) fstpt (%rsp) callq 0xb110 fldt 0xd8(%rsp) fmulp %st, %st(1) fldt 0x48(%rsp) faddp %st, %st(1) jmp 0x289fa7 fldt 0x48(%rsp) fldt 0xe4(%rsp) fxch %st(1) fucomi %st(1), %st fstp %st(1) ja 0x28a015 movq %r14, %rdi fstpt 0x48(%rsp) callq 0x28a8f8 testb %al, %al jne 0x28a01d fldt 0x54(%rsp) fldt 0x48(%rsp) incl %ebx jmp 0x289f00 vmovsd 0x88(%rsp), %xmm0 movq 0x80(%rsp), %rdi movq %rbp, %rsi callq 0x8b74a movl 0xc0(%r12), %eax imull 0x6c(%rsp), %eax subl %ebx, %eax movslq %eax, %rdx movq %r14, %rdi movq %rbp, %rsi callq 0x28a876 fldt 0x54(%rsp) fldt 0x48(%rsp) jmp 0x28a025 fldt 0x54(%rsp) fxch %st(1) jmp 0x28a025 fldt 0x54(%rsp) fldt 0x48(%rsp) movl 0x90(%rsp), %ebx decl 0x6c(%rsp) incl %ebx jmp 0x289e51 jmp 0x28a072 jmp 0x28a04b fstp %st(1) fstp %st(0) jmp 0x28a072 movq %rax, %rbx jmp 0x28a082 movq %rax, %rbx jmp 0x28a08f movq %rax, %rbx jmp 0x28a09c movq %rax, %rbx jmp 0x28a0a9 movq %rax, %rbx jmp 0x28a0b3 jmp 0x28a072 jmp 0x28a072 jmp 0x28a072 leaq 0x1d8(%rsp), %rdi movq %rax, %rbx callq 0x357fe jmp 0x28a082 leaq 0x1d0(%rsp), %rdi movq %rax, %rbx callq 0x435d0 leaq 0x118(%rsp), %rdi callq 0x393ac leaq 0x150(%rsp), %rdi callq 0x28518c leaq 0xc0(%rsp), %rdi callq 0x8affa leaq 0x78(%rsp), %rdi callq 0x357fe movq 0x60(%rsp), %rdi callq 0xea60 movq %rbx, %rdi callq 0xb3d0 nop
/quesnel[P]baryonyx/lib/src/itm-solver-common.hpp
baryonyx::itm::solver_functor<baryonyx::itm::solver_inequalities_01coeff<long double, baryonyx::itm::maximize_tag, baryonyx::itm::default_cost_type<long double>, false>, long double, baryonyx::itm::maximize_tag, baryonyx::itm::default_cost_type<long double>, baryonyx::none_observer>::operator()(std::vector<baryonyx::itm::merged_constraint, std::allocator<baryonyx::itm::merged_constraint>> const&, int, baryonyx::itm::default_cost_type<long double> const&, double)
result operator()(const std::vector<merged_constraint>& constraints, int variables, const Cost& original_costs, double cost_constant) { result r; bit_array x(variables); int best_remaining = INT_MAX; auto& p = m_ctx.parameters; auto norm_costs = normalize_costs<Float, Cost>( m_ctx, original_costs, m_rng, variables); const auto kappa_step = static_cast<Float>(p.kappa_step); const auto kappa_max = static_cast<Float>(p.kappa_max); const auto alpha = static_cast<Float>(p.alpha); const auto theta = static_cast<Float>(p.theta); const auto delta = p.delta < 0 ? compute_delta<Float, Cost>(m_ctx, norm_costs, theta, variables) : static_cast<Float>(p.delta); const auto pushing_k_factor = static_cast<Float>(p.pushing_k_factor); const auto pushing_objective_amplifier = static_cast<Float>(p.pushing_objective_amplifier); const long int w_limit = static_cast<long int>(p.w); Solver slv( m_rng, length(constraints), variables, norm_costs, constraints); compute_order compute(p.order, variables); { std::bernoulli_distribution choose_mutation( m_ctx.parameters.init_policy_random); bit_array empty_x; switch (p.init_policy) { case solver_parameters::init_policy_type::pessimistic_solve: init_with_pre_solve<Cost, Mode>( x, empty_x, m_rng, original_costs, constraints, 1.0); break; case solver_parameters::init_policy_type::optimistic_solve: init_with_pre_solve<Cost, Mode>( empty_x, x, m_rng, original_costs, constraints, 1.0); break; case solver_parameters::init_policy_type::bastert: init_with_bastert<Cost, Mode>(x, original_costs, variables, 0); break; } for (int i = 0, e = x.size(); i != e; ++i) if (choose_mutation(m_rng)) x.invert(i); } bool start_push = false; auto kappa = static_cast<Float>(p.kappa_min); Observer obs("img", slv.m, slv.n, p.limit); m_begin = std::chrono::steady_clock::now(); m_end = std::chrono::steady_clock::now(); compute.init(slv, x); for (long int i = 0; i != p.limit; ++i) { auto remaining = compute.run(slv, x, m_rng, kappa, delta, theta); obs.make_observation(slv.ap, slv.P.get(), slv.pi.get()); if (remaining == 0) { store_if_better( x, original_costs.results(x, cost_constant), i); best_remaining = remaining; start_push = true; break; } if (remaining < best_remaining) { store_if_better(x, remaining, i); best_remaining = remaining; } if (i > w_limit) kappa += kappa_step * std::pow(static_cast<Float>(remaining) / static_cast<Float>(slv.m), alpha); if (kappa > kappa_max) { r.status = result_status::kappa_max_reached; break; } if (is_timelimit_reached()) { r.status = result_status::time_limit_reached; break; } } if (!start_push) { r.status = result_status::limit_reached; } else { for (int push = 0; push < p.pushes_limit; ++push) { auto remaining = compute.push_and_run(slv, x, m_rng, pushing_k_factor * kappa, delta, theta, pushing_objective_amplifier); if (remaining == 0) store_if_better(x, original_costs.results(x, cost_constant), -push * p.pushing_iteration_limit - 1); if (is_timelimit_reached()) break; for (int iter = 0; iter < p.pushing_iteration_limit; ++iter) { remaining = compute.run(slv, x, m_rng, kappa, delta, theta); if (remaining == 0) { store_if_better( x, original_costs.results(x, cost_constant), -push * p.pushing_iteration_limit - iter - 1); break; } if (iter > p.w) kappa += kappa_step * std::pow(static_cast<Float>(remaining) / static_cast<Float>(slv.m), alpha); if (kappa > kappa_max) break; if (is_timelimit_reached()) break; } } } if (m_best.remaining_constraints == 0) r.status = result_status::success; if (!m_best.x.empty()) { r.solutions.resize(1); convert(m_best, r.solutions[0], variables); } return r; }
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x1d8, %rsp # imm = 0x1D8 vmovsd %xmm0, 0x80(%rsp) vxorpd %xmm0, %xmm0, %xmm0 movq %r8, %r13 movl %ecx, %ebx movq %rdx, %r15 movq %rsi, %r14 movq %rdi, 0x58(%rsp) vmovupd %zmm0, 0x58(%rdi) vmovupd %zmm0, 0x40(%rdi) vmovupd %zmm0, (%rdi) movq $0x7fffffff, 0x98(%rdi) # imm = 0x7FFFFFFF movl $0x2, 0xa0(%rdi) leaq 0x68(%rsp), %rdi movl %ecx, %esi vzeroupper callq 0x368ae movq 0x10(%r14), %r12 movq 0x18(%r14), %rcx leaq 0xc0(%rsp), %rdi movq %r12, %rsi movq %r13, %rdx movl %ebx, %r8d callq 0x8d226 fldl 0x20(%r12) fstpt 0xe0(%rsp) vxorpd %xmm1, %xmm1, %xmm1 fldl 0x28(%r12) fstpt 0xec(%rsp) fldl 0x30(%r12) fstpt 0xd4(%rsp) fldl 0x8(%r12) vmovsd 0x10(%r12), %xmm0 fld %st(0) fstpt 0x4c(%rsp) vucomisd %xmm0, %xmm1 jbe 0x28a1ac movq 0x10(%r14), %rdi leaq 0xc0(%rsp), %rsi movl %ebx, %edx fstpt (%rsp) callq 0x8d304 fstpt 0x94(%rsp) jmp 0x28a1c5 fstp %st(0) vmovsd %xmm0, 0x100(%rsp) fldl 0x100(%rsp) fstpt 0x94(%rsp) fldl 0x38(%r12) fstpt 0x114(%rsp) fldl 0x40(%r12) fstpt 0x108(%rsp) vcvttsd2si 0xa0(%r12), %rax movq 0x18(%r14), %rsi movq %rax, 0x88(%rsp) movq 0x8(%r15), %rax subq (%r15), %rax pushq $0x28 popq %rcx cqto idivq %rcx leaq 0x158(%rsp), %rdi leaq 0xc0(%rsp), %r8 movl %eax, %edx movl %ebx, %ecx movq %r15, %r9 callq 0x284acc movl 0xcc(%r12), %esi leaq 0x120(%rsp), %rdi movl %ebx, %edx callq 0x38cd8 movq 0x10(%r14), %rax movl %ebx, 0x90(%rsp) vmovsd 0x48(%rax), %xmm0 vmovsd %xmm0, 0x40(%rsp) vxorpd %xmm0, %xmm0, %xmm0 vmovapd %xmm0, 0xb0(%rsp) movl 0xd4(%r12), %eax testl %eax, %eax je 0x28a2a7 cmpl $0x1, %eax je 0x28a281 cmpl $0x2, %eax jne 0x28a2b8 movq 0x18(%r14), %rdx leaq 0xb0(%rsp), %rdi leaq 0x68(%rsp), %rsi jmp 0x28a292 movq 0x18(%r14), %rdx leaq 0x68(%rsp), %rdi leaq 0xb0(%rsp), %rsi vmovsd 0x1c9d9e(%rip), %xmm0 # 0x454038 movq %r13, %rcx movq %r15, %r8 callq 0x8b497 jmp 0x28a2b8 leaq 0x68(%rsp), %rdi movq %r13, %rsi movl %ebx, %edx xorl %ecx, %ecx callq 0x8b41f leaq 0x68(%rsp), %r15 movq %r13, 0x78(%rsp) xorl %ebx, %ebx movl (%r15), %ebp cmpl %ebx, %ebp je 0x28a2ee movq 0x18(%r14), %rdi callq 0x36ac3 vmovsd 0x40(%rsp), %xmm1 vucomisd %xmm0, %xmm1 jbe 0x28a2ea movq %r15, %rdi movl %ebx, %esi callq 0x3626e incl %ebx jmp 0x28a2c7 leaq 0xb8(%rsp), %rdi callq 0x357fe vmovsd 0x18(%r12), %xmm0 vmovsd %xmm0, 0x40(%rsp) callq 0xb5a0 movq %rax, (%r14) callq 0xb5a0 movq %rax, 0x8(%r14) leaq 0x120(%rsp), %rdi leaq 0x158(%rsp), %rsi leaq 0x68(%rsp), %rdx callq 0x284c76 vmovsd 0x40(%rsp), %xmm0 movl $0x7fffffff, %ebp # imm = 0x7FFFFFFF leaq 0x68(%rsp), %rbx xorl %r15d, %r15d vmovsd %xmm0, 0xf8(%rsp) fldl 0xf8(%rsp) fldt 0x4c(%rsp) cmpq 0xb0(%r12), %r15 je 0x28a449 movq 0x18(%r14), %rcx fstpt 0x20(%rsp) leaq 0x120(%rsp), %rdi leaq 0x158(%rsp), %rsi movq %rbx, %rdx fldt 0x94(%rsp) fstpt 0x10(%rsp) fld %st(0) fstpt 0x40(%rsp) fstpt (%rsp) callq 0x284cb8 movl %eax, %r13d testl %eax, %eax je 0x28a4f0 cmpl %ebp, %r13d jge 0x28a3c0 movq %r14, %rdi movq %rbx, %rsi movl %r13d, %edx movq %r15, %rcx callq 0x28a9a2 movl %r13d, %ebp cmpq 0x88(%rsp), %r15 jle 0x28a412 movl %r13d, 0xa8(%rsp) fildl 0xa8(%rsp) movl 0x1d0(%rsp), %eax movl %eax, 0xac(%rsp) fidivl 0xac(%rsp) fldt 0xd4(%rsp) fstpt 0x10(%rsp) fstpt (%rsp) callq 0xb110 fldt 0xe0(%rsp) fmulp %st, %st(1) fldt 0x40(%rsp) faddp %st, %st(1) jmp 0x28a416 fldt 0x40(%rsp) fldt 0xec(%rsp) fxch %st(1) fucomi %st(1), %st fstp %st(1) fldz ja 0x28a449 fstp %st(0) movq %r14, %rdi fstpt 0x40(%rsp) callq 0x28a9da fldt 0x40(%rsp) fldt 0x4c(%rsp) testb %al, %al jne 0x28a449 incq %r15 jmp 0x28a35a fstp %st(0) fstp %st(0) movq 0x58(%rsp), %rax movl $0x5, 0xa0(%rax) fldz fldz fstp %st(1) fstp %st(0) cmpl $0x0, 0x50(%r14) jne 0x28a477 movq 0x58(%rsp), %rax andl $0x0, 0xa0(%rax) cmpq $0x0, 0x28(%r14) je 0x28a4a8 movq 0x58(%rsp), %rax leaq 0x68(%rax), %rbx pushq $0x1 popq %rsi movq %rbx, %rdi callq 0x35292 movq (%rbx), %rsi addq $0x20, %r14 movl 0x90(%rsp), %edx movq %r14, %rdi callq 0x35781 leaq 0x120(%rsp), %rdi callq 0x393ac leaq 0x158(%rsp), %rdi callq 0x28518c leaq 0xc8(%rsp), %rdi callq 0x8affa leaq 0x70(%rsp), %rdi callq 0x357fe movq 0x58(%rsp), %rax addq $0x1d8, %rsp # imm = 0x1D8 popq %rbx popq %r12 popq %r13 popq %r14 popq %r15 popq %rbp retq vmovsd 0x80(%rsp), %xmm0 movq 0x78(%rsp), %rdi leaq 0x68(%rsp), %rbx movq %rbx, %rsi callq 0x8b74a movq %r14, %rdi movq %rbx, %rsi movq %r15, %rdx callq 0x28a958 movl $0x0, 0x64(%rsp) leaq 0x158(%rsp), %r15 leaq 0x68(%rsp), %rbp xorl %ebx, %ebx fldt 0x4c(%rsp) fldt 0x40(%rsp) cmpl 0xbc(%r12), %ebx jge 0x28a460 movq 0x18(%r14), %rcx fldt 0x114(%rsp) fld %st(1) fstpt 0x40(%rsp) fmulp %st, %st(1) fldt 0x108(%rsp) fstpt 0x30(%rsp) fxch %st(1) fstpt 0x20(%rsp) leaq 0x120(%rsp), %rdi movq %r15, %rsi movq %rbp, %rdx fldt 0x94(%rsp) fstpt 0x10(%rsp) fstpt (%rsp) callq 0x284ee8 testl %eax, %eax jne 0x28a5c0 vmovsd 0x80(%rsp), %xmm0 movq 0x78(%rsp), %rdi movq %rbp, %rsi callq 0x8b74a movl 0xc0(%r12), %eax imull %ebx, %eax notl %eax movslq %eax, %rdx movq %r14, %rdi movq %rbp, %rsi callq 0x28a958 movq %r14, %rdi callq 0x28a9da fldt 0x4c(%rsp) fldt 0x40(%rsp) testb %al, %al jne 0x28a460 movl %ebx, 0x88(%rsp) pushq $0x1 popq %rax movl %eax, %ebx leal -0x1(%rbx), %r13d cmpl 0xc0(%r12), %r13d jge 0x28a706 movq 0x18(%r14), %rcx fxch %st(1) fstpt 0x20(%rsp) leaq 0x120(%rsp), %rdi movq %r15, %rsi movq %rbp, %rdx fldt 0x94(%rsp) fstpt 0x10(%rsp) fld %st(0) fstpt 0x40(%rsp) fstpt (%rsp) callq 0x284cb8 testl %eax, %eax je 0x28a6b9 vcvtsi2sd %r13d, %xmm2, %xmm0 vucomisd 0xa0(%r12), %xmm0 jbe 0x28a687 movl %eax, 0xa0(%rsp) fildl 0xa0(%rsp) movl 0x1d0(%rsp), %eax movl %eax, 0xa4(%rsp) fidivl 0xa4(%rsp) fldt 0xd4(%rsp) fstpt 0x10(%rsp) fstpt (%rsp) callq 0xb110 fldt 0xe0(%rsp) fmulp %st, %st(1) fldt 0x40(%rsp) faddp %st, %st(1) jmp 0x28a68b fldt 0x40(%rsp) fldt 0xec(%rsp) fxch %st(1) fucomi %st(1), %st fstp %st(1) ja 0x28a6f6 movq %r14, %rdi fstpt 0x40(%rsp) callq 0x28a9da testb %al, %al jne 0x28a6fe fldt 0x4c(%rsp) fldt 0x40(%rsp) incl %ebx jmp 0x28a5e4 vmovsd 0x80(%rsp), %xmm0 movq 0x78(%rsp), %rdi movq %rbp, %rsi callq 0x8b74a movl 0xc0(%r12), %eax imull 0x64(%rsp), %eax subl %ebx, %eax movslq %eax, %rdx movq %r14, %rdi movq %rbp, %rsi callq 0x28a958 fldt 0x4c(%rsp) fldt 0x40(%rsp) jmp 0x28a706 fldt 0x4c(%rsp) fxch %st(1) jmp 0x28a706 fldt 0x4c(%rsp) fldt 0x40(%rsp) movl 0x88(%rsp), %ebx decl 0x64(%rsp) incl %ebx jmp 0x28a538 jmp 0x28a74e jmp 0x28a727 fstp %st(1) fstp %st(0) jmp 0x28a74e movq %rax, %rbx jmp 0x28a75e movq %rax, %rbx jmp 0x28a76b movq %rax, %rbx jmp 0x28a778 movq %rax, %rbx jmp 0x28a782 jmp 0x28a74e jmp 0x28a74e jmp 0x28a74e leaq 0xb8(%rsp), %rdi movq %rax, %rbx callq 0x357fe jmp 0x28a751 movq %rax, %rbx leaq 0x120(%rsp), %rdi callq 0x393ac leaq 0x158(%rsp), %rdi callq 0x28518c leaq 0xc8(%rsp), %rdi callq 0x8affa leaq 0x70(%rsp), %rdi callq 0x357fe movq 0x58(%rsp), %rdi callq 0xea60 movq %rbx, %rdi callq 0xb3d0
/quesnel[P]baryonyx/lib/src/itm-solver-common.hpp
baryonyx::itm::solver_functor<baryonyx::itm::solver_inequalities_01coeff<long double, baryonyx::itm::maximize_tag, baryonyx::itm::default_cost_type<long double>, false>, long double, baryonyx::itm::maximize_tag, baryonyx::itm::default_cost_type<long double>, baryonyx::pnm_observer>::store_if_better(baryonyx::bit_array const&, double, long)
inline bool is_better_solution(Float lhs, Float rhs) noexcept { if constexpr (std::is_same_v<Mode, minimize_tag>) return lhs < rhs; else return lhs > rhs; }
vucomisd 0x30(%rdi), %xmm0 jbe 0x28a7dc pushq %r14 pushq %rbx pushq %rax movq %rdi, %r14 addq $0x20, %rdi movq %rdx, %rbx vmovsd %xmm0, (%rsp) callq 0x36a44 movq %r14, %rdi callq 0x28a856 vmovsd (%rsp), %xmm1 vmovsd %xmm0, 0x38(%r14) movq %rbx, 0x48(%r14) andl $0x0, 0x50(%r14) vmovsd %xmm1, 0x30(%r14) addq $0x8, %rsp popq %rbx popq %r14 retq nop
/quesnel[P]baryonyx/lib/src/itm-common.hpp
baryonyx::itm::solver_functor<baryonyx::itm::solver_inequalities_01coeff<long double, baryonyx::itm::maximize_tag, baryonyx::itm::default_cost_type<long double>, false>, long double, baryonyx::itm::maximize_tag, baryonyx::itm::default_cost_type<long double>, baryonyx::pnm_observer>::is_timelimit_reached()
bool is_timelimit_reached() { m_end = std::chrono::steady_clock::now(); return is_time_limit(m_ctx.parameters.time_limit, m_begin, m_end); }
pushq %rbx movq %rdi, %rbx callq 0xb5a0 movq %rax, 0x8(%rbx) vxorpd %xmm2, %xmm2, %xmm2 movq 0x10(%rbx), %rcx subq (%rbx), %rax vmovsd (%rcx), %xmm0 vcvtsi2sd %rax, %xmm1, %xmm1 vdivsd 0x1cf241(%rip), %xmm1, %xmm1 # 0x459a80 vcmpnlepd %xmm2, %xmm0, %k1 vcmpltpd %xmm1, %xmm0, %k0 {%k1} kmovd %k0, %eax andb $0x1, %al popq %rbx retq nop
/quesnel[P]baryonyx/lib/src/itm-solver-common.hpp
baryonyx::itm::solver_functor<baryonyx::itm::solver_inequalities_01coeff<long double, baryonyx::itm::maximize_tag, baryonyx::itm::default_cost_type<long double>, false>, long double, baryonyx::itm::maximize_tag, baryonyx::itm::default_cost_type<long double>, baryonyx::file_observer>::store_if_better(baryonyx::bit_array const&, double, long)
inline bool is_better_solution(Float lhs, Float rhs) noexcept { if constexpr (std::is_same_v<Mode, minimize_tag>) return lhs < rhs; else return lhs > rhs; }
vucomisd 0x30(%rdi), %xmm0 jbe 0x28a8be pushq %r14 pushq %rbx pushq %rax movq %rdi, %r14 addq $0x20, %rdi movq %rdx, %rbx vmovsd %xmm0, (%rsp) callq 0x36a44 movq %r14, %rdi callq 0x28a938 vmovsd (%rsp), %xmm1 vmovsd %xmm0, 0x38(%r14) movq %rbx, 0x48(%r14) andl $0x0, 0x50(%r14) vmovsd %xmm1, 0x30(%r14) addq $0x8, %rsp popq %rbx popq %r14 retq nop
/quesnel[P]baryonyx/lib/src/itm-common.hpp
baryonyx::itm::solver_functor<baryonyx::itm::solver_inequalities_01coeff<long double, baryonyx::itm::maximize_tag, baryonyx::itm::default_cost_type<long double>, false>, long double, baryonyx::itm::maximize_tag, baryonyx::itm::default_cost_type<long double>, baryonyx::file_observer>::is_timelimit_reached()
bool is_timelimit_reached() { m_end = std::chrono::steady_clock::now(); return is_time_limit(m_ctx.parameters.time_limit, m_begin, m_end); }
pushq %rbx movq %rdi, %rbx callq 0xb5a0 movq %rax, 0x8(%rbx) vxorpd %xmm2, %xmm2, %xmm2 movq 0x10(%rbx), %rcx subq (%rbx), %rax vmovsd (%rcx), %xmm0 vcvtsi2sd %rax, %xmm1, %xmm1 vdivsd 0x1cf15f(%rip), %xmm1, %xmm1 # 0x459a80 vcmpnlepd %xmm2, %xmm0, %k1 vcmpltpd %xmm1, %xmm0, %k0 {%k1} kmovd %k0, %eax andb $0x1, %al popq %rbx retq nop
/quesnel[P]baryonyx/lib/src/itm-solver-common.hpp
baryonyx::itm::solver_functor<baryonyx::itm::solver_inequalities_01coeff<long double, baryonyx::itm::maximize_tag, baryonyx::itm::default_cost_type<long double>, false>, long double, baryonyx::itm::maximize_tag, baryonyx::itm::default_cost_type<long double>, baryonyx::none_observer>::store_if_better(baryonyx::bit_array const&, double, long)
inline bool is_better_solution(Float lhs, Float rhs) noexcept { if constexpr (std::is_same_v<Mode, minimize_tag>) return lhs < rhs; else return lhs > rhs; }
vucomisd 0x30(%rdi), %xmm0 jbe 0x28a9a0 pushq %r14 pushq %rbx pushq %rax movq %rdi, %r14 addq $0x20, %rdi movq %rdx, %rbx vmovsd %xmm0, (%rsp) callq 0x36a44 movq %r14, %rdi callq 0x28aa1a vmovsd (%rsp), %xmm1 vmovsd %xmm0, 0x38(%r14) movq %rbx, 0x48(%r14) andl $0x0, 0x50(%r14) vmovsd %xmm1, 0x30(%r14) addq $0x8, %rsp popq %rbx popq %r14 retq nop
/quesnel[P]baryonyx/lib/src/itm-common.hpp
baryonyx::itm::solver_functor<baryonyx::itm::solver_inequalities_01coeff<long double, baryonyx::itm::maximize_tag, baryonyx::itm::default_cost_type<long double>, false>, long double, baryonyx::itm::maximize_tag, baryonyx::itm::default_cost_type<long double>, baryonyx::none_observer>::is_timelimit_reached()
bool is_timelimit_reached() { m_end = std::chrono::steady_clock::now(); return is_time_limit(m_ctx.parameters.time_limit, m_begin, m_end); }
pushq %rbx movq %rdi, %rbx callq 0xb5a0 movq %rax, 0x8(%rbx) vxorpd %xmm2, %xmm2, %xmm2 movq 0x10(%rbx), %rcx subq (%rbx), %rax vmovsd (%rcx), %xmm0 vcvtsi2sd %rax, %xmm1, %xmm1 vdivsd 0x1cf07d(%rip), %xmm1, %xmm1 # 0x459a80 vcmpnlepd %xmm2, %xmm0, %k1 vcmpltpd %xmm1, %xmm0, %k0 {%k1} kmovd %k0, %eax andb $0x1, %al popq %rbx retq nop
/quesnel[P]baryonyx/lib/src/itm-solver-common.hpp
baryonyx::result baryonyx::itm::optimize_problem<baryonyx::itm::solver_inequalities_01coeff<long double, baryonyx::itm::maximize_tag, baryonyx::itm::quadratic_cost_type<long double>, true>, long double, baryonyx::itm::maximize_tag, baryonyx::itm::quadratic_cost_type<long double>>(baryonyx::context const&, baryonyx::problem const&)
inline result optimize_problem(const context& ctx, const problem& pb) { result r; if (ctx.start) ctx.start(ctx.parameters); auto constraints{ make_merged_constraints(ctx, pb) }; if (constraints.empty() || pb.vars.values.empty()) { r.status = result_status::success; r.solutions.resize(1); r.solutions.back().value = pb.objective.value; r.strings = pb.strings; r.affected_vars = pb.affected_vars; r.variable_name = pb.vars.names; return r; } random_engine rng(init_random_generator_seed(ctx)); auto variables = length(pb.vars.values); auto cost = Cost(pb.objective, variables); auto cost_constant = pb.objective.value; const auto thread = get_thread_number(ctx); std::vector<optimize_functor<Solver, Float, Mode, Cost>> functors; std::vector<std::thread> pool(thread); best_solution_recorder<Cost, Float, Mode> best_recorder( rng, thread, cost, cost_constant, variables, constraints, ctx.parameters.init_population_size); auto seeds = generate_seed(rng, thread); std::atomic_bool stop_task; stop_task.store(false); for (unsigned i = 0u; i != thread; ++i) functors.emplace_back(ctx, i, seeds[i]); for (unsigned i = 0u; i != thread; ++i) pool[i] = std::thread(std::ref(functors[i]), std::ref(stop_task), std::ref(best_recorder), std::cref(constraints), variables, std::cref(cost), cost_constant); const auto start = std::chrono::steady_clock::now(); auto end = start; do { std::this_thread::sleep_for(std::chrono::seconds{ 1L }); if (ctx.update) { auto call_number = 0L; for (auto i = 0u; i != thread; ++i) call_number += functors[i].m_call_number; int constraints_remaining; long int loop; double value; double duration; best_recorder.get_best( constraints_remaining, value, duration, loop); ctx.update( constraints_remaining, value, loop, duration, call_number); } end = std::chrono::steady_clock::now(); } while (!is_time_limit(ctx.parameters.time_limit, start, end)); stop_task.store(true); for (auto& t : pool) t.join(); r.strings = pb.strings; r.affected_vars = pb.affected_vars; r.variable_name = pb.vars.names; r.variables = variables; r.constraints = length(constraints); const auto& first = best_recorder.get_best(0); if (!first.is_solution()) r.status = result_status::time_limit_reached; else r.status = result_status::success; r.duration = first.duration; r.loop = first.loop; r.remaining_constraints = first.remaining_constraints; switch (ctx.parameters.storage) { case solver_parameters::storage_type::one: { r.solutions.resize(1); convert(first, r.solutions[0], variables); } break; case solver_parameters::storage_type::bound: { r.solutions.resize(2); convert(first, r.solutions[0], variables); convert(best_recorder.get_worst(), r.solutions[1], variables); } break; case solver_parameters::storage_type::five: { r.solutions.resize(5); for (int i = 0; i != 5; ++i) convert(best_recorder.get_best(i), r.solutions[i], variables); } break; } best_recorder.show_population(ctx); if (ctx.finish) ctx.finish(r); return r; }
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x198, %rsp # imm = 0x198 vxorpd %xmm0, %xmm0, %xmm0 movq %rdx, %r13 movq %rsi, %r15 movq %rdi, %r12 vmovupd %zmm0, 0x58(%rdi) vmovupd %zmm0, 0x40(%rdi) vmovupd %zmm0, (%rdi) movq $0x7fffffff, 0x98(%rdi) # imm = 0x7FFFFFFF movl $0x2, 0xa0(%rdi) cmpq $0x0, 0x128(%rsi) je 0x28aa9f leaq 0x118(%r15), %rdi movq %r15, %rsi vzeroupper callq 0x3527e leaq 0x48(%rsp), %rdi movq %r15, %rsi movq %r13, %rdx vzeroupper callq 0x450078 movq 0x48(%rsp), %rax cmpq 0x50(%rsp), %rax je 0x28abfe movq 0xa8(%r13), %rax cmpq 0xb0(%r13), %rax je 0x28abfe movq %r15, %rdi callq 0x353ba movl $0x7fffffff, %ecx # imm = 0x7FFFFFFF xorl %edx, %edx divq %rcx cmpq $0x1, %rdx adcq $0x0, %rdx movq %rdx, 0x40(%rsp) movq 0xb0(%r13), %rax subq 0xa8(%r13), %rax pushq $0xc popq %rcx cqto leaq 0x10(%r13), %rsi idivq %rcx movl %eax, 0x4(%rsp) leaq 0xb8(%rsp), %rdi movl %eax, %edx movq %r12, 0x10(%rsp) callq 0x9650c vmovsd 0x40(%r13), %xmm0 vmovsd %xmm0, 0x38(%rsp) movq %r15, %rdi callq 0x35447 andq $0x0, 0x70(%rsp) vxorpd %xmm0, %xmm0, %xmm0 movl %eax, %ebp vmovapd %xmm0, 0x60(%rsp) leaq 0xa0(%rsp), %rdi leaq 0xd8(%rsp), %rdx movq %rbp, %rsi callq 0x35462 vmovsd 0x38(%rsp), %xmm0 movl 0xc4(%r15), %eax movl 0x4(%rsp), %r8d subq $0x8, %rsp movl %ebp, %edx leaq 0xe0(%rsp), %rdi leaq 0x48(%rsp), %rsi leaq 0xc0(%rsp), %rcx leaq 0x50(%rsp), %r9 pushq %rax callq 0x96720 popq %rax popq %rcx leaq 0x30(%rsp), %rdi leaq 0x40(%rsp), %rsi movl %ebp, %edx callq 0x3551c xorl %eax, %eax movb %al, 0x3(%rsp) xorl %ecx, %ecx leaq 0x60(%rsp), %rbx leaq 0x8(%rsp), %r14 xchgb %cl, 0x3(%rsp) leaq 0x90(%r13), %rcx movq %rcx, 0x88(%rsp) movl %eax, 0x8(%rsp) cmpl %ebp, %eax je 0x28ac59 movl %eax, %ecx shlq $0x3, %rcx addq 0x30(%rsp), %rcx movq %rbx, %rdi movq %r15, %rsi movq %r14, %rdx callq 0x28bd3c movl 0x8(%rsp), %eax incl %eax jmp 0x28abd5 andl $0x0, 0xa0(%r12) leaq 0x68(%r12), %rdi pushq $0x1 popq %rsi callq 0x35292 vmovsd 0x40(%r13), %xmm0 movq 0x70(%r12), %rax vmovsd %xmm0, -0x8(%rax) movq %r12, %rdi movq %r13, %rsi callq 0x31d28 leaq 0xc0(%r13), %rsi leaq 0x20(%r12), %rdi callq 0x352c2 addq $0x90, %r13 leaq 0x50(%r12), %rdi movq %r13, %rsi callq 0x352ec jmp 0x28aff4 movq %r13, 0x80(%rsp) movq %r15, %r13 leaq 0x8(%rsp), %r15 leaq 0x98(%rsp), %r14 leaq 0x4(%rsp), %rbx xorl %r12d, %r12d cmpq %rbp, %r12 je 0x28ad1d imulq $0xc0, %r12, %rax addq 0x60(%rsp), %rax leaq 0x3(%rsp), %r10 leaq 0xd8(%rsp), %r11 movq %r10, 0x20(%rsp) leaq 0x48(%rsp), %r10 movq %r11, 0x18(%rsp) movq %r10, 0x98(%rsp) movq %rax, 0x28(%rsp) leaq 0xb8(%rsp), %rax movq %rax, 0x90(%rsp) leaq 0x20(%rsp), %rdx leaq 0x38(%rsp), %rax movq %r15, %rdi leaq 0x28(%rsp), %rsi leaq 0x18(%rsp), %rcx movq %r14, %r8 movq %rbx, %r9 pushq %rax leaq 0x98(%rsp), %rax pushq %rax callq 0x28bd88 popq %rax popq %rcx leaq (,%r12,8), %rdi addq 0xa0(%rsp), %rdi movq %r15, %rsi callq 0x356d6 movq %r15, %rdi callq 0x356f0 incq %r12 jmp 0x28ac79 callq 0xb5a0 imulq $0xc0, %rbp, %r15 leaq 0x138(%r13), %r12 leaq 0x8(%rsp), %rbp movq %rax, %rbx movq $0x1, 0x8(%rsp) movq %rbp, %rdi callq 0x356fd cmpq $0x0, 0x148(%r13) je 0x28adb2 movq 0x60(%rsp), %rax xorl %ecx, %ecx xorl %r14d, %r14d cmpq %rcx, %r15 je 0x28ad73 addq 0xb0(%rax,%rcx), %r14 addq $0xc0, %rcx jmp 0x28ad5d leaq 0xe0(%rsp), %rdi leaq 0x18(%rsp), %rsi leaq 0x28(%rsp), %rdx leaq 0x20(%rsp), %rcx movq %rbp, %r8 callq 0x9f4ce vmovsd 0x28(%rsp), %xmm0 vmovsd 0x20(%rsp), %xmm1 movq 0x8(%rsp), %rdx movl 0x18(%rsp), %esi movq %r12, %rdi movq %r14, %rcx callq 0x35736 callq 0xb5a0 vmovsd (%r13), %xmm0 subq %rbx, %rax vxorpd %xmm1, %xmm1, %xmm1 vucomisd %xmm0, %xmm1 vcvtsi2sd %rax, %xmm2, %xmm1 vdivsd 0x1cecab(%rip), %xmm1, %xmm1 # 0x459a80 jae 0x28ad38 vucomisd %xmm0, %xmm1 jbe 0x28ad38 movb $0x1, %al xchgb %al, 0x3(%rsp) movq 0xa0(%rsp), %rbx movq 0xa8(%rsp), %r14 movq 0x10(%rsp), %r12 cmpq %r14, %rbx je 0x28ae13 movq %rbx, %rdi callq 0xb1a0 addq $0x8, %rbx jmp 0x28ae00 movq 0x80(%rsp), %rbx movq %r12, %rdi movq %rbx, %rsi callq 0x31d28 addq $0xc0, %rbx leaq 0x20(%r12), %rdi movq %rbx, %rsi callq 0x352c2 leaq 0x50(%r12), %rdi movq 0x88(%rsp), %rsi callq 0x352ec movl 0x4(%rsp), %eax movl %eax, 0x90(%r12) movq 0x50(%rsp), %rax subq 0x48(%rsp), %rax pushq $0x28 popq %rcx cqto xorl %esi, %esi idivq %rcx movq 0x118(%rsp), %rdi movq 0x130(%rsp), %rcx movl %eax, 0x94(%r12) movslq (%rdi), %rax imulq $0x38, %rax, %rax movl 0x30(%rcx,%rax), %edx leaq (%rcx,%rax), %rbx testl %edx, %edx setne %sil leal (%rsi,%rsi,2), %esi movl %esi, 0xa0(%r12) vmovsd 0x18(%rcx,%rax), %xmm0 vmovsd %xmm0, 0x80(%r12) movq 0x28(%rcx,%rax), %rax movq %rax, 0x88(%r12) movl %edx, 0x98(%r12) movl 0xe8(%r13), %eax testl %eax, %eax je 0x28af6e cmpl $0x1, %eax je 0x28af36 cmpl $0x2, %eax jne 0x28af8d leaq 0x68(%r12), %rbx pushq $0x5 popq %rsi movq %rbx, %rdi callq 0x35292 xorl %r14d, %r14d xorl %r15d, %r15d cmpq $0x5, %r15 je 0x28af8d movq 0x118(%rsp), %rax movq (%rbx), %rsi movl 0x4(%rsp), %edx movslq (%rax,%r15,4), %rax addq %r14, %rsi imulq $0x38, %rax, %rdi addq 0x130(%rsp), %rdi callq 0x35781 incq %r15 addq $0x20, %r14 jmp 0x28aefc leaq 0x68(%r12), %r14 pushq $0x2 popq %rsi movq %r14, %rdi callq 0x35292 movl 0x4(%rsp), %edx movq (%r14), %rsi movq %rbx, %rdi callq 0x35781 leaq 0xe0(%rsp), %rdi callq 0x9f54c movq (%r14), %rsi movq %rax, %rbx addq $0x20, %rsi jmp 0x28af81 leaq 0x68(%r12), %r14 pushq $0x1 popq %rsi movq %r14, %rdi callq 0x35292 movq (%r14), %rsi movl 0x4(%rsp), %edx movq %rbx, %rdi callq 0x35781 leaq 0xd8(%rsp), %rdi movq %r13, %rsi callq 0x96884 cmpq $0x0, 0x168(%r13) je 0x28afb9 addq $0x158, %r13 # imm = 0x158 movq %r13, %rdi movq %r12, %rsi callq 0x357ea leaq 0x30(%rsp), %rdi callq 0x357fe leaq 0xd8(%rsp), %rdi callq 0x96898 leaq 0xa0(%rsp), %rdi callq 0x35836 leaq 0x60(%rsp), %rdi callq 0x2915f4 leaq 0xb8(%rsp), %rdi callq 0x968b6 leaq 0x48(%rsp), %rdi callq 0x35858 movq %r12, %rax addq $0x198, %rsp # imm = 0x198 popq %rbx popq %r12 popq %r13 popq %r14 popq %r15 popq %rbp retq movq %rax, %rbx jmp 0x28b055 movq %rax, %rbx jmp 0x28b062 movq %rax, %rbx jmp 0x28b06f jmp 0x28b039 jmp 0x28b02a jmp 0x28b048 jmp 0x28b048 movq %r12, 0x10(%rsp) movq %rax, %rbx jmp 0x28b090 movq %r12, 0x10(%rsp) movq %rax, %rbx jmp 0x28b086 jmp 0x28b048 jmp 0x28b048 jmp 0x28b048 jmp 0x28b048 jmp 0x28b048 movq %rax, %rbx leaq 0x30(%rsp), %rdi callq 0x357fe leaq 0xd8(%rsp), %rdi callq 0x96898 leaq 0xa0(%rsp), %rdi callq 0x35836 leaq 0x60(%rsp), %rdi callq 0x2915f4 leaq 0xb8(%rsp), %rdi callq 0x968b6 leaq 0x48(%rsp), %rdi callq 0x35858 movq 0x10(%rsp), %rdi callq 0xea60 movq %rbx, %rdi callq 0xb3d0
/quesnel[P]baryonyx/lib/src/itm-optimizer-common.hpp
baryonyx::result baryonyx::itm::solve_problem<baryonyx::itm::solver_inequalities_01coeff<long double, baryonyx::itm::maximize_tag, baryonyx::itm::quadratic_cost_type<long double>, true>, long double, baryonyx::itm::maximize_tag, baryonyx::itm::quadratic_cost_type<long double>>(baryonyx::context const&, baryonyx::problem const&)
inline result solve_problem(const context& ctx, const problem& pb) { if (ctx.start) ctx.start(ctx.parameters); result ret; auto variables = length(pb.vars.values); auto constraints{ make_merged_constraints(ctx, pb) }; if (!constraints.empty() && !pb.vars.values.empty()) { random_engine rng(init_random_generator_seed(ctx)); auto cost = Cost(pb.objective, variables); auto cost_constant = pb.objective.value; switch (ctx.parameters.observer) { case solver_parameters::observer_type::pnm: { using obs = pnm_observer; solver_functor<Solver, Float, Mode, Cost, obs> slv(ctx, rng); ret = slv(constraints, variables, cost, cost_constant); } break; case solver_parameters::observer_type::file: { using obs = file_observer; solver_functor<Solver, Float, Mode, Cost, obs> slv(ctx, rng); ret = slv(constraints, variables, cost, cost_constant); } break; default: { using obs = none_observer; solver_functor<Solver, Float, Mode, Cost, obs> slv(ctx, rng); ret = slv(constraints, variables, cost, cost_constant); break; } } } else { ret.status = result_status::success; ret.solutions.resize(1); ret.solutions.back().value = pb.objective.value; } ret.strings = pb.strings; ret.variable_name = std::move(pb.vars.names); ret.affected_vars = std::move(pb.affected_vars); ret.variables = variables; ret.constraints = length(constraints); if (ctx.finish) ctx.finish(ret); return ret; }
pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x140, %rsp # imm = 0x140 cmpq $0x0, 0x128(%rsi) movq %rdx, %r15 movq %rsi, %r14 movq %rdi, %rbx je 0x28b0d4 leaq 0x118(%r14), %rdi movq %r14, %rsi callq 0x3527e vxorps %xmm0, %xmm0, %xmm0 vmovups %zmm0, 0x58(%rbx) vmovups %zmm0, 0x40(%rbx) vmovups %zmm0, (%rbx) movq $0x7fffffff, 0x98(%rbx) # imm = 0x7FFFFFFF movl $0x2, 0xa0(%rbx) movq 0xb0(%r15), %rax subq 0xa8(%r15), %rax pushq $0xc popq %rcx cqto idivq %rcx movq %rax, %r12 leaq 0x8(%rsp), %rdi movq %r14, %rsi movq %r15, %rdx vzeroupper callq 0x450078 movq 0x8(%rsp), %rax cmpq 0x10(%rsp), %rax je 0x28b1fd movq 0xa8(%r15), %rax cmpq 0xb0(%r15), %rax je 0x28b1fd movq %r14, %rdi callq 0x353ba movl $0x7fffffff, %ecx # imm = 0x7FFFFFFF xorl %edx, %edx leaq 0x10(%r15), %rsi divq %rcx cmpq $0x1, %rdx adcq $0x0, %rdx movq %rdx, (%rsp) leaq 0x20(%rsp), %rdi movl %r12d, %edx callq 0x9650c vmovsd 0x40(%r15), %xmm0 movl 0xe4(%r14), %eax vxorps %xmm1, %xmm1, %xmm1 leaq 0x40(%rsp), %rsi cmpl $0x1, %eax je 0x28b224 cmpl $0x2, %eax jne 0x28b273 movabsq $-0x10000000000000, %r9 # imm = 0xFFF0000000000000 vmovaps %xmm1, (%rsi) movq %r14, 0x10(%rsi) movq %rsp, %rax movq %rax, 0x18(%rsi) vmovaps %xmm1, 0x20(%rsi) movq %r9, 0x30(%rsi) andq $0x0, 0x48(%rsi) vmovups %xmm1, 0x38(%rsi) movl $0x7fffffff, 0x50(%rsi) # imm = 0x7FFFFFFF leaq 0x98(%rsp), %rdi leaq 0x8(%rsp), %rdx leaq 0x20(%rsp), %r8 movl %r12d, %ecx callq 0x291d64 jmp 0x28b2c0 andl $0x0, 0xa0(%rbx) leaq 0x68(%rbx), %rdi pushq $0x1 popq %rsi callq 0x35292 vmovsd 0x40(%r15), %xmm0 movq 0x70(%rbx), %rax vmovsd %xmm0, -0x8(%rax) jmp 0x28b2ef movabsq $-0x10000000000000, %r9 # imm = 0xFFF0000000000000 vmovaps %xmm1, (%rsi) movq %r14, 0x10(%rsi) movq %rsp, %rax movq %rax, 0x18(%rsi) vmovaps %xmm1, 0x20(%rsi) movq %r9, 0x30(%rsi) andq $0x0, 0x48(%rsi) vmovups %xmm1, 0x38(%rsi) movl $0x7fffffff, 0x50(%rsi) # imm = 0x7FFFFFFF leaq 0x98(%rsp), %rdi leaq 0x8(%rsp), %rdx leaq 0x20(%rsp), %r8 movl %r12d, %ecx callq 0x29161c jmp 0x28b2c0 movabsq $-0x10000000000000, %r9 # imm = 0xFFF0000000000000 vmovaps %xmm1, (%rsi) movq %r14, 0x10(%rsi) movq %rsp, %rax movq %rax, 0x18(%rsi) vmovaps %xmm1, 0x20(%rsi) movq %r9, 0x30(%rsi) andq $0x0, 0x48(%rsi) vmovups %xmm1, 0x38(%rsi) movl $0x7fffffff, 0x50(%rsi) # imm = 0x7FFFFFFF leaq 0x98(%rsp), %rdi leaq 0x8(%rsp), %rdx leaq 0x20(%rsp), %r8 movl %r12d, %ecx callq 0x2924ac leaq 0x98(%rsp), %r13 movq %rbx, %rdi movq %r13, %rsi callq 0x41d9e movq %r13, %rdi callq 0xea60 leaq 0x68(%rsp), %rdi callq 0x357fe leaq 0x20(%rsp), %rdi callq 0x968b6 leaq 0x90(%r15), %r13 movq %rbx, %rdi movq %r15, %rsi callq 0x31d28 leaq 0x50(%rbx), %rdi movq %r13, %rsi callq 0x352ec addq $0xc0, %r15 leaq 0x20(%rbx), %rdi movq %r15, %rsi callq 0x352c2 movq 0x10(%rsp), %rax movl %r12d, 0x90(%rbx) subq 0x8(%rsp), %rax pushq $0x28 popq %rcx cqto idivq %rcx movl %eax, 0x94(%rbx) cmpq $0x0, 0x168(%r14) je 0x28b35b addq $0x158, %r14 # imm = 0x158 movq %r14, %rdi movq %rbx, %rsi callq 0x357ea leaq 0x8(%rsp), %rdi callq 0x35858 movq %rbx, %rax addq $0x140, %rsp # imm = 0x140 popq %rbx popq %r12 popq %r13 popq %r14 popq %r15 retq jmp 0x28b37d jmp 0x28b37d leaq 0x68(%rsp), %rdi movq %rax, %r14 callq 0x357fe leaq 0x20(%rsp), %rdi callq 0x968b6 jmp 0x28b3a0 jmp 0x28b39d movq %rax, %r14 jmp 0x28b3aa movq %rax, %r14 leaq 0x8(%rsp), %rdi callq 0x35858 movq %rbx, %rdi callq 0xea60 movq %r14, %rdi callq 0xb3d0
/quesnel[P]baryonyx/lib/src/itm-solver-common.hpp
baryonyx::result baryonyx::itm::optimize_problem<baryonyx::itm::solver_inequalities_01coeff<long double, baryonyx::itm::maximize_tag, baryonyx::itm::quadratic_cost_type<long double>, false>, long double, baryonyx::itm::maximize_tag, baryonyx::itm::quadratic_cost_type<long double>>(baryonyx::context const&, baryonyx::problem const&)
inline result optimize_problem(const context& ctx, const problem& pb) { result r; if (ctx.start) ctx.start(ctx.parameters); auto constraints{ make_merged_constraints(ctx, pb) }; if (constraints.empty() || pb.vars.values.empty()) { r.status = result_status::success; r.solutions.resize(1); r.solutions.back().value = pb.objective.value; r.strings = pb.strings; r.affected_vars = pb.affected_vars; r.variable_name = pb.vars.names; return r; } random_engine rng(init_random_generator_seed(ctx)); auto variables = length(pb.vars.values); auto cost = Cost(pb.objective, variables); auto cost_constant = pb.objective.value; const auto thread = get_thread_number(ctx); std::vector<optimize_functor<Solver, Float, Mode, Cost>> functors; std::vector<std::thread> pool(thread); best_solution_recorder<Cost, Float, Mode> best_recorder( rng, thread, cost, cost_constant, variables, constraints, ctx.parameters.init_population_size); auto seeds = generate_seed(rng, thread); std::atomic_bool stop_task; stop_task.store(false); for (unsigned i = 0u; i != thread; ++i) functors.emplace_back(ctx, i, seeds[i]); for (unsigned i = 0u; i != thread; ++i) pool[i] = std::thread(std::ref(functors[i]), std::ref(stop_task), std::ref(best_recorder), std::cref(constraints), variables, std::cref(cost), cost_constant); const auto start = std::chrono::steady_clock::now(); auto end = start; do { std::this_thread::sleep_for(std::chrono::seconds{ 1L }); if (ctx.update) { auto call_number = 0L; for (auto i = 0u; i != thread; ++i) call_number += functors[i].m_call_number; int constraints_remaining; long int loop; double value; double duration; best_recorder.get_best( constraints_remaining, value, duration, loop); ctx.update( constraints_remaining, value, loop, duration, call_number); } end = std::chrono::steady_clock::now(); } while (!is_time_limit(ctx.parameters.time_limit, start, end)); stop_task.store(true); for (auto& t : pool) t.join(); r.strings = pb.strings; r.affected_vars = pb.affected_vars; r.variable_name = pb.vars.names; r.variables = variables; r.constraints = length(constraints); const auto& first = best_recorder.get_best(0); if (!first.is_solution()) r.status = result_status::time_limit_reached; else r.status = result_status::success; r.duration = first.duration; r.loop = first.loop; r.remaining_constraints = first.remaining_constraints; switch (ctx.parameters.storage) { case solver_parameters::storage_type::one: { r.solutions.resize(1); convert(first, r.solutions[0], variables); } break; case solver_parameters::storage_type::bound: { r.solutions.resize(2); convert(first, r.solutions[0], variables); convert(best_recorder.get_worst(), r.solutions[1], variables); } break; case solver_parameters::storage_type::five: { r.solutions.resize(5); for (int i = 0; i != 5; ++i) convert(best_recorder.get_best(i), r.solutions[i], variables); } break; } best_recorder.show_population(ctx); if (ctx.finish) ctx.finish(r); return r; }
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x198, %rsp # imm = 0x198 vxorpd %xmm0, %xmm0, %xmm0 movq %rdx, %r13 movq %rsi, %r15 movq %rdi, %r12 vmovupd %zmm0, 0x58(%rdi) vmovupd %zmm0, 0x40(%rdi) vmovupd %zmm0, (%rdi) movq $0x7fffffff, 0x98(%rdi) # imm = 0x7FFFFFFF movl $0x2, 0xa0(%rdi) cmpq $0x0, 0x128(%rsi) je 0x28b420 leaq 0x118(%r15), %rdi movq %r15, %rsi vzeroupper callq 0x3527e leaq 0x48(%rsp), %rdi movq %r15, %rsi movq %r13, %rdx vzeroupper callq 0x450078 movq 0x48(%rsp), %rax cmpq 0x50(%rsp), %rax je 0x28b57f movq 0xa8(%r13), %rax cmpq 0xb0(%r13), %rax je 0x28b57f movq %r15, %rdi callq 0x353ba movl $0x7fffffff, %ecx # imm = 0x7FFFFFFF xorl %edx, %edx divq %rcx cmpq $0x1, %rdx adcq $0x0, %rdx movq %rdx, 0x40(%rsp) movq 0xb0(%r13), %rax subq 0xa8(%r13), %rax pushq $0xc popq %rcx cqto leaq 0x10(%r13), %rsi idivq %rcx movl %eax, 0x4(%rsp) leaq 0xb8(%rsp), %rdi movl %eax, %edx movq %r12, 0x10(%rsp) callq 0x9650c vmovsd 0x40(%r13), %xmm0 vmovsd %xmm0, 0x38(%rsp) movq %r15, %rdi callq 0x35447 andq $0x0, 0x70(%rsp) vxorpd %xmm0, %xmm0, %xmm0 movl %eax, %ebp vmovapd %xmm0, 0x60(%rsp) leaq 0xa0(%rsp), %rdi leaq 0xd8(%rsp), %rdx movq %rbp, %rsi callq 0x35462 vmovsd 0x38(%rsp), %xmm0 movl 0xc4(%r15), %eax movl 0x4(%rsp), %r8d subq $0x8, %rsp movl %ebp, %edx leaq 0xe0(%rsp), %rdi leaq 0x48(%rsp), %rsi leaq 0xc0(%rsp), %rcx leaq 0x50(%rsp), %r9 pushq %rax callq 0x96720 popq %rax popq %rcx leaq 0x30(%rsp), %rdi leaq 0x40(%rsp), %rsi movl %ebp, %edx callq 0x3551c xorl %eax, %eax movb %al, 0x3(%rsp) xorl %ecx, %ecx leaq 0x60(%rsp), %rbx leaq 0x8(%rsp), %r14 xchgb %cl, 0x3(%rsp) leaq 0x90(%r13), %rcx movq %rcx, 0x88(%rsp) movl %eax, 0x8(%rsp) cmpl %ebp, %eax je 0x28b5da movl %eax, %ecx shlq $0x3, %rcx addq 0x30(%rsp), %rcx movq %rbx, %rdi movq %r15, %rsi movq %r14, %rdx callq 0x292e20 movl 0x8(%rsp), %eax incl %eax jmp 0x28b556 andl $0x0, 0xa0(%r12) leaq 0x68(%r12), %rdi pushq $0x1 popq %rsi callq 0x35292 vmovsd 0x40(%r13), %xmm0 movq 0x70(%r12), %rax vmovsd %xmm0, -0x8(%rax) movq %r12, %rdi movq %r13, %rsi callq 0x31d28 leaq 0xc0(%r13), %rsi leaq 0x20(%r12), %rdi callq 0x352c2 addq $0x90, %r13 leaq 0x50(%r12), %rdi movq %r13, %rsi callq 0x352ec jmp 0x28b975 movq %r13, 0x80(%rsp) movq %r15, %r13 leaq 0x8(%rsp), %r15 leaq 0x98(%rsp), %r14 leaq 0x4(%rsp), %rbx xorl %r12d, %r12d cmpq %rbp, %r12 je 0x28b69e imulq $0xc0, %r12, %rax addq 0x60(%rsp), %rax leaq 0x3(%rsp), %r10 leaq 0xd8(%rsp), %r11 movq %r10, 0x20(%rsp) leaq 0x48(%rsp), %r10 movq %r11, 0x18(%rsp) movq %r10, 0x98(%rsp) movq %rax, 0x28(%rsp) leaq 0xb8(%rsp), %rax movq %rax, 0x90(%rsp) leaq 0x20(%rsp), %rdx leaq 0x38(%rsp), %rax movq %r15, %rdi leaq 0x28(%rsp), %rsi leaq 0x18(%rsp), %rcx movq %r14, %r8 movq %rbx, %r9 pushq %rax leaq 0x98(%rsp), %rax pushq %rax callq 0x292e6c popq %rax popq %rcx leaq (,%r12,8), %rdi addq 0xa0(%rsp), %rdi movq %r15, %rsi callq 0x356d6 movq %r15, %rdi callq 0x356f0 incq %r12 jmp 0x28b5fa callq 0xb5a0 imulq $0xc0, %rbp, %r15 leaq 0x138(%r13), %r12 leaq 0x8(%rsp), %rbp movq %rax, %rbx movq $0x1, 0x8(%rsp) movq %rbp, %rdi callq 0x356fd cmpq $0x0, 0x148(%r13) je 0x28b733 movq 0x60(%rsp), %rax xorl %ecx, %ecx xorl %r14d, %r14d cmpq %rcx, %r15 je 0x28b6f4 addq 0xb0(%rax,%rcx), %r14 addq $0xc0, %rcx jmp 0x28b6de leaq 0xe0(%rsp), %rdi leaq 0x18(%rsp), %rsi leaq 0x28(%rsp), %rdx leaq 0x20(%rsp), %rcx movq %rbp, %r8 callq 0x9f4ce vmovsd 0x28(%rsp), %xmm0 vmovsd 0x20(%rsp), %xmm1 movq 0x8(%rsp), %rdx movl 0x18(%rsp), %esi movq %r12, %rdi movq %r14, %rcx callq 0x35736 callq 0xb5a0 vmovsd (%r13), %xmm0 subq %rbx, %rax vxorpd %xmm1, %xmm1, %xmm1 vucomisd %xmm0, %xmm1 vcvtsi2sd %rax, %xmm2, %xmm1 vdivsd 0x1ce32a(%rip), %xmm1, %xmm1 # 0x459a80 jae 0x28b6b9 vucomisd %xmm0, %xmm1 jbe 0x28b6b9 movb $0x1, %al xchgb %al, 0x3(%rsp) movq 0xa0(%rsp), %rbx movq 0xa8(%rsp), %r14 movq 0x10(%rsp), %r12 cmpq %r14, %rbx je 0x28b794 movq %rbx, %rdi callq 0xb1a0 addq $0x8, %rbx jmp 0x28b781 movq 0x80(%rsp), %rbx movq %r12, %rdi movq %rbx, %rsi callq 0x31d28 addq $0xc0, %rbx leaq 0x20(%r12), %rdi movq %rbx, %rsi callq 0x352c2 leaq 0x50(%r12), %rdi movq 0x88(%rsp), %rsi callq 0x352ec movl 0x4(%rsp), %eax movl %eax, 0x90(%r12) movq 0x50(%rsp), %rax subq 0x48(%rsp), %rax pushq $0x28 popq %rcx cqto xorl %esi, %esi idivq %rcx movq 0x118(%rsp), %rdi movq 0x130(%rsp), %rcx movl %eax, 0x94(%r12) movslq (%rdi), %rax imulq $0x38, %rax, %rax movl 0x30(%rcx,%rax), %edx leaq (%rcx,%rax), %rbx testl %edx, %edx setne %sil leal (%rsi,%rsi,2), %esi movl %esi, 0xa0(%r12) vmovsd 0x18(%rcx,%rax), %xmm0 vmovsd %xmm0, 0x80(%r12) movq 0x28(%rcx,%rax), %rax movq %rax, 0x88(%r12) movl %edx, 0x98(%r12) movl 0xe8(%r13), %eax testl %eax, %eax je 0x28b8ef cmpl $0x1, %eax je 0x28b8b7 cmpl $0x2, %eax jne 0x28b90e leaq 0x68(%r12), %rbx pushq $0x5 popq %rsi movq %rbx, %rdi callq 0x35292 xorl %r14d, %r14d xorl %r15d, %r15d cmpq $0x5, %r15 je 0x28b90e movq 0x118(%rsp), %rax movq (%rbx), %rsi movl 0x4(%rsp), %edx movslq (%rax,%r15,4), %rax addq %r14, %rsi imulq $0x38, %rax, %rdi addq 0x130(%rsp), %rdi callq 0x35781 incq %r15 addq $0x20, %r14 jmp 0x28b87d leaq 0x68(%r12), %r14 pushq $0x2 popq %rsi movq %r14, %rdi callq 0x35292 movl 0x4(%rsp), %edx movq (%r14), %rsi movq %rbx, %rdi callq 0x35781 leaq 0xe0(%rsp), %rdi callq 0x9f54c movq (%r14), %rsi movq %rax, %rbx addq $0x20, %rsi jmp 0x28b902 leaq 0x68(%r12), %r14 pushq $0x1 popq %rsi movq %r14, %rdi callq 0x35292 movq (%r14), %rsi movl 0x4(%rsp), %edx movq %rbx, %rdi callq 0x35781 leaq 0xd8(%rsp), %rdi movq %r13, %rsi callq 0x96884 cmpq $0x0, 0x168(%r13) je 0x28b93a addq $0x158, %r13 # imm = 0x158 movq %r13, %rdi movq %r12, %rsi callq 0x357ea leaq 0x30(%rsp), %rdi callq 0x357fe leaq 0xd8(%rsp), %rdi callq 0x96898 leaq 0xa0(%rsp), %rdi callq 0x35836 leaq 0x60(%rsp), %rdi callq 0x297f8e leaq 0xb8(%rsp), %rdi callq 0x968b6 leaq 0x48(%rsp), %rdi callq 0x35858 movq %r12, %rax addq $0x198, %rsp # imm = 0x198 popq %rbx popq %r12 popq %r13 popq %r14 popq %r15 popq %rbp retq movq %rax, %rbx jmp 0x28b9d6 movq %rax, %rbx jmp 0x28b9e3 movq %rax, %rbx jmp 0x28b9f0 jmp 0x28b9ba jmp 0x28b9ab jmp 0x28b9c9 jmp 0x28b9c9 movq %r12, 0x10(%rsp) movq %rax, %rbx jmp 0x28ba11 movq %r12, 0x10(%rsp) movq %rax, %rbx jmp 0x28ba07 jmp 0x28b9c9 jmp 0x28b9c9 jmp 0x28b9c9 jmp 0x28b9c9 jmp 0x28b9c9 movq %rax, %rbx leaq 0x30(%rsp), %rdi callq 0x357fe leaq 0xd8(%rsp), %rdi callq 0x96898 leaq 0xa0(%rsp), %rdi callq 0x35836 leaq 0x60(%rsp), %rdi callq 0x297f8e leaq 0xb8(%rsp), %rdi callq 0x968b6 leaq 0x48(%rsp), %rdi callq 0x35858 movq 0x10(%rsp), %rdi callq 0xea60 movq %rbx, %rdi callq 0xb3d0
/quesnel[P]baryonyx/lib/src/itm-optimizer-common.hpp
baryonyx::result baryonyx::itm::solve_problem<baryonyx::itm::solver_inequalities_01coeff<long double, baryonyx::itm::maximize_tag, baryonyx::itm::quadratic_cost_type<long double>, false>, long double, baryonyx::itm::maximize_tag, baryonyx::itm::quadratic_cost_type<long double>>(baryonyx::context const&, baryonyx::problem const&)
inline result solve_problem(const context& ctx, const problem& pb) { if (ctx.start) ctx.start(ctx.parameters); result ret; auto variables = length(pb.vars.values); auto constraints{ make_merged_constraints(ctx, pb) }; if (!constraints.empty() && !pb.vars.values.empty()) { random_engine rng(init_random_generator_seed(ctx)); auto cost = Cost(pb.objective, variables); auto cost_constant = pb.objective.value; switch (ctx.parameters.observer) { case solver_parameters::observer_type::pnm: { using obs = pnm_observer; solver_functor<Solver, Float, Mode, Cost, obs> slv(ctx, rng); ret = slv(constraints, variables, cost, cost_constant); } break; case solver_parameters::observer_type::file: { using obs = file_observer; solver_functor<Solver, Float, Mode, Cost, obs> slv(ctx, rng); ret = slv(constraints, variables, cost, cost_constant); } break; default: { using obs = none_observer; solver_functor<Solver, Float, Mode, Cost, obs> slv(ctx, rng); ret = slv(constraints, variables, cost, cost_constant); break; } } } else { ret.status = result_status::success; ret.solutions.resize(1); ret.solutions.back().value = pb.objective.value; } ret.strings = pb.strings; ret.variable_name = std::move(pb.vars.names); ret.affected_vars = std::move(pb.affected_vars); ret.variables = variables; ret.constraints = length(constraints); if (ctx.finish) ctx.finish(ret); return ret; }
pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x140, %rsp # imm = 0x140 cmpq $0x0, 0x128(%rsi) movq %rdx, %r15 movq %rsi, %r14 movq %rdi, %rbx je 0x28ba55 leaq 0x118(%r14), %rdi movq %r14, %rsi callq 0x3527e vxorps %xmm0, %xmm0, %xmm0 vmovups %zmm0, 0x58(%rbx) vmovups %zmm0, 0x40(%rbx) vmovups %zmm0, (%rbx) movq $0x7fffffff, 0x98(%rbx) # imm = 0x7FFFFFFF movl $0x2, 0xa0(%rbx) movq 0xb0(%r15), %rax subq 0xa8(%r15), %rax pushq $0xc popq %rcx cqto idivq %rcx movq %rax, %r12 leaq 0x8(%rsp), %rdi movq %r14, %rsi movq %r15, %rdx vzeroupper callq 0x450078 movq 0x8(%rsp), %rax cmpq 0x10(%rsp), %rax je 0x28bb7e movq 0xa8(%r15), %rax cmpq 0xb0(%r15), %rax je 0x28bb7e movq %r14, %rdi callq 0x353ba movl $0x7fffffff, %ecx # imm = 0x7FFFFFFF xorl %edx, %edx leaq 0x10(%r15), %rsi divq %rcx cmpq $0x1, %rdx adcq $0x0, %rdx movq %rdx, (%rsp) leaq 0x20(%rsp), %rdi movl %r12d, %edx callq 0x9650c vmovsd 0x40(%r15), %xmm0 movl 0xe4(%r14), %eax vxorps %xmm1, %xmm1, %xmm1 leaq 0x40(%rsp), %rsi cmpl $0x1, %eax je 0x28bba5 cmpl $0x2, %eax jne 0x28bbf4 movabsq $-0x10000000000000, %r9 # imm = 0xFFF0000000000000 vmovaps %xmm1, (%rsi) movq %r14, 0x10(%rsi) movq %rsp, %rax movq %rax, 0x18(%rsi) vmovaps %xmm1, 0x20(%rsi) movq %r9, 0x30(%rsi) andq $0x0, 0x48(%rsi) vmovups %xmm1, 0x38(%rsi) movl $0x7fffffff, 0x50(%rsi) # imm = 0x7FFFFFFF leaq 0x98(%rsp), %rdi leaq 0x8(%rsp), %rdx leaq 0x20(%rsp), %r8 movl %r12d, %ecx callq 0x2986fe jmp 0x28bc41 andl $0x0, 0xa0(%rbx) leaq 0x68(%rbx), %rdi pushq $0x1 popq %rsi callq 0x35292 vmovsd 0x40(%r15), %xmm0 movq 0x70(%rbx), %rax vmovsd %xmm0, -0x8(%rax) jmp 0x28bc70 movabsq $-0x10000000000000, %r9 # imm = 0xFFF0000000000000 vmovaps %xmm1, (%rsi) movq %r14, 0x10(%rsi) movq %rsp, %rax movq %rax, 0x18(%rsi) vmovaps %xmm1, 0x20(%rsi) movq %r9, 0x30(%rsi) andq $0x0, 0x48(%rsi) vmovups %xmm1, 0x38(%rsi) movl $0x7fffffff, 0x50(%rsi) # imm = 0x7FFFFFFF leaq 0x98(%rsp), %rdi leaq 0x8(%rsp), %rdx leaq 0x20(%rsp), %r8 movl %r12d, %ecx callq 0x297fb6 jmp 0x28bc41 movabsq $-0x10000000000000, %r9 # imm = 0xFFF0000000000000 vmovaps %xmm1, (%rsi) movq %r14, 0x10(%rsi) movq %rsp, %rax movq %rax, 0x18(%rsi) vmovaps %xmm1, 0x20(%rsi) movq %r9, 0x30(%rsi) andq $0x0, 0x48(%rsi) vmovups %xmm1, 0x38(%rsi) movl $0x7fffffff, 0x50(%rsi) # imm = 0x7FFFFFFF leaq 0x98(%rsp), %rdi leaq 0x8(%rsp), %rdx leaq 0x20(%rsp), %r8 movl %r12d, %ecx callq 0x298e46 leaq 0x98(%rsp), %r13 movq %rbx, %rdi movq %r13, %rsi callq 0x41d9e movq %r13, %rdi callq 0xea60 leaq 0x68(%rsp), %rdi callq 0x357fe leaq 0x20(%rsp), %rdi callq 0x968b6 leaq 0x90(%r15), %r13 movq %rbx, %rdi movq %r15, %rsi callq 0x31d28 leaq 0x50(%rbx), %rdi movq %r13, %rsi callq 0x352ec addq $0xc0, %r15 leaq 0x20(%rbx), %rdi movq %r15, %rsi callq 0x352c2 movq 0x10(%rsp), %rax movl %r12d, 0x90(%rbx) subq 0x8(%rsp), %rax pushq $0x28 popq %rcx cqto idivq %rcx movl %eax, 0x94(%rbx) cmpq $0x0, 0x168(%r14) je 0x28bcdc addq $0x158, %r14 # imm = 0x158 movq %r14, %rdi movq %rbx, %rsi callq 0x357ea leaq 0x8(%rsp), %rdi callq 0x35858 movq %rbx, %rax addq $0x140, %rsp # imm = 0x140 popq %rbx popq %r12 popq %r13 popq %r14 popq %r15 retq jmp 0x28bcfe jmp 0x28bcfe leaq 0x68(%rsp), %rdi movq %rax, %r14 callq 0x357fe leaq 0x20(%rsp), %rdi callq 0x968b6 jmp 0x28bd21 jmp 0x28bd1e movq %rax, %r14 jmp 0x28bd2b movq %rax, %r14 leaq 0x8(%rsp), %rdi callq 0x35858 movq %rbx, %rdi callq 0xea60 movq %r14, %rdi callq 0xb3d0 nop
/quesnel[P]baryonyx/lib/src/itm-solver-common.hpp
baryonyx::itm::optimize_functor<baryonyx::itm::solver_inequalities_01coeff<long double, baryonyx::itm::maximize_tag, baryonyx::itm::quadratic_cost_type<long double>, true>, long double, baryonyx::itm::maximize_tag, baryonyx::itm::quadratic_cost_type<long double>>::operator()(std::atomic<bool> const&, baryonyx::itm::best_solution_recorder<baryonyx::itm::quadratic_cost_type<long double>, long double, baryonyx::itm::maximize_tag>&, std::vector<baryonyx::itm::merged_constraint, std::allocator<baryonyx::itm::merged_constraint>> const&, int, baryonyx::itm::quadratic_cost_type<long double> const&, double)
void operator()(const std::atomic_bool& stop_task, best_solution_recorder<Cost, Float, Mode>& best_recorder, const std::vector<merged_constraint>& constraints, int variables, const Cost& original_costs, double cost_constant) { bit_array x(variables); auto& p = m_ctx.parameters; auto norm_costs = normalize_costs<Float, Cost>( m_ctx, original_costs, m_local_ctx.rng, variables); const auto kappa_step = static_cast<Float>(p.kappa_step); const auto kappa_max = static_cast<Float>(p.kappa_max); const auto alpha = static_cast<Float>(p.alpha); const auto theta = static_cast<Float>(p.theta); const auto delta = p.delta < 0 ? compute_delta<Float, Cost>(m_ctx, norm_costs, theta, variables) : static_cast<Float>(p.delta); const auto pushing_k_factor = static_cast<Float>(p.pushing_k_factor); const auto pushing_objective_amplifier = static_cast<Float>(p.pushing_objective_amplifier); const auto w_limit = static_cast<long int>(p.w); Solver slv(m_local_ctx.rng, length(constraints), variables, norm_costs, constraints); compute_order compute(p.order, variables); bool is_a_solution = false; while (!stop_task.load()) { ++m_call_number; const auto kappa_start = static_cast<Float>(best_recorder.reinit( m_local_ctx, is_a_solution, p.kappa_min, p.kappa_max, x)); auto kappa = kappa_start; compute.init(slv, x); auto best_remaining = INT_MAX; is_a_solution = false; for (long int i = 0; !stop_task.load() && i != p.limit; ++i) { auto remaining = compute.run(slv, x, m_local_ctx.rng, kappa, delta, theta); if (remaining == 0) { best_recorder.try_update( m_local_ctx, x, original_costs.results(x, cost_constant), i); best_remaining = 0; is_a_solution = true; break; } else { best_remaining = std::min(remaining, best_remaining); } if (i > w_limit) kappa += kappa_step * std::pow(static_cast<Float>(remaining) / static_cast<Float>(slv.m), alpha); if (kappa > kappa_max) break; } if (best_remaining > 0) { best_recorder.try_advance( m_local_ctx, x, best_remaining, p.limit); continue; } for (int push = 0; !stop_task.load() && push < p.pushes_limit; ++push) { auto remaining = compute.push_and_run(slv, x, m_local_ctx.rng, pushing_k_factor, delta, theta, pushing_objective_amplifier); if (remaining == 0) { best_recorder.try_update( m_local_ctx, x, original_costs.results(x, cost_constant), -push * p.pushing_iteration_limit - 1); is_a_solution = true; } kappa = kappa_start; for (int iter = 0; !stop_task.load() && iter < p.pushing_iteration_limit; ++iter) { remaining = compute.run( slv, x, m_local_ctx.rng, kappa, delta, theta); if (remaining == 0) { best_recorder.try_update( m_local_ctx, x, original_costs.results(x, cost_constant), -push * p.pushing_iteration_limit - iter - 1); is_a_solution = true; break; } if (iter > p.w) kappa += kappa_step * std::pow(static_cast<Float>(remaining) / static_cast<Float>(slv.m), alpha); if (kappa > kappa_max) break; } } } }
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x1f8, %rsp # imm = 0x1F8 movq %rdi, %rbp leaq 0x60(%rsp), %rdi movq %rsi, %r15 movl %r8d, %esi vmovsd %xmm0, 0x90(%rsp) movq %r9, %r12 movl %r8d, %ebx movq %rcx, %r14 movq %rdx, 0x70(%rsp) callq 0x368ae movq (%rbp), %r13 movq %rbp, 0xb8(%rsp) addq $0x8, %rbp leaq 0x120(%rsp), %rdi movq %r13, %rsi movq %r12, 0x88(%rsp) movq %r12, %rdx movq %rbp, %rcx movl %ebx, %r8d callq 0x98bf0 fldl 0x20(%r13) fstpt 0xd8(%rsp) vxorpd %xmm1, %xmm1, %xmm1 fldl 0x28(%r13) fstpt 0xe4(%rsp) fldl 0x30(%r13) fstpt 0xcc(%rsp) fldl 0x8(%r13) vmovsd 0x10(%r13), %xmm0 fld %st(0) fstpt 0x4c(%rsp) vucomisd %xmm0, %xmm1 jbe 0x28c21d movq 0xb8(%rsp), %rax movq (%rax), %rdi leaq 0x120(%rsp), %rsi movl %ebx, %edx fstpt (%rsp) callq 0x98cca fstpt 0x9c(%rsp) jmp 0x28c236 fstp %st(0) vmovsd %xmm0, 0x100(%rsp) fldl 0x100(%rsp) fstpt 0x9c(%rsp) fldl 0x38(%r13) fstpt 0x114(%rsp) fldl 0x40(%r13) fstpt 0x108(%rsp) vcvttsd2si 0xa0(%r13), %rax movq %rax, 0xf0(%rsp) movq 0x8(%r14), %rax subq (%r14), %rax pushq $0x28 popq %rcx cqto idivq %rcx leaq 0x178(%rsp), %rdi leaq 0x120(%rsp), %r8 movq %rbp, %rsi movl %eax, %edx movl %ebx, %ecx movq %r14, %r9 callq 0x28c702 movl 0xcc(%r13), %esi leaq 0x140(%rsp), %rdi movl %ebx, %edx callq 0x38cd8 movq $0x0, 0x58(%rsp) movb (%r15), %al testb $0x1, %al jne 0x28c66a movq 0xb8(%rsp), %rax incq 0xb0(%rax) vmovsd 0x18(%r13), %xmm0 vmovsd 0x28(%r13), %xmm1 movzbl 0x58(%rsp), %edx movq 0x70(%rsp), %rdi leaq 0x60(%rsp), %rbx movq %rbp, %rsi movq %rbx, %rcx andl $0x1, %edx callq 0x98f66 vmovsd %xmm0, 0xf8(%rsp) fldl 0xf8(%rsp) fstpt 0xc0(%rsp) leaq 0x140(%rsp), %rdi leaq 0x178(%rsp), %rsi movq %rbx, %rdx callq 0x28c8ca fldt 0xc0(%rsp) fldt 0x4c(%rsp) movl $0x7fffffff, %ebx # imm = 0x7FFFFFFF xorl %r14d, %r14d movb (%r15), %al testb $0x1, %al jne 0x28c3ff cmpq 0xb0(%r13), %r14 je 0x28c3ff fstpt 0x20(%rsp) leaq 0x140(%rsp), %rdi leaq 0x178(%rsp), %rsi leaq 0x60(%rsp), %rdx movq %rbp, %rcx fldt 0x9c(%rsp) fstpt 0x10(%rsp) fld %st(0) fstpt 0x7c(%rsp) fstpt (%rsp) callq 0x28c90c testl %eax, %eax je 0x28c419 cmpl %eax, %ebx cmovgel %eax, %ebx cmpq 0xf0(%rsp), %r14 jle 0x28c3de movl %eax, 0xb0(%rsp) fildl 0xb0(%rsp) movl 0x1f0(%rsp), %eax movl %eax, 0xb4(%rsp) fidivl 0xb4(%rsp) fldt 0xcc(%rsp) fstpt 0x10(%rsp) fstpt (%rsp) callq 0xb110 fldt 0xd8(%rsp) fmulp %st, %st(1) fldt 0x7c(%rsp) faddp %st, %st(1) fldt 0x4c(%rsp) jmp 0x28c3e8 fldt 0x4c(%rsp) fldt 0x7c(%rsp) fxch %st(1) fldt 0xe4(%rsp) fxch %st(2) fucomi %st(2), %st fstp %st(2) ja 0x28c3ff incq %r14 jmp 0x28c32f fstp %st(1) testl %ebx, %ebx jg 0x28c648 movq $0x0, 0x58(%rsp) leaq 0x60(%rsp), %rbx jmp 0x28c455 vmovsd 0x90(%rsp), %xmm0 movq 0x88(%rsp), %rdi leaq 0x60(%rsp), %rbx movq %rbx, %rsi callq 0x9706e movq 0x70(%rsp), %rdi movq %rbp, %rsi movq %rbx, %rdx movq %r14, %rcx callq 0x992d8 movb $0x1, %al movq %rax, 0x58(%rsp) fldt 0x4c(%rsp) xorl %r14d, %r14d movl $0x0, 0x78(%rsp) movb (%r15), %al testb $0x1, %al jne 0x28c641 cmpl 0xbc(%r13), %r14d jge 0x28c641 fldt 0x108(%rsp) fstpt 0x30(%rsp) fstpt 0x20(%rsp) leaq 0x140(%rsp), %rdi leaq 0x178(%rsp), %rsi movq %rbx, %rdx movq %rbp, %rcx fldt 0x9c(%rsp) fstpt 0x10(%rsp) fldt 0x114(%rsp) fstpt (%rsp) callq 0x28cb3c testl %eax, %eax jne 0x28c4fb vmovsd 0x90(%rsp), %xmm0 movq 0x88(%rsp), %rdi movq %rbx, %rsi callq 0x9706e movl 0xc0(%r13), %eax movq 0x70(%rsp), %rdi movq %rbp, %rsi movq %rbx, %rdx imull %r14d, %eax notl %eax movslq %eax, %rcx callq 0x992d8 movb $0x1, %al movq %rax, 0x58(%rsp) pushq $0x1 popq %rax fldt 0xc0(%rsp) fldt 0x4c(%rsp) movl %eax, %ebx movb (%r15), %al testb $0x1, %al jne 0x28c5df leal -0x1(%rbx), %r12d cmpl 0xc0(%r13), %r12d jge 0x28c5df fstpt 0x20(%rsp) leaq 0x140(%rsp), %rdi leaq 0x178(%rsp), %rsi leaq 0x60(%rsp), %rdx movq %rbp, %rcx fldt 0x9c(%rsp) fstpt 0x10(%rsp) fld %st(0) fstpt 0x7c(%rsp) fstpt (%rsp) callq 0x28c90c testl %eax, %eax je 0x28c5e8 vcvtsi2sd %r12d, %xmm2, %xmm0 vucomisd 0xa0(%r13), %xmm0 jbe 0x28c5bf movl %eax, 0xa8(%rsp) fildl 0xa8(%rsp) movl 0x1f0(%rsp), %eax movl %eax, 0xac(%rsp) fidivl 0xac(%rsp) fldt 0xcc(%rsp) fstpt 0x10(%rsp) fstpt (%rsp) callq 0xb110 fldt 0xd8(%rsp) fmulp %st, %st(1) fldt 0x7c(%rsp) faddp %st, %st(1) fldt 0x4c(%rsp) jmp 0x28c5c9 fldt 0x4c(%rsp) fldt 0x7c(%rsp) fxch %st(1) fldt 0xe4(%rsp) fxch %st(2) fucomi %st(2), %st fstp %st(2) ja 0x28c5df incl %ebx jmp 0x28c50b fstp %st(1) leaq 0x60(%rsp), %rbx jmp 0x28c635 vmovsd 0x90(%rsp), %xmm0 movq 0x88(%rsp), %rdi leaq 0x60(%rsp), %r12 movq %r12, %rsi callq 0x9706e movl 0xc0(%r13), %eax movq 0x70(%rsp), %rdi movq %rbp, %rsi movq %r12, %rdx imull 0x78(%rsp), %eax subl %ebx, %eax movq %r12, %rbx movslq %eax, %rcx callq 0x992d8 movb $0x1, %al movq %rax, 0x58(%rsp) fldt 0x4c(%rsp) decl 0x78(%rsp) incl %r14d jmp 0x28c460 fstp %st(0) jmp 0x28c2aa fstp %st(0) movq 0xb0(%r13), %r8 movq 0x70(%rsp), %rdi leaq 0x60(%rsp), %rdx movq %rbp, %rsi movl %ebx, %ecx callq 0x99360 jmp 0x28c2a1 leaq 0x140(%rsp), %rdi callq 0x393ac leaq 0x178(%rsp), %rdi callq 0x28cde0 leaq 0x120(%rsp), %rdi callq 0x968b6 leaq 0x68(%rsp), %rdi callq 0x357fe addq $0x1f8, %rsp # imm = 0x1F8 popq %rbx popq %r12 popq %r13 popq %r14 popq %r15 popq %rbp retq jmp 0x28c6b4 movq %rax, %rbx jmp 0x28c6d6 movq %rax, %rbx jmp 0x28c6e3 movq %rax, %rbx jmp 0x28c6f0 jmp 0x28c6c6 jmp 0x28c6c6 jmp 0x28c6c6 jmp 0x28c6c6 leaq 0x140(%rsp), %rdi movq %rax, %rbx callq 0x393ac leaq 0x178(%rsp), %rdi callq 0x28cde0 leaq 0x120(%rsp), %rdi callq 0x968b6 leaq 0x68(%rsp), %rdi callq 0x357fe movq %rbx, %rdi callq 0xb3d0
/quesnel[P]baryonyx/lib/src/itm-optimizer-common.hpp
int baryonyx::itm::compute_order::push_and_run<baryonyx::itm::solver_inequalities_01coeff<long double, baryonyx::itm::maximize_tag, baryonyx::itm::quadratic_cost_type<long double>, true>, baryonyx::bit_array, long double>(baryonyx::itm::solver_inequalities_01coeff<long double, baryonyx::itm::maximize_tag, baryonyx::itm::quadratic_cost_type<long double>, true>&, baryonyx::bit_array&, std::linear_congruential_engine<unsigned long, 16807ul, 0ul, 2147483647ul>&, long double, long double, long double, long double)
int push_and_run(Solver& solver, Xtype& x, random_engine& rng, Float kappa, Float delta, Float theta, Float objective_amplifier) { bool pi_changed = 0; int remaining = 0; if (use_cycle) order = next_state(order); switch (order) { case solver_parameters::constraint_order::reversing: solver.push_and_compute_update_row(x, R.crbegin(), R.crend(), kappa, delta, theta, objective_amplifier); return compute_violated_constraints(solver, x, R); case solver_parameters::constraint_order::random_sorting: std::shuffle(R.begin(), R.end(), rng); solver.push_and_compute_update_row( x, R.begin(), R.end(), kappa, delta, theta, objective_amplifier); return compute_violated_constraints(solver, x, R); case solver_parameters::constraint_order::infeasibility_decr: std::sort(m_order.begin(), m_order.end(), [](const auto& lhs, const auto& rhs) { return rhs.second < lhs.second; }); solver.push_and_compute_update_row(x, m_order.cbegin(), m_order.cend(), kappa, delta, theta, objective_amplifier); return infeasibility_local_compute_violated_constraints(solver, x); case solver_parameters::constraint_order::infeasibility_incr: std::sort(m_order.begin(), m_order.end(), [](const auto& lhs, const auto& rhs) { return lhs.second < rhs.second; }); solver.push_and_compute_update_row(x, m_order.cbegin(), m_order.cend(), kappa, delta, theta, objective_amplifier); return infeasibility_local_compute_violated_constraints(solver, x); case solver_parameters::constraint_order::lagrangian_decr: std::sort( R.begin(), R.end(), [&solver](const auto lhs, const auto rhs) { return solver.pi[rhs] < solver.pi[lhs]; }); solver.push_and_compute_update_row(x, R.cbegin(), R.cend(), kappa, delta, theta, objective_amplifier); return compute_violated_constraints(solver, x, R); case solver_parameters::constraint_order::lagrangian_incr: std::sort( R.begin(), R.end(), [&solver](const auto lhs, const auto rhs) { return solver.pi[lhs] < solver.pi[rhs]; }); solver.push_and_compute_update_row(x, R.cbegin(), R.cend(), kappa, delta, theta, objective_amplifier); return compute_violated_constraints(solver, x, R); case solver_parameters::constraint_order::pi_sign_change: std::shuffle(R.begin(), R.end(), rng); pi_changed = solver.push_and_compute_update_row( x, R.begin(), R.end(), kappa, delta, theta, objective_amplifier); remaining = local_compute_violated_constraints(solver, x); if (!pi_changed && remaining == 0) return 0; return remaining; case solver_parameters::constraint_order::none: default: solver.push_and_compute_update_row(x, R.cbegin(), R.cend(), kappa, delta, theta, objective_amplifier); return compute_violated_constraints(solver, x, R); } }
pushq %r15 pushq %r14 pushq %rbx subq $0x80, %rsp fldt 0xd0(%rsp) fldt 0xc0(%rsp) fldt 0xb0(%rsp) fldt 0xa0(%rsp) movl 0x30(%rdi), %eax cmpb $0x1, 0x34(%rdi) movq %rdx, %rbx movq %rsi, %r15 movq %rdi, %r14 jne 0x28cb87 xorl %esi, %esi leal 0x1(%rax), %edx cmpl $0x7, %eax cmovll %edx, %esi movl %esi, %eax movl %esi, 0x30(%r14) decl %eax cmpl $0x6, %eax ja 0x28cc9e leaq 0x1d42b3(%rip), %rdx # 0x460e4c movslq (%rdx,%rax,4), %rax addq %rdx, %rax jmpq *%rax movq 0x8(%r14), %rax leaq 0x78(%rsp), %rdx leaq 0x70(%rsp), %rcx movq %r15, %rdi movq %rbx, %rsi movq %rax, (%rdx) movq (%r14), %rax movq %rax, (%rcx) fxch %st(3) fstpt 0x30(%rsp) fxch %st(1) fstpt 0x20(%rsp) fstpt 0x10(%rsp) fstpt (%rsp) callq 0x28f894 movq %r15, %rdi movq %rbx, %rsi movq %r14, %rdx callq 0x28cfd1 addq $0x80, %rsp popq %rbx popq %r14 popq %r15 retq movq (%r14), %rdi movq 0x8(%r14), %rsi fstpt 0x64(%rsp) fstpt 0x58(%rsp) fstpt 0x4c(%rsp) movq %r15, %rdx fstpt 0x40(%rsp) callq 0x290b4f jmp 0x28cd2b movq 0x18(%r14), %rdi movq 0x20(%r14), %rsi fstpt 0x64(%rsp) fstpt 0x58(%rsp) fstpt 0x4c(%rsp) fstpt 0x40(%rsp) callq 0x290279 jmp 0x28cc52 movq 0x18(%r14), %rdi movq 0x20(%r14), %rsi fstpt 0x64(%rsp) fstpt 0x58(%rsp) fstpt 0x4c(%rsp) fstpt 0x40(%rsp) callq 0x2906e4 movq 0x18(%r14), %rdx movq 0x20(%r14), %rcx fldt 0x40(%rsp) fstpt 0x30(%rsp) movq %r15, %rdi movq %rbx, %rsi fldt 0x4c(%rsp) fstpt 0x20(%rsp) fldt 0x58(%rsp) fstpt 0x10(%rsp) fldt 0x64(%rsp) fstpt (%rsp) callq 0x28fd8e movq %r14, %rdi movq %r15, %rsi movq %rbx, %rdx addq $0x80, %rsp popq %rbx popq %r14 popq %r15 jmp 0x28d6be movq (%r14), %rdx movq 0x8(%r14), %rcx fxch %st(3) fstpt 0x30(%rsp) fxch %st(1) fstpt 0x20(%rsp) fstpt 0x10(%rsp) jmp 0x28cd4e movq (%r14), %rdi movq 0x8(%r14), %rsi fstpt 0x64(%rsp) fstpt 0x58(%rsp) fstpt 0x4c(%rsp) movq %rcx, %rdx fstpt 0x40(%rsp) callq 0x3ceed movq (%r14), %rdx movq 0x8(%r14), %rcx fldt 0x40(%rsp) fstpt 0x30(%rsp) movq %r15, %rdi movq %rbx, %rsi fldt 0x4c(%rsp) fstpt 0x20(%rsp) fldt 0x58(%rsp) fstpt 0x10(%rsp) fldt 0x64(%rsp) fstpt (%rsp) callq 0x28fb18 jmp 0x28cd5c movq (%r14), %rdi movq 0x8(%r14), %rsi fstpt 0x64(%rsp) fstpt 0x58(%rsp) fstpt 0x4c(%rsp) movq %r15, %rdx fstpt 0x40(%rsp) callq 0x2910a3 movq (%r14), %rdx movq 0x8(%r14), %rcx fldt 0x40(%rsp) fstpt 0x30(%rsp) fldt 0x4c(%rsp) fstpt 0x20(%rsp) fldt 0x58(%rsp) fstpt 0x10(%rsp) fldt 0x64(%rsp) fstpt (%rsp) movq %r15, %rdi movq %rbx, %rsi callq 0x290004 movq %r15, %rdi movq %rbx, %rsi movq %r14, %rdx addq $0x80, %rsp popq %rbx popq %r14 popq %r15 jmp 0x28cfd1 movq (%r14), %rdi movq 0x8(%r14), %rsi fstpt 0x64(%rsp) fstpt 0x58(%rsp) fstpt 0x4c(%rsp) movq %rcx, %rdx fstpt 0x40(%rsp) callq 0x3ceed movq (%r14), %rdx movq 0x8(%r14), %rcx fldt 0x40(%rsp) fstpt 0x30(%rsp) movq %r15, %rdi movq %rbx, %rsi fldt 0x4c(%rsp) fstpt 0x20(%rsp) fldt 0x58(%rsp) fstpt 0x10(%rsp) fldt 0x64(%rsp) fstpt (%rsp) callq 0x28fb18 movq %r14, %rdi movq %r15, %rsi movq %rbx, %rdx addq $0x80, %rsp popq %rbx popq %r14 popq %r15 jmp 0x28d99e
/quesnel[P]baryonyx/lib/src/itm-common.hpp
int baryonyx::itm::compute_order::infeasibility_local_compute_violated_constraints<baryonyx::itm::solver_inequalities_01coeff<long double, baryonyx::itm::maximize_tag, baryonyx::itm::quadratic_cost_type<long double>, true> const, baryonyx::bit_array>(baryonyx::itm::solver_inequalities_01coeff<long double, baryonyx::itm::maximize_tag, baryonyx::itm::quadratic_cost_type<long double>, true> const&, baryonyx::bit_array const&)
int infeasibility_local_compute_violated_constraints(Solver& solver, const Xtype& x) { m_order.clear(); for (int k = 0, e = solver.m; k != e; ++k) { sparse_matrix<int>::const_row_iterator it, et; std::tie(it, et) = solver.ap.row(k); int v = 0; for (; it != et; ++it) v += solver.factor(it->value) * x[it->column]; if (solver.bound_min(k) > v) m_order.push_back(std::make_pair(k, solver.bound_min(k) - v)); else if (solver.bound_max(k) < v) m_order.push_back(std::make_pair(k, v - solver.bound_max(k))); } return length(m_order); }
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x38, %rsp movq 0x18(%rdi), %rax movq %rdx, %r14 movq %rsi, 0x20(%rsp) cmpq %rax, 0x20(%rdi) je 0x28cf10 movq %rax, 0x20(%rdi) movq 0x20(%rsp), %rcx leaq 0x18(%rdi), %rax movq %rdi, 0x28(%rsp) xorl %ebp, %ebp movq %rax, 0x18(%rsp) movl 0x78(%rcx), %r12d leaq 0x10(%rcx), %rax movq %rax, 0x30(%rsp) cmpq %r12, %rbp je 0x28cfb1 movq 0x30(%rsp), %rsi leaq 0x8(%rsp), %rdi movl %ebp, %edx callq 0x3c868 movq 0x8(%rsp), %r15 movq 0x10(%rsp), %r13 xorl %ebx, %ebx cmpq %r15, %r13 je 0x28cf6d movl 0x4(%r13), %esi movq %r14, %rdi callq 0x3664a addl %eax, %ebx addq $0x8, %r13 jmp 0x28cf54 movq 0x20(%rsp), %rax movq 0x60(%rax), %rcx movl (%rcx,%rbp,8), %eax subl %ebx, %eax jle 0x28cf8b shlq $0x20, %rax orq %rbp, %rax movq %rax, 0x8(%rsp) jmp 0x28cf9d subl 0x4(%rcx,%rbp,8), %ebx jle 0x28cfac shlq $0x20, %rbx orq %rbp, %rbx movq %rbx, 0x8(%rsp) movq 0x18(%rsp), %rdi leaq 0x8(%rsp), %rsi callq 0x3c8c2 incq %rbp jmp 0x28cf32 movq 0x28(%rsp), %rcx movq 0x20(%rcx), %rax subq 0x18(%rcx), %rax shrq $0x3, %rax addq $0x38, %rsp popq %rbx popq %r12 popq %r13 popq %r14 popq %r15 popq %rbp retq
/quesnel[P]baryonyx/lib/src/itm-common.hpp
int baryonyx::itm::compute_violated_constraints<baryonyx::itm::solver_inequalities_01coeff<long double, baryonyx::itm::maximize_tag, baryonyx::itm::quadratic_cost_type<long double>, true>, baryonyx::bit_array>(baryonyx::itm::solver_inequalities_01coeff<long double, baryonyx::itm::maximize_tag, baryonyx::itm::quadratic_cost_type<long double>, true> const&, baryonyx::bit_array const&, std::vector<int, std::allocator<int>>&)
int compute_violated_constraints(const Solver& slv, const Xtype& x, std::vector<int>& out) { out.clear(); for (int k = 0; k != slv.m; ++k) if (!is_valid_constraint(slv, k, x)) out.emplace_back(k); return length(out); }
pushq %r15 pushq %r14 pushq %r12 pushq %rbx pushq %rax movq (%rdx), %rax movq %rdx, %rbx movq %rsi, %r14 movq %rdi, %r15 cmpq %rax, 0x8(%rdx) je 0x28cfef movq %rax, 0x8(%rbx) leaq 0x4(%rsp), %r12 xorl %esi, %esi movl %esi, 0x4(%rsp) cmpl 0x78(%r15), %esi je 0x28d022 movq %r15, %rdi movq %r14, %rdx callq 0x28d039 testb %al, %al jne 0x28d01a movq %rbx, %rdi movq %r12, %rsi callq 0x31460 movl 0x4(%rsp), %esi incl %esi jmp 0x28cff6 movq 0x8(%rbx), %rax subq (%rbx), %rax shrq $0x2, %rax addq $0x8, %rsp popq %rbx popq %r12 popq %r14 popq %r15 retq
/quesnel[P]baryonyx/lib/src/itm-common.hpp
bool baryonyx::itm::is_valid_constraint<baryonyx::itm::solver_inequalities_01coeff<long double, baryonyx::itm::maximize_tag, baryonyx::itm::quadratic_cost_type<long double>, true>, baryonyx::bit_array>(baryonyx::itm::solver_inequalities_01coeff<long double, baryonyx::itm::maximize_tag, baryonyx::itm::quadratic_cost_type<long double>, true> const&, int, baryonyx::bit_array const&)
bool is_valid_constraint(const Solver& slv, int k, const Xtype& x) { typename sparse_matrix<int>::const_row_iterator it, et; std::tie(it, et) = slv.ap.row(k); int v = 0; for (; it != et; ++it) v += slv.factor(it->value) * x[it->column]; return slv.bound_min(k) <= v && v <= slv.bound_max(k); }
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x18, %rsp movl %esi, %ebp leaq 0x10(%rdi), %rsi leaq 0x8(%rsp), %r15 movq %rdx, %r14 movq %rdi, %rbx movl %ebp, %edx movq %r15, %rdi callq 0x3c868 movq (%r15), %r12 movq 0x8(%r15), %r13 xorl %r15d, %r15d cmpq %r12, %r13 je 0x28d086 movl 0x4(%r13), %esi movq %r14, %rdi callq 0x3664a addl %eax, %r15d addq $0x8, %r13 jmp 0x28d06c movq 0x60(%rbx), %rcx movslq %ebp, %rax cmpl %r15d, (%rcx,%rax,8) jle 0x28d097 xorl %eax, %eax jmp 0x28d09f cmpl 0x4(%rcx,%rax,8), %r15d setle %al addq $0x18, %rsp popq %rbx popq %r12 popq %r13 popq %r14 popq %r15 popq %rbp retq
/quesnel[P]baryonyx/lib/src/itm-common.hpp
bool baryonyx::itm::solver_inequalities_01coeff<long double, baryonyx::itm::maximize_tag, baryonyx::itm::quadratic_cost_type<long double>, true>::compute_update_row<baryonyx::bit_array, std::reverse_iterator<__gnu_cxx::__normal_iterator<int const*, std::vector<int, std::allocator<int>>>>>(baryonyx::bit_array&, std::reverse_iterator<__gnu_cxx::__normal_iterator<int const*, std::vector<int, std::allocator<int>>>>, std::reverse_iterator<__gnu_cxx::__normal_iterator<int const*, std::vector<int, std::allocator<int>>>>, long double, long double, long double)
bool compute_update_row(Xtype& x, Iterator first, Iterator last, Float kappa, Float delta, Float theta) { auto at_least_one_pi_changed{ false }; logger::log("update-row {} {} {}\n", kappa, delta, theta); for (; first != last; ++first) { auto k = constraint(first); const auto it = ap.row(k); decrease_preference(std::get<0>(it), std::get<1>(it), theta); const auto r_size = compute_reduced_costs(std::get<0>(it), std::get<1>(it), x); calculator_sort<Mode>(R.get(), R.get() + r_size, rng); int selected = select_variables(r_size, b[k].min, b[k].max); logger::log("constraints {}: {} <= ", k, b[k].min); for (int i = 0; i < r_size; ++i) logger::log("({} {}) ", R[i].value, R[i].id); logger::log("<= {} => Selected: {}\n", b[k].max, selected); auto pi_change = affect( *this, x, std::get<0>(it), k, selected, r_size, kappa, delta); at_least_one_pi_changed = at_least_one_pi_changed || pi_change; } return at_least_one_pi_changed; }
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0xa8, %rsp movq %rcx, 0x60(%rsp) movq %rdx, %rbx leaq 0x1cf276(%rip), %rdx # 0x45c344 movq %rsi, %r14 movq %rdi, %r12 fldt 0x100(%rsp) fldt 0xf0(%rsp) fldt 0xe0(%rsp) fstpt 0x90(%rsp) fstpt 0x80(%rsp) fstpt 0x70(%rsp) pushq $0x14 popq %rsi leaq 0x90(%rsp), %rcx leaq 0x80(%rsp), %r8 leaq 0x70(%rsp), %r9 callq 0x126c32 leaq 0x10(%r12), %rax movq %rbx, 0x58(%rsp) xorl %edx, %edx movq %rax, 0x48(%rsp) movq (%rbx), %rax pushq $0x16 popq %r13 leaq 0x1d6607(%rip), %rbp # 0x46373e pushq $0x8 popq %rbx movq %r14, 0x50(%rsp) movq 0x60(%rsp), %rcx cmpq (%rcx), %rax je 0x28d2b2 movq %rdx, 0x68(%rsp) movl -0x4(%rax), %edx movq 0x48(%rsp), %rsi leaq 0x38(%rsp), %rdi movl %edx, 0x2c(%rsp) callq 0x3de66 movq 0x38(%rsp), %rdx movq 0x40(%rsp), %rsi fldt 0x70(%rsp) movq %r12, %rdi fstpt (%rsp) callq 0x28d9de movq 0x38(%rsp), %rdx movq 0x40(%rsp), %rsi movq %r12, %rdi movq %r14, %rcx callq 0x28da04 movq 0x58(%r12), %rdi movq 0x8(%r12), %rdx movslq %eax, %r14 movl %eax, 0x34(%rsp) movq %r14, %rsi shlq $0x5, %rsi addq %rdi, %rsi callq 0x28dada movslq 0x2c(%rsp), %rax movq 0x60(%r12), %rcx movq %r12, %rdi movl %r14d, %esi movl (%rcx,%rax,8), %edx movl 0x4(%rcx,%rax,8), %ecx callq 0x28db0e movslq 0x2c(%rsp), %r8 movl %eax, 0x30(%rsp) leaq 0x1d3e63(%rip), %rdx # 0x461045 leaq 0x2c(%rsp), %rcx movq %r12, %rdi movq %r13, %r15 movq %r13, %rsi shlq $0x3, %r8 addq 0x60(%r12), %r8 callq 0xb91c2 movl %r14d, %eax sarl $0x1f, %eax xorl %r13d, %r13d andnl %r14d, %eax, %r14d shlq $0x5, %r14 cmpq %r13, %r14 je 0x28d236 movq 0x58(%r12), %rcx movq %r12, %rdi movq %rbx, %rsi movq %rbp, %rdx leaq 0x10(%rcx,%r13), %r8 addq %r13, %rcx callq 0x126d84 addq $0x20, %r13 jmp 0x28d210 movslq 0x2c(%rsp), %rax movq 0x60(%r12), %rcx leaq 0x1d3e15(%rip), %rdx # 0x46105c leaq 0x30(%rsp), %r8 movq %r12, %rdi movq %r15, %r13 movq %r15, %rsi leaq 0x4(%rcx,%rax,8), %rcx callq 0xb91c2 movq 0x40(%rsp), %rdx movl 0x2c(%rsp), %ecx movl 0x30(%rsp), %r8d fldt 0x90(%rsp) fldt 0x80(%rsp) fstpt 0x10(%rsp) fstpt (%rsp) movq %r12, %rdi movq 0x50(%rsp), %r14 movl 0x34(%rsp), %r9d movq %r14, %rsi callq 0x28db7f movq 0x68(%rsp), %rdx movq 0x58(%rsp), %rcx orb %al, %dl movq (%rcx), %rax addq $-0x4, %rax movq %rax, (%rcx) jmp 0x28d13f andb $0x1, %dl movl %edx, %eax addq $0xa8, %rsp popq %rbx popq %r12 popq %r13 popq %r14 popq %r15 popq %rbp retq nop
/quesnel[P]baryonyx/lib/src/itm-solver-inequalities-01.cpp
bool baryonyx::itm::solver_inequalities_01coeff<long double, baryonyx::itm::maximize_tag, baryonyx::itm::quadratic_cost_type<long double>, true>::compute_update_row<baryonyx::bit_array, __gnu_cxx::__normal_iterator<int*, std::vector<int, std::allocator<int>>>>(baryonyx::bit_array&, __gnu_cxx::__normal_iterator<int*, std::vector<int, std::allocator<int>>>, __gnu_cxx::__normal_iterator<int*, std::vector<int, std::allocator<int>>>, long double, long double, long double)
bool compute_update_row(Xtype& x, Iterator first, Iterator last, Float kappa, Float delta, Float theta) { auto at_least_one_pi_changed{ false }; logger::log("update-row {} {} {}\n", kappa, delta, theta); for (; first != last; ++first) { auto k = constraint(first); const auto it = ap.row(k); decrease_preference(std::get<0>(it), std::get<1>(it), theta); const auto r_size = compute_reduced_costs(std::get<0>(it), std::get<1>(it), x); calculator_sort<Mode>(R.get(), R.get() + r_size, rng); int selected = select_variables(r_size, b[k].min, b[k].max); logger::log("constraints {}: {} <= ", k, b[k].min); for (int i = 0; i < r_size; ++i) logger::log("({} {}) ", R[i].value, R[i].id); logger::log("<= {} => Selected: {}\n", b[k].max, selected); auto pi_change = affect( *this, x, std::get<0>(it), k, selected, r_size, kappa, delta); at_least_one_pi_changed = at_least_one_pi_changed || pi_change; } return at_least_one_pi_changed; }
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x98, %rsp movq %rcx, 0x48(%rsp) movq %rdx, %r14 movq %rsi, %r13 movq %rdi, %r12 leaq 0x1cf054(%rip), %rdx # 0x45c344 fldt 0xf0(%rsp) fldt 0xe0(%rsp) fldt 0xd0(%rsp) fstpt 0x80(%rsp) fstpt 0x70(%rsp) fstpt 0x60(%rsp) pushq $0x14 popq %rsi leaq 0x80(%rsp), %rcx leaq 0x70(%rsp), %r8 leaq 0x60(%rsp), %r9 callq 0x126c32 leaq 0x10(%r12), %rax leaq 0x1d6404(%rip), %rbp # 0x46373e movq %rax, 0x38(%rsp) xorl %eax, %eax pushq $0x8 popq %rbx movq %r13, 0x40(%rsp) cmpq 0x48(%rsp), %r14 je 0x28d4af movl (%r14), %edx movq 0x38(%rsp), %rsi leaq 0x28(%rsp), %rdi movq %rax, 0x50(%rsp) movq %r14, 0x58(%rsp) movl %edx, 0x20(%rsp) callq 0x3de66 movq 0x28(%rsp), %rdx movq 0x30(%rsp), %rsi fldt 0x60(%rsp) movq %r12, %rdi fstpt (%rsp) callq 0x28d9de movq 0x28(%rsp), %rdx movq 0x30(%rsp), %rsi movq %r12, %rdi movq %r13, %rcx callq 0x28da04 movq 0x58(%r12), %rdi movq 0x8(%r12), %rdx movslq %eax, %r13 movl %eax, %r15d movq %r13, %rsi shlq $0x5, %rsi addq %rdi, %rsi callq 0x28dada movslq 0x20(%rsp), %rax movq 0x60(%r12), %rcx movq %r12, %rdi movl %r13d, %esi movl (%rcx,%rax,8), %edx movl 0x4(%rcx,%rax,8), %ecx callq 0x28db0e movslq 0x20(%rsp), %r8 movl %eax, 0x24(%rsp) movq %r12, %rdi shlq $0x3, %r8 addq 0x60(%r12), %r8 pushq $0x16 popq %rsi leaq 0x1d3c49(%rip), %rdx # 0x461045 leaq 0x20(%rsp), %rcx callq 0xb91c2 movl %r13d, %eax sarl $0x1f, %eax xorl %r14d, %r14d andnl %r13d, %eax, %r13d shlq $0x5, %r13 cmpq %r14, %r13 je 0x28d43e movq 0x58(%r12), %rcx movq %r12, %rdi movq %rbx, %rsi movq %rbp, %rdx leaq 0x10(%rcx,%r14), %r8 addq %r14, %rcx callq 0x126d84 addq $0x20, %r14 jmp 0x28d418 movslq 0x20(%rsp), %rax movq 0x60(%r12), %rcx movq %r12, %rdi leaq 0x4(%rcx,%rax,8), %rcx pushq $0x16 popq %rsi leaq 0x1d3c02(%rip), %rdx # 0x46105c leaq 0x24(%rsp), %r8 callq 0xb91c2 movq 0x30(%rsp), %rdx movl 0x20(%rsp), %ecx movl 0x24(%rsp), %r8d fldt 0x80(%rsp) fldt 0x70(%rsp) fstpt 0x10(%rsp) fstpt (%rsp) movq %r12, %rdi movl %r15d, %r9d movq 0x40(%rsp), %r13 movq %r13, %rsi callq 0x28db7f movq 0x50(%rsp), %rcx movq 0x58(%rsp), %r14 orb %al, %cl addq $0x4, %r14 movq %rcx, %rax jmp 0x28d349 andb $0x1, %al addq $0x98, %rsp popq %rbx popq %r12 popq %r13 popq %r14 popq %r15 popq %rbp retq nop
/quesnel[P]baryonyx/lib/src/itm-solver-inequalities-01.cpp
bool baryonyx::itm::solver_inequalities_01coeff<long double, baryonyx::itm::maximize_tag, baryonyx::itm::quadratic_cost_type<long double>, true>::compute_update_row<baryonyx::bit_array, __gnu_cxx::__normal_iterator<std::pair<int, int> const*, std::vector<std::pair<int, int>, std::allocator<std::pair<int, int>>>>>(baryonyx::bit_array&, __gnu_cxx::__normal_iterator<std::pair<int, int> const*, std::vector<std::pair<int, int>, std::allocator<std::pair<int, int>>>>, __gnu_cxx::__normal_iterator<std::pair<int, int> const*, std::vector<std::pair<int, int>, std::allocator<std::pair<int, int>>>>, long double, long double, long double)
bool compute_update_row(Xtype& x, Iterator first, Iterator last, Float kappa, Float delta, Float theta) { auto at_least_one_pi_changed{ false }; logger::log("update-row {} {} {}\n", kappa, delta, theta); for (; first != last; ++first) { auto k = constraint(first); const auto it = ap.row(k); decrease_preference(std::get<0>(it), std::get<1>(it), theta); const auto r_size = compute_reduced_costs(std::get<0>(it), std::get<1>(it), x); calculator_sort<Mode>(R.get(), R.get() + r_size, rng); int selected = select_variables(r_size, b[k].min, b[k].max); logger::log("constraints {}: {} <= ", k, b[k].min); for (int i = 0; i < r_size; ++i) logger::log("({} {}) ", R[i].value, R[i].id); logger::log("<= {} => Selected: {}\n", b[k].max, selected); auto pi_change = affect( *this, x, std::get<0>(it), k, selected, r_size, kappa, delta); at_least_one_pi_changed = at_least_one_pi_changed || pi_change; } return at_least_one_pi_changed; }
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x98, %rsp movq %rcx, 0x48(%rsp) movq %rdx, %r14 movq %rsi, %r13 movq %rdi, %r12 leaq 0x1cee5a(%rip), %rdx # 0x45c344 fldt 0xf0(%rsp) fldt 0xe0(%rsp) fldt 0xd0(%rsp) fstpt 0x80(%rsp) fstpt 0x70(%rsp) fstpt 0x60(%rsp) pushq $0x14 popq %rsi leaq 0x80(%rsp), %rcx leaq 0x70(%rsp), %r8 leaq 0x60(%rsp), %r9 callq 0x126c32 leaq 0x10(%r12), %rax leaq 0x1d620a(%rip), %rbp # 0x46373e movq %rax, 0x38(%rsp) xorl %eax, %eax pushq $0x8 popq %rbx movq %r13, 0x40(%rsp) cmpq 0x48(%rsp), %r14 je 0x28d6a9 movl (%r14), %edx movq 0x38(%rsp), %rsi leaq 0x28(%rsp), %rdi movq %rax, 0x50(%rsp) movq %r14, 0x58(%rsp) movl %edx, 0x20(%rsp) callq 0x3de66 movq 0x28(%rsp), %rdx movq 0x30(%rsp), %rsi fldt 0x60(%rsp) movq %r12, %rdi fstpt (%rsp) callq 0x28d9de movq 0x28(%rsp), %rdx movq 0x30(%rsp), %rsi movq %r12, %rdi movq %r13, %rcx callq 0x28da04 movq 0x58(%r12), %rdi movq 0x8(%r12), %rdx movslq %eax, %r13 movl %eax, %r15d movq %r13, %rsi shlq $0x5, %rsi addq %rdi, %rsi callq 0x28dada movslq 0x20(%rsp), %rax movq 0x60(%r12), %rcx movq %r12, %rdi movl %r13d, %esi movl (%rcx,%rax,8), %edx movl 0x4(%rcx,%rax,8), %ecx callq 0x28db0e movslq 0x20(%rsp), %r8 movl %eax, 0x24(%rsp) movq %r12, %rdi shlq $0x3, %r8 addq 0x60(%r12), %r8 pushq $0x16 popq %rsi leaq 0x1d3a4f(%rip), %rdx # 0x461045 leaq 0x20(%rsp), %rcx callq 0xb91c2 movl %r13d, %eax sarl $0x1f, %eax xorl %r14d, %r14d andnl %r13d, %eax, %r13d shlq $0x5, %r13 cmpq %r14, %r13 je 0x28d638 movq 0x58(%r12), %rcx movq %r12, %rdi movq %rbx, %rsi movq %rbp, %rdx leaq 0x10(%rcx,%r14), %r8 addq %r14, %rcx callq 0x126d84 addq $0x20, %r14 jmp 0x28d612 movslq 0x20(%rsp), %rax movq 0x60(%r12), %rcx movq %r12, %rdi leaq 0x4(%rcx,%rax,8), %rcx pushq $0x16 popq %rsi leaq 0x1d3a08(%rip), %rdx # 0x46105c leaq 0x24(%rsp), %r8 callq 0xb91c2 movq 0x30(%rsp), %rdx movl 0x20(%rsp), %ecx movl 0x24(%rsp), %r8d fldt 0x80(%rsp) fldt 0x70(%rsp) fstpt 0x10(%rsp) fstpt (%rsp) movq %r12, %rdi movl %r15d, %r9d movq 0x40(%rsp), %r13 movq %r13, %rsi callq 0x28db7f movq 0x50(%rsp), %rcx movq 0x58(%rsp), %r14 orb %al, %cl addq $0x8, %r14 movq %rcx, %rax jmp 0x28d543 andb $0x1, %al addq $0x98, %rsp popq %rbx popq %r12 popq %r13 popq %r14 popq %r15 popq %rbp retq nop
/quesnel[P]baryonyx/lib/src/itm-solver-inequalities-01.cpp
int baryonyx::itm::compute_order::infeasibility_local_compute_violated_constraints<baryonyx::itm::solver_inequalities_01coeff<long double, baryonyx::itm::maximize_tag, baryonyx::itm::quadratic_cost_type<long double>, true>, baryonyx::bit_array>(baryonyx::itm::solver_inequalities_01coeff<long double, baryonyx::itm::maximize_tag, baryonyx::itm::quadratic_cost_type<long double>, true>&, baryonyx::bit_array const&)
int infeasibility_local_compute_violated_constraints(Solver& solver, const Xtype& x) { m_order.clear(); for (int k = 0, e = solver.m; k != e; ++k) { sparse_matrix<int>::const_row_iterator it, et; std::tie(it, et) = solver.ap.row(k); int v = 0; for (; it != et; ++it) v += solver.factor(it->value) * x[it->column]; if (solver.bound_min(k) > v) m_order.push_back(std::make_pair(k, solver.bound_min(k) - v)); else if (solver.bound_max(k) < v) m_order.push_back(std::make_pair(k, v - solver.bound_max(k))); } return length(m_order); }
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x38, %rsp movq 0x18(%rdi), %rax movq %rdx, %r14 movq %rsi, 0x20(%rsp) cmpq %rax, 0x20(%rdi) je 0x28d6e2 movq %rax, 0x20(%rdi) movq 0x20(%rsp), %rcx leaq 0x18(%rdi), %rax movq %rdi, 0x28(%rsp) xorl %ebp, %ebp movq %rax, 0x18(%rsp) movl 0x78(%rcx), %r12d leaq 0x10(%rcx), %rax movq %rax, 0x30(%rsp) cmpq %r12, %rbp je 0x28d783 movq 0x30(%rsp), %rsi leaq 0x8(%rsp), %rdi movl %ebp, %edx callq 0x3de66 movq 0x8(%rsp), %r15 movq 0x10(%rsp), %r13 xorl %ebx, %ebx cmpq %r15, %r13 je 0x28d73f movl 0x4(%r13), %esi movq %r14, %rdi callq 0x3664a addl %eax, %ebx addq $0x8, %r13 jmp 0x28d726 movq 0x20(%rsp), %rax movq 0x60(%rax), %rcx movl (%rcx,%rbp,8), %eax subl %ebx, %eax jle 0x28d75d shlq $0x20, %rax orq %rbp, %rax movq %rax, 0x8(%rsp) jmp 0x28d76f subl 0x4(%rcx,%rbp,8), %ebx jle 0x28d77e shlq $0x20, %rbx orq %rbp, %rbx movq %rbx, 0x8(%rsp) movq 0x18(%rsp), %rdi leaq 0x8(%rsp), %rsi callq 0x3c8c2 incq %rbp jmp 0x28d704 movq 0x28(%rsp), %rcx movq 0x20(%rcx), %rax subq 0x18(%rcx), %rax shrq $0x3, %rax addq $0x38, %rsp popq %rbx popq %r12 popq %r13 popq %r14 popq %r15 popq %rbp retq nop
/quesnel[P]baryonyx/lib/src/itm-common.hpp
bool baryonyx::itm::solver_inequalities_01coeff<long double, baryonyx::itm::maximize_tag, baryonyx::itm::quadratic_cost_type<long double>, true>::compute_update_row<baryonyx::bit_array, __gnu_cxx::__normal_iterator<int const*, std::vector<int, std::allocator<int>>>>(baryonyx::bit_array&, __gnu_cxx::__normal_iterator<int const*, std::vector<int, std::allocator<int>>>, __gnu_cxx::__normal_iterator<int const*, std::vector<int, std::allocator<int>>>, long double, long double, long double)
bool compute_update_row(Xtype& x, Iterator first, Iterator last, Float kappa, Float delta, Float theta) { auto at_least_one_pi_changed{ false }; logger::log("update-row {} {} {}\n", kappa, delta, theta); for (; first != last; ++first) { auto k = constraint(first); const auto it = ap.row(k); decrease_preference(std::get<0>(it), std::get<1>(it), theta); const auto r_size = compute_reduced_costs(std::get<0>(it), std::get<1>(it), x); calculator_sort<Mode>(R.get(), R.get() + r_size, rng); int selected = select_variables(r_size, b[k].min, b[k].max); logger::log("constraints {}: {} <= ", k, b[k].min); for (int i = 0; i < r_size; ++i) logger::log("({} {}) ", R[i].value, R[i].id); logger::log("<= {} => Selected: {}\n", b[k].max, selected); auto pi_change = affect( *this, x, std::get<0>(it), k, selected, r_size, kappa, delta); at_least_one_pi_changed = at_least_one_pi_changed || pi_change; } return at_least_one_pi_changed; }
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x98, %rsp movq %rcx, 0x48(%rsp) movq %rdx, %r14 movq %rsi, %r13 movq %rdi, %r12 leaq 0x1ceb7a(%rip), %rdx # 0x45c344 fldt 0xf0(%rsp) fldt 0xe0(%rsp) fldt 0xd0(%rsp) fstpt 0x80(%rsp) fstpt 0x70(%rsp) fstpt 0x60(%rsp) pushq $0x14 popq %rsi leaq 0x80(%rsp), %rcx leaq 0x70(%rsp), %r8 leaq 0x60(%rsp), %r9 callq 0x126c32 leaq 0x10(%r12), %rax leaq 0x1d5f2a(%rip), %rbp # 0x46373e movq %rax, 0x38(%rsp) xorl %eax, %eax pushq $0x8 popq %rbx movq %r13, 0x40(%rsp) cmpq 0x48(%rsp), %r14 je 0x28d989 movl (%r14), %edx movq 0x38(%rsp), %rsi leaq 0x28(%rsp), %rdi movq %rax, 0x50(%rsp) movq %r14, 0x58(%rsp) movl %edx, 0x20(%rsp) callq 0x3de66 movq 0x28(%rsp), %rdx movq 0x30(%rsp), %rsi fldt 0x60(%rsp) movq %r12, %rdi fstpt (%rsp) callq 0x28d9de movq 0x28(%rsp), %rdx movq 0x30(%rsp), %rsi movq %r12, %rdi movq %r13, %rcx callq 0x28da04 movq 0x58(%r12), %rdi movq 0x8(%r12), %rdx movslq %eax, %r13 movl %eax, %r15d movq %r13, %rsi shlq $0x5, %rsi addq %rdi, %rsi callq 0x28dada movslq 0x20(%rsp), %rax movq 0x60(%r12), %rcx movq %r12, %rdi movl %r13d, %esi movl (%rcx,%rax,8), %edx movl 0x4(%rcx,%rax,8), %ecx callq 0x28db0e movslq 0x20(%rsp), %r8 movl %eax, 0x24(%rsp) movq %r12, %rdi shlq $0x3, %r8 addq 0x60(%r12), %r8 pushq $0x16 popq %rsi leaq 0x1d376f(%rip), %rdx # 0x461045 leaq 0x20(%rsp), %rcx callq 0xb91c2 movl %r13d, %eax sarl $0x1f, %eax xorl %r14d, %r14d andnl %r13d, %eax, %r13d shlq $0x5, %r13 cmpq %r14, %r13 je 0x28d918 movq 0x58(%r12), %rcx movq %r12, %rdi movq %rbx, %rsi movq %rbp, %rdx leaq 0x10(%rcx,%r14), %r8 addq %r14, %rcx callq 0x126d84 addq $0x20, %r14 jmp 0x28d8f2 movslq 0x20(%rsp), %rax movq 0x60(%r12), %rcx movq %r12, %rdi leaq 0x4(%rcx,%rax,8), %rcx pushq $0x16 popq %rsi leaq 0x1d3728(%rip), %rdx # 0x46105c leaq 0x24(%rsp), %r8 callq 0xb91c2 movq 0x30(%rsp), %rdx movl 0x20(%rsp), %ecx movl 0x24(%rsp), %r8d fldt 0x80(%rsp) fldt 0x70(%rsp) fstpt 0x10(%rsp) fstpt (%rsp) movq %r12, %rdi movl %r15d, %r9d movq 0x40(%rsp), %r13 movq %r13, %rsi callq 0x28db7f movq 0x50(%rsp), %rcx movq 0x58(%rsp), %r14 orb %al, %cl addq $0x4, %r14 movq %rcx, %rax jmp 0x28d823 andb $0x1, %al addq $0x98, %rsp popq %rbx popq %r12 popq %r13 popq %r14 popq %r15 popq %rbp retq nop
/quesnel[P]baryonyx/lib/src/itm-solver-inequalities-01.cpp
baryonyx::itm::solver_functor<baryonyx::itm::solver_inequalities_01coeff<long double, baryonyx::itm::maximize_tag, baryonyx::itm::quadratic_cost_type<long double>, true>, long double, baryonyx::itm::maximize_tag, baryonyx::itm::quadratic_cost_type<long double>, baryonyx::file_observer>::is_timelimit_reached()
bool is_timelimit_reached() { m_end = std::chrono::steady_clock::now(); return is_time_limit(m_ctx.parameters.time_limit, m_begin, m_end); }
pushq %rbx movq %rdi, %rbx callq 0xb5a0 movq %rax, 0x8(%rbx) vxorpd %xmm2, %xmm2, %xmm2 movq 0x10(%rbx), %rcx subq (%rbx), %rax vmovsd (%rcx), %xmm0 vcvtsi2sd %rax, %xmm1, %xmm1 vdivsd 0x1c6d79(%rip), %xmm1, %xmm1 # 0x459a80 vcmpnlepd %xmm2, %xmm0, %k1 vcmpltpd %xmm1, %xmm0, %k0 {%k1} kmovd %k0, %eax andb $0x1, %al popq %rbx retq nop
/quesnel[P]baryonyx/lib/src/itm-solver-common.hpp
baryonyx::itm::solver_inequalities_01coeff<long double, baryonyx::itm::maximize_tag, baryonyx::itm::quadratic_cost_type<long double>, false>::solver_inequalities_01coeff(std::linear_congruential_engine<unsigned long, 16807ul, 0ul, 2147483647ul>&, int, int, baryonyx::itm::quadratic_cost_type<long double> const&, std::vector<baryonyx::itm::merged_constraint, std::allocator<baryonyx::itm::merged_constraint>> const&)
solver_inequalities_01coeff(random_engine& rng_, int m_, int n_, const cost_type& c_, const std::vector<merged_constraint>& csts) : logger("solver_inequalities_01coeff") , rng(rng_) , ap(csts, m_, n_) , P(std::make_unique<Float[]>(ap.size())) , R(std::make_unique<rc_data[]>(compute_reduced_costs_vector_size(csts))) , b(std::make_unique<bound_factor[]>(m_)) , pi(std::make_unique<Float[]>(m_)) , c(c_) , m(m_) , n(n_) { for (int i = 0; i != m; ++i) { #if !defined(BARYONYX_FULL_OPTIMIZATION) // mscv 15.9.6 fail to build this line: // for ([[maybe_unused]] const auto& cst : csts[i].elements) // bx_ensures(cst.factor == 1); for (const auto& cst : csts[i].elements) bx_ensures(cst.factor == 1); #endif if (csts[i].min == csts[i].max) { b[i].min = csts[i].min; b[i].max = csts[i].max; } else { b[i].min = std::max(0, csts[i].min); b[i].max = std::min(length(csts[i].elements), csts[i].max); } bx_ensures(b[i].min <= b[i].max); } }
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x28, %rsp andq $0x0, (%rdi) leaq 0x10(%rdi), %rbx movq %rsi, 0x8(%rdi) movq %rdi, %r13 movq %r9, %r15 movq %r8, 0x20(%rsp) movl %ecx, %r12d movl %edx, %ebp movq %r9, %rsi movq %rbx, %rdi callq 0x3a4ca movl 0x38(%r13), %esi leaq 0x50(%r13), %r14 movq %r14, %rdi movq %rbx, 0x18(%rsp) callq 0x8afb9 movq %r15, %rdi movq %r14, 0x10(%rsp) leaq 0x58(%r13), %rbx callq 0x3a89d movq %rbx, 0x8(%rsp) movq %rbx, %rdi movq %rax, %rsi callq 0x293edb leaq 0x60(%r13), %rbx movslq %ebp, %r14 movq %rbx, %rdi movq %r14, %rsi callq 0x293f1b leaq 0x68(%r13), %rdi movq %r14, %rsi callq 0x8afb9 movq 0x20(%rsp), %rax movl %ebp, %edx xorl %esi, %esi movq %rax, 0x70(%r13) movl %ebp, 0x78(%r13) movl %r12d, 0x7c(%r13) movq (%r15), %rax movq 0x60(%r13), %rcx cmpq %rdx, %rsi je 0x29391f imulq $0x28, %rsi, %rdi movq (%rax,%rdi), %r9 leaq (%rax,%rdi), %r8 movq 0x8(%rax,%rdi), %rdi movq %r9, %r10 cmpq %rdi, %r10 je 0x2938b9 cmpl $0x1, (%r10) leaq 0x8(%r10), %r10 je 0x2938a8 jmp 0x29392e movl 0x18(%r8), %r11d cmpl 0x1c(%r8), %r11d jne 0x2938cc movl %r11d, (%rcx,%rsi,8) movl %r11d, %r10d jmp 0x2938f1 movl %r11d, %r10d sarl $0x1f, %r10d subq %r9, %rdi andnl %r11d, %r10d, %r10d shrq $0x3, %rdi movl %r10d, (%rcx,%rsi,8) movl 0x1c(%r8), %r8d cmpl %edi, %r8d cmovll %r8d, %edi movl %edi, %r11d movl %r11d, 0x4(%rcx,%rsi,8) incq %rsi cmpl %r11d, %r10d jle 0x29388b leaq 0x1c4b25(%rip), %rdi # 0x45842a leaq 0x1c699d(%rip), %rsi # 0x45a2a9 leaq 0x1cd6c2(%rip), %rdx # 0x460fd5 leaq 0x1cd727(%rip), %rcx # 0x461041 callq 0x2813f addq $0x28, %rsp popq %rbx popq %r12 popq %r13 popq %r14 popq %r15 popq %rbp retq leaq 0x1c4af5(%rip), %rdi # 0x45842a leaq 0x1c891f(%rip), %rsi # 0x45c25b leaq 0x1cd692(%rip), %rdx # 0x460fd5 leaq 0x1c6cce(%rip), %rcx # 0x45a618 callq 0x2813f movq %rbx, %rdi movq %rax, %r15 callq 0x293f5c jmp 0x29395f movq %rax, %r15 movq 0x8(%rsp), %rdi callq 0x293f76 jmp 0x29396e movq %rax, %r15 movq 0x10(%rsp), %rdi callq 0x8affa jmp 0x29397d movq %rax, %r15 movq 0x18(%rsp), %rdi callq 0x3a954 movq %r15, %rdi callq 0xb3d0 nop
/quesnel[P]baryonyx/lib/src/itm-solver-inequalities-01.cpp
int baryonyx::itm::compute_order::run<baryonyx::itm::solver_inequalities_01coeff<long double, baryonyx::itm::maximize_tag, baryonyx::itm::quadratic_cost_type<long double>, false>, baryonyx::bit_array, long double>(baryonyx::itm::solver_inequalities_01coeff<long double, baryonyx::itm::maximize_tag, baryonyx::itm::quadratic_cost_type<long double>, false>&, baryonyx::bit_array&, std::linear_congruential_engine<unsigned long, 16807ul, 0ul, 2147483647ul>&, long double, long double, long double)
int run(Solver& solver, Xtype& x, random_engine& rng, Float kappa, Float delta, Float theta) { bool pi_changed = false; int remaining = 0; switch (order) { case solver_parameters::constraint_order::reversing: solver.compute_update_row( x, R.crbegin(), R.crend(), kappa, delta, theta); return compute_violated_constraints(solver, x, R); case solver_parameters::constraint_order::random_sorting: std::shuffle(R.begin(), R.end(), rng); solver.compute_update_row( x, R.begin(), R.end(), kappa, delta, theta); return compute_violated_constraints(solver, x, R); case solver_parameters::constraint_order::infeasibility_decr: std::sort(m_order.begin(), m_order.end(), [](const auto& lhs, const auto& rhs) { return rhs.second < lhs.second; }); solver.compute_update_row( x, m_order.cbegin(), m_order.cend(), kappa, delta, theta); return infeasibility_local_compute_violated_constraints(solver, x); case solver_parameters::constraint_order::infeasibility_incr: std::sort(m_order.begin(), m_order.end(), [](const auto& lhs, const auto& rhs) { return lhs.second < rhs.second; }); solver.compute_update_row( x, m_order.cbegin(), m_order.cend(), kappa, delta, theta); return infeasibility_local_compute_violated_constraints(solver, x); case solver_parameters::constraint_order::lagrangian_decr: std::sort( R.begin(), R.end(), [&solver](const auto lhs, const auto rhs) { return solver.pi[rhs] < solver.pi[lhs]; }); solver.compute_update_row( x, R.cbegin(), R.cend(), kappa, delta, theta); return compute_violated_constraints(solver, x, R); case solver_parameters::constraint_order::lagrangian_incr: std::sort( R.begin(), R.end(), [&solver](const auto lhs, const auto rhs) { return solver.pi[lhs] < solver.pi[rhs]; }); solver.compute_update_row( x, R.cbegin(), R.cend(), kappa, delta, theta); return compute_violated_constraints(solver, x, R); case solver_parameters::constraint_order::pi_sign_change: std::shuffle(R.begin(), R.end(), rng); pi_changed = solver.compute_update_row( x, R.begin(), R.end(), kappa, delta, theta); remaining = local_compute_violated_constraints(solver, x); if (!pi_changed && remaining == 0) return 0; return remaining; case solver_parameters::constraint_order::none: default: solver.compute_update_row( x, R.begin(), R.end(), kappa, delta, theta); return compute_violated_constraints(solver, x, R); } }
pushq %r15 pushq %r14 pushq %rbx subq $0x70, %rsp fldt 0xb0(%rsp) fldt 0xa0(%rsp) fldt 0x90(%rsp) movl 0x30(%rdi), %eax movq %rdx, %rbx movq %rsi, %r15 movq %rdi, %r14 decl %eax cmpl $0x6, %eax ja 0x293af3 leaq 0x1cd45a(%rip), %rdx # 0x460e68 movslq (%rdx,%rax,4), %rax addq %rdx, %rax jmpq *%rax movq 0x8(%r14), %rax leaq 0x68(%rsp), %rdx leaq 0x60(%rsp), %rcx movq %r15, %rdi movq %rbx, %rsi movq %rax, (%rdx) movq (%r14), %rax movq %rax, (%rcx) fxch %st(2) fstpt 0x20(%rsp) fstpt 0x10(%rsp) fstpt (%rsp) callq 0x29416e movq %r15, %rdi movq %rbx, %rsi movq %r14, %rdx callq 0x294091 addq $0x70, %rsp popq %rbx popq %r14 popq %r15 retq movq (%r14), %rdi movq 0x8(%r14), %rsi fstpt 0x54(%rsp) fstpt 0x48(%rsp) movq %r15, %rdx fstpt 0x3c(%rsp) callq 0x295af3 jmp 0x293b67 movq 0x18(%r14), %rdi movq 0x20(%r14), %rsi fstpt 0x54(%rsp) fstpt 0x48(%rsp) fstpt 0x3c(%rsp) callq 0x29521d jmp 0x293ab2 movq 0x18(%r14), %rdi movq 0x20(%r14), %rsi fstpt 0x54(%rsp) fstpt 0x48(%rsp) fstpt 0x3c(%rsp) callq 0x295688 movq 0x18(%r14), %rdx movq 0x20(%r14), %rcx fldt 0x3c(%rsp) fstpt 0x20(%rsp) movq %r15, %rdi movq %rbx, %rsi fldt 0x48(%rsp) fstpt 0x10(%rsp) fldt 0x54(%rsp) fstpt (%rsp) callq 0x29439c movq %r14, %rdi movq %r15, %rsi movq %rbx, %rdx addq $0x70, %rsp popq %rbx popq %r14 popq %r15 jmp 0x2944ac movq (%r14), %rdx movq 0x8(%r14), %rcx fxch %st(2) fstpt 0x20(%rsp) fstpt 0x10(%rsp) jmp 0x293b3c movq (%r14), %rdi movq 0x8(%r14), %rsi fstpt 0x54(%rsp) fstpt 0x48(%rsp) movq %rcx, %rdx fstpt 0x3c(%rsp) callq 0x3ceed movq (%r14), %rdx movq 0x8(%r14), %rcx fldt 0x3c(%rsp) fstpt 0x20(%rsp) fldt 0x48(%rsp) fstpt 0x10(%rsp) fldt 0x54(%rsp) fstpt (%rsp) movq %r15, %rdi movq %rbx, %rsi callq 0x29428c jmp 0x293b90 movq (%r14), %rdi movq 0x8(%r14), %rsi fstpt 0x54(%rsp) fstpt 0x48(%rsp) movq %r15, %rdx fstpt 0x3c(%rsp) callq 0x296047 movq (%r14), %rdx movq 0x8(%r14), %rcx fldt 0x3c(%rsp) fstpt 0x20(%rsp) movq %r15, %rdi movq %rbx, %rsi fldt 0x48(%rsp) fstpt 0x10(%rsp) fldt 0x54(%rsp) fstpt (%rsp) callq 0x294592 movq %r15, %rdi movq %rbx, %rsi movq %r14, %rdx addq $0x70, %rsp popq %rbx popq %r14 popq %r15 jmp 0x294091 movq (%r14), %rdi movq 0x8(%r14), %rsi fstpt 0x54(%rsp) fstpt 0x48(%rsp) movq %rcx, %rdx fstpt 0x3c(%rsp) callq 0x3ceed movq (%r14), %rdx movq 0x8(%r14), %rcx fldt 0x3c(%rsp) fstpt 0x20(%rsp) movq %r15, %rdi movq %rbx, %rsi fldt 0x48(%rsp) fstpt 0x10(%rsp) fldt 0x54(%rsp) fstpt (%rsp) callq 0x29428c movq %r14, %rdi movq %r15, %rsi movq %rbx, %rdx addq $0x70, %rsp popq %rbx popq %r14 popq %r15 jmp 0x2946a2
/quesnel[P]baryonyx/lib/src/itm-common.hpp
baryonyx::itm::solver_functor<baryonyx::itm::solver_inequalities_01coeff<long double, baryonyx::itm::maximize_tag, baryonyx::itm::quadratic_cost_type<long double>, false>, long double, baryonyx::itm::maximize_tag, baryonyx::itm::quadratic_cost_type<long double>, baryonyx::file_observer>::operator()(std::vector<baryonyx::itm::merged_constraint, std::allocator<baryonyx::itm::merged_constraint>> const&, int, baryonyx::itm::quadratic_cost_type<long double> const&, double)
result operator()(const std::vector<merged_constraint>& constraints, int variables, const Cost& original_costs, double cost_constant) { result r; bit_array x(variables); int best_remaining = INT_MAX; auto& p = m_ctx.parameters; auto norm_costs = normalize_costs<Float, Cost>( m_ctx, original_costs, m_rng, variables); const auto kappa_step = static_cast<Float>(p.kappa_step); const auto kappa_max = static_cast<Float>(p.kappa_max); const auto alpha = static_cast<Float>(p.alpha); const auto theta = static_cast<Float>(p.theta); const auto delta = p.delta < 0 ? compute_delta<Float, Cost>(m_ctx, norm_costs, theta, variables) : static_cast<Float>(p.delta); const auto pushing_k_factor = static_cast<Float>(p.pushing_k_factor); const auto pushing_objective_amplifier = static_cast<Float>(p.pushing_objective_amplifier); const long int w_limit = static_cast<long int>(p.w); Solver slv( m_rng, length(constraints), variables, norm_costs, constraints); compute_order compute(p.order, variables); { std::bernoulli_distribution choose_mutation( m_ctx.parameters.init_policy_random); bit_array empty_x; switch (p.init_policy) { case solver_parameters::init_policy_type::pessimistic_solve: init_with_pre_solve<Cost, Mode>( x, empty_x, m_rng, original_costs, constraints, 1.0); break; case solver_parameters::init_policy_type::optimistic_solve: init_with_pre_solve<Cost, Mode>( empty_x, x, m_rng, original_costs, constraints, 1.0); break; case solver_parameters::init_policy_type::bastert: init_with_bastert<Cost, Mode>(x, original_costs, variables, 0); break; } for (int i = 0, e = x.size(); i != e; ++i) if (choose_mutation(m_rng)) x.invert(i); } bool start_push = false; auto kappa = static_cast<Float>(p.kappa_min); Observer obs("img", slv.m, slv.n, p.limit); m_begin = std::chrono::steady_clock::now(); m_end = std::chrono::steady_clock::now(); compute.init(slv, x); for (long int i = 0; i != p.limit; ++i) { auto remaining = compute.run(slv, x, m_rng, kappa, delta, theta); obs.make_observation(slv.ap, slv.P.get(), slv.pi.get()); if (remaining == 0) { store_if_better( x, original_costs.results(x, cost_constant), i); best_remaining = remaining; start_push = true; break; } if (remaining < best_remaining) { store_if_better(x, remaining, i); best_remaining = remaining; } if (i > w_limit) kappa += kappa_step * std::pow(static_cast<Float>(remaining) / static_cast<Float>(slv.m), alpha); if (kappa > kappa_max) { r.status = result_status::kappa_max_reached; break; } if (is_timelimit_reached()) { r.status = result_status::time_limit_reached; break; } } if (!start_push) { r.status = result_status::limit_reached; } else { for (int push = 0; push < p.pushes_limit; ++push) { auto remaining = compute.push_and_run(slv, x, m_rng, pushing_k_factor * kappa, delta, theta, pushing_objective_amplifier); if (remaining == 0) store_if_better(x, original_costs.results(x, cost_constant), -push * p.pushing_iteration_limit - 1); if (is_timelimit_reached()) break; for (int iter = 0; iter < p.pushing_iteration_limit; ++iter) { remaining = compute.run(slv, x, m_rng, kappa, delta, theta); if (remaining == 0) { store_if_better( x, original_costs.results(x, cost_constant), -push * p.pushing_iteration_limit - iter - 1); break; } if (iter > p.w) kappa += kappa_step * std::pow(static_cast<Float>(remaining) / static_cast<Float>(slv.m), alpha); if (kappa > kappa_max) break; if (is_timelimit_reached()) break; } } } if (m_best.remaining_constraints == 0) r.status = result_status::success; if (!m_best.x.empty()) { r.solutions.resize(1); convert(m_best, r.solutions[0], variables); } return r; }
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x438, %rsp # imm = 0x438 vmovsd %xmm0, 0x88(%rsp) vxorpd %xmm0, %xmm0, %xmm0 movq %r8, %r13 movl %ecx, %ebx movq %rdx, %r15 movq %rsi, %r14 movq %rdi, 0x60(%rsp) vmovupd %zmm0, 0x58(%rdi) vmovupd %zmm0, 0x40(%rdi) vmovupd %zmm0, (%rdi) movq $0x7fffffff, 0x98(%rdi) # imm = 0x7FFFFFFF movl $0x2, 0xa0(%rdi) leaq 0x70(%rsp), %rdi movl %ecx, %esi vzeroupper callq 0x368ae movq 0x10(%r14), %r12 movq 0x18(%r14), %rcx leaq 0x108(%rsp), %rdi movq %r12, %rsi movq %r13, %rdx movl %ebx, %r8d callq 0x98bf0 fldl 0x20(%r12) fstpt 0xc8(%rsp) vxorpd %xmm1, %xmm1, %xmm1 fldl 0x28(%r12) fstpt 0xd4(%rsp) fldl 0x30(%r12) fstpt 0xbc(%rsp) fldl 0x8(%r12) vmovsd 0x10(%r12), %xmm0 fld %st(0) fstpt 0x54(%rsp) vucomisd %xmm0, %xmm1 jbe 0x2987e4 movq 0x10(%r14), %rdi leaq 0x108(%rsp), %rsi movl %ebx, %edx fstpt (%rsp) callq 0x98cca fstpt 0xa0(%rsp) jmp 0x2987fd fstp %st(0) vmovsd %xmm0, 0xe8(%rsp) fldl 0xe8(%rsp) fstpt 0xa0(%rsp) fldl 0x38(%r12) fstpt 0xfc(%rsp) fldl 0x40(%r12) fstpt 0xf0(%rsp) vcvttsd2si 0xa0(%r12), %rax movq 0x18(%r14), %rsi movq %rax, 0x90(%rsp) movq 0x8(%r15), %rax subq (%r15), %rax pushq $0x28 popq %rcx cqto idivq %rcx leaq 0x160(%rsp), %rdi leaq 0x108(%rsp), %r8 movl %eax, %edx movl %ebx, %ecx movq %r15, %r9 callq 0x2937e6 movl 0xcc(%r12), %esi leaq 0x128(%rsp), %rdi movl %ebx, %edx callq 0x38cd8 movq 0x10(%r14), %rax movl %ebx, 0x9c(%rsp) vmovsd 0x48(%rax), %xmm0 vmovsd %xmm0, 0x48(%rsp) vxorpd %xmm0, %xmm0, %xmm0 vmovapd %xmm0, 0x1e0(%rsp) movl 0xd4(%r12), %eax testl %eax, %eax je 0x2988df cmpl $0x1, %eax je 0x2988b9 cmpl $0x2, %eax jne 0x2988f0 movq 0x18(%r14), %rdx leaq 0x1e0(%rsp), %rdi leaq 0x70(%rsp), %rsi jmp 0x2988ca movq 0x18(%r14), %rdx leaq 0x70(%rsp), %rdi leaq 0x1e0(%rsp), %rsi vmovsd 0x1bb766(%rip), %xmm0 # 0x454038 movq %r13, %rcx movq %r15, %r8 callq 0x96db2 jmp 0x2988f0 leaq 0x70(%rsp), %rdi movq %r13, %rsi movl %ebx, %edx xorl %ecx, %ecx callq 0x96d3f leaq 0x70(%rsp), %r15 movq %r13, 0x80(%rsp) xorl %ebx, %ebx movl (%r15), %ebp cmpl %ebx, %ebp je 0x298929 movq 0x18(%r14), %rdi callq 0x36ac3 vmovsd 0x48(%rsp), %xmm1 vucomisd %xmm0, %xmm1 jbe 0x298925 movq %r15, %rdi movl %ebx, %esi callq 0x3626e incl %ebx jmp 0x298902 leaq 0x1e8(%rsp), %rdi callq 0x357fe vmovsd 0x18(%r12), %xmm0 movl 0x1d8(%rsp), %ecx movl 0x1dc(%rsp), %r8d movq 0xb0(%r12), %r9 vmovsd %xmm0, 0x48(%rsp) leaq 0x1c1c16(%rip), %rdx # 0x45a577 leaq 0x1e0(%rsp), %rdi pushq $0x3 popq %rsi callq 0x4346e callq 0xb5a0 movq %rax, (%r14) callq 0xb5a0 movq %rax, 0x8(%r14) leaq 0x128(%rsp), %rdi leaq 0x160(%rsp), %rsi leaq 0x70(%rsp), %rdx callq 0x293990 vmovsd 0x48(%rsp), %xmm0 movl $0x7fffffff, %ebp # imm = 0x7FFFFFFF leaq 0x70(%rsp), %r13 xorl %r15d, %r15d vmovsd %xmm0, 0xe0(%rsp) fldl 0xe0(%rsp) fldt 0x54(%rsp) cmpq 0xb0(%r12), %r15 je 0x298ad2 movq 0x18(%r14), %rcx fstpt 0x20(%rsp) leaq 0x128(%rsp), %rdi leaq 0x160(%rsp), %rsi movq %r13, %rdx fldt 0xa0(%rsp) fstpt 0x10(%rsp) fld %st(0) fstpt 0x48(%rsp) fstpt (%rsp) callq 0x2939d2 movq 0x1b0(%rsp), %rdx movq 0x1c8(%rsp), %rcx movl %eax, %ebx leaq 0x1e0(%rsp), %rdi leaq 0x170(%rsp), %rsi callq 0x95692 testl %ebx, %ebx je 0x298b86 cmpl %ebp, %ebx jge 0x298a4a movq %r14, %rdi movq %r13, %rsi movl %ebx, %edx movq %r15, %rcx callq 0x299640 movl %ebx, %ebp cmpq 0x90(%rsp), %r15 jle 0x298a9b movl %ebx, 0xb4(%rsp) fildl 0xb4(%rsp) movl 0x1d8(%rsp), %eax movl %eax, 0xb8(%rsp) fidivl 0xb8(%rsp) fldt 0xbc(%rsp) fstpt 0x10(%rsp) fstpt (%rsp) callq 0xb110 fldt 0xc8(%rsp) fmulp %st, %st(1) fldt 0x48(%rsp) faddp %st, %st(1) jmp 0x298a9f fldt 0x48(%rsp) fldt 0xd4(%rsp) fxch %st(1) fucomi %st(1), %st fstp %st(1) fldz ja 0x298ad2 fstp %st(0) movq %r14, %rdi fstpt 0x48(%rsp) callq 0x299678 fldt 0x48(%rsp) fldt 0x54(%rsp) testb %al, %al jne 0x298ad2 incq %r15 jmp 0x2989c3 fstp %st(0) fstp %st(0) movq 0x60(%rsp), %rax movl $0x5, 0xa0(%rax) fldz fldz fstp %st(1) fstp %st(0) cmpl $0x0, 0x50(%r14) jne 0x298b00 movq 0x60(%rsp), %rax andl $0x0, 0xa0(%rax) cmpq $0x0, 0x28(%r14) je 0x298b31 movq 0x60(%rsp), %rax leaq 0x68(%rax), %rbx pushq $0x1 popq %rsi movq %rbx, %rdi callq 0x35292 movq (%rbx), %rsi addq $0x20, %r14 movl 0x9c(%rsp), %edx movq %r14, %rdi callq 0x35781 leaq 0x1e0(%rsp), %rdi callq 0x435d0 leaq 0x128(%rsp), %rdi callq 0x393ac leaq 0x160(%rsp), %rdi callq 0x293ea6 leaq 0x108(%rsp), %rdi callq 0x968b6 leaq 0x78(%rsp), %rdi callq 0x357fe movq 0x60(%rsp), %rax addq $0x438, %rsp # imm = 0x438 popq %rbx popq %r12 popq %r13 popq %r14 popq %r15 popq %rbp retq vmovsd 0x88(%rsp), %xmm0 movq 0x80(%rsp), %rdi leaq 0x70(%rsp), %rbx movq %rbx, %rsi callq 0x9706e movq %r14, %rdi movq %rbx, %rsi movq %r15, %rdx callq 0x2995f6 movl $0x0, 0x6c(%rsp) leaq 0x160(%rsp), %r15 leaq 0x70(%rsp), %rbp xorl %ebx, %ebx fldt 0x54(%rsp) fldt 0x48(%rsp) cmpl 0xbc(%r12), %ebx jge 0x298ae9 movq 0x18(%r14), %rcx fldt 0xfc(%rsp) fld %st(1) fstpt 0x48(%rsp) fmulp %st, %st(1) fldt 0xf0(%rsp) fstpt 0x30(%rsp) fxch %st(1) fstpt 0x20(%rsp) leaq 0x128(%rsp), %rdi movq %r15, %rsi movq %rbp, %rdx fldt 0xa0(%rsp) fstpt 0x10(%rsp) fstpt (%rsp) callq 0x293c02 testl %eax, %eax jne 0x298c5c vmovsd 0x88(%rsp), %xmm0 movq 0x80(%rsp), %rdi movq %rbp, %rsi callq 0x9706e movl 0xc0(%r12), %eax imull %ebx, %eax notl %eax movslq %eax, %rdx movq %r14, %rdi movq %rbp, %rsi callq 0x2995f6 movq %r14, %rdi callq 0x299678 fldt 0x54(%rsp) fldt 0x48(%rsp) testb %al, %al jne 0x298ae9 movl %ebx, 0x90(%rsp) pushq $0x1 popq %rax movl %eax, %ebx leal -0x1(%rbx), %r13d cmpl 0xc0(%r12), %r13d jge 0x298da5 movq 0x18(%r14), %rcx fxch %st(1) fstpt 0x20(%rsp) leaq 0x128(%rsp), %rdi movq %r15, %rsi movq %rbp, %rdx fldt 0xa0(%rsp) fstpt 0x10(%rsp) fld %st(0) fstpt 0x48(%rsp) fstpt (%rsp) callq 0x2939d2 testl %eax, %eax je 0x298d55 vcvtsi2sd %r13d, %xmm1, %xmm0 vucomisd 0xa0(%r12), %xmm0 jbe 0x298d23 movl %eax, 0xac(%rsp) fildl 0xac(%rsp) movl 0x1d8(%rsp), %eax movl %eax, 0xb0(%rsp) fidivl 0xb0(%rsp) fldt 0xbc(%rsp) fstpt 0x10(%rsp) fstpt (%rsp) callq 0xb110 fldt 0xc8(%rsp) fmulp %st, %st(1) fldt 0x48(%rsp) faddp %st, %st(1) jmp 0x298d27 fldt 0x48(%rsp) fldt 0xd4(%rsp) fxch %st(1) fucomi %st(1), %st fstp %st(1) ja 0x298d95 movq %r14, %rdi fstpt 0x48(%rsp) callq 0x299678 testb %al, %al jne 0x298d9d fldt 0x54(%rsp) fldt 0x48(%rsp) incl %ebx jmp 0x298c80 vmovsd 0x88(%rsp), %xmm0 movq 0x80(%rsp), %rdi movq %rbp, %rsi callq 0x9706e movl 0xc0(%r12), %eax imull 0x6c(%rsp), %eax subl %ebx, %eax movslq %eax, %rdx movq %r14, %rdi movq %rbp, %rsi callq 0x2995f6 fldt 0x54(%rsp) fldt 0x48(%rsp) jmp 0x298da5 fldt 0x54(%rsp) fxch %st(1) jmp 0x298da5 fldt 0x54(%rsp) fldt 0x48(%rsp) movl 0x90(%rsp), %ebx decl 0x6c(%rsp) incl %ebx jmp 0x298bd1 jmp 0x298df2 jmp 0x298dcb fstp %st(1) fstp %st(0) jmp 0x298df2 movq %rax, %rbx jmp 0x298e02 movq %rax, %rbx jmp 0x298e0f movq %rax, %rbx jmp 0x298e1c movq %rax, %rbx jmp 0x298e29 movq %rax, %rbx jmp 0x298e33 jmp 0x298df2 jmp 0x298df2 jmp 0x298df2 leaq 0x1e8(%rsp), %rdi movq %rax, %rbx callq 0x357fe jmp 0x298e02 leaq 0x1e0(%rsp), %rdi movq %rax, %rbx callq 0x435d0 leaq 0x128(%rsp), %rdi callq 0x393ac leaq 0x160(%rsp), %rdi callq 0x293ea6 leaq 0x108(%rsp), %rdi callq 0x968b6 leaq 0x78(%rsp), %rdi callq 0x357fe movq 0x60(%rsp), %rdi callq 0xea60 movq %rbx, %rdi callq 0xb3d0 nop
/quesnel[P]baryonyx/lib/src/itm-solver-common.hpp
baryonyx::itm::solver_functor<baryonyx::itm::solver_inequalities_01coeff<long double, baryonyx::itm::maximize_tag, baryonyx::itm::quadratic_cost_type<long double>, false>, long double, baryonyx::itm::maximize_tag, baryonyx::itm::quadratic_cost_type<long double>, baryonyx::none_observer>::operator()(std::vector<baryonyx::itm::merged_constraint, std::allocator<baryonyx::itm::merged_constraint>> const&, int, baryonyx::itm::quadratic_cost_type<long double> const&, double)
result operator()(const std::vector<merged_constraint>& constraints, int variables, const Cost& original_costs, double cost_constant) { result r; bit_array x(variables); int best_remaining = INT_MAX; auto& p = m_ctx.parameters; auto norm_costs = normalize_costs<Float, Cost>( m_ctx, original_costs, m_rng, variables); const auto kappa_step = static_cast<Float>(p.kappa_step); const auto kappa_max = static_cast<Float>(p.kappa_max); const auto alpha = static_cast<Float>(p.alpha); const auto theta = static_cast<Float>(p.theta); const auto delta = p.delta < 0 ? compute_delta<Float, Cost>(m_ctx, norm_costs, theta, variables) : static_cast<Float>(p.delta); const auto pushing_k_factor = static_cast<Float>(p.pushing_k_factor); const auto pushing_objective_amplifier = static_cast<Float>(p.pushing_objective_amplifier); const long int w_limit = static_cast<long int>(p.w); Solver slv( m_rng, length(constraints), variables, norm_costs, constraints); compute_order compute(p.order, variables); { std::bernoulli_distribution choose_mutation( m_ctx.parameters.init_policy_random); bit_array empty_x; switch (p.init_policy) { case solver_parameters::init_policy_type::pessimistic_solve: init_with_pre_solve<Cost, Mode>( x, empty_x, m_rng, original_costs, constraints, 1.0); break; case solver_parameters::init_policy_type::optimistic_solve: init_with_pre_solve<Cost, Mode>( empty_x, x, m_rng, original_costs, constraints, 1.0); break; case solver_parameters::init_policy_type::bastert: init_with_bastert<Cost, Mode>(x, original_costs, variables, 0); break; } for (int i = 0, e = x.size(); i != e; ++i) if (choose_mutation(m_rng)) x.invert(i); } bool start_push = false; auto kappa = static_cast<Float>(p.kappa_min); Observer obs("img", slv.m, slv.n, p.limit); m_begin = std::chrono::steady_clock::now(); m_end = std::chrono::steady_clock::now(); compute.init(slv, x); for (long int i = 0; i != p.limit; ++i) { auto remaining = compute.run(slv, x, m_rng, kappa, delta, theta); obs.make_observation(slv.ap, slv.P.get(), slv.pi.get()); if (remaining == 0) { store_if_better( x, original_costs.results(x, cost_constant), i); best_remaining = remaining; start_push = true; break; } if (remaining < best_remaining) { store_if_better(x, remaining, i); best_remaining = remaining; } if (i > w_limit) kappa += kappa_step * std::pow(static_cast<Float>(remaining) / static_cast<Float>(slv.m), alpha); if (kappa > kappa_max) { r.status = result_status::kappa_max_reached; break; } if (is_timelimit_reached()) { r.status = result_status::time_limit_reached; break; } } if (!start_push) { r.status = result_status::limit_reached; } else { for (int push = 0; push < p.pushes_limit; ++push) { auto remaining = compute.push_and_run(slv, x, m_rng, pushing_k_factor * kappa, delta, theta, pushing_objective_amplifier); if (remaining == 0) store_if_better(x, original_costs.results(x, cost_constant), -push * p.pushing_iteration_limit - 1); if (is_timelimit_reached()) break; for (int iter = 0; iter < p.pushing_iteration_limit; ++iter) { remaining = compute.run(slv, x, m_rng, kappa, delta, theta); if (remaining == 0) { store_if_better( x, original_costs.results(x, cost_constant), -push * p.pushing_iteration_limit - iter - 1); break; } if (iter > p.w) kappa += kappa_step * std::pow(static_cast<Float>(remaining) / static_cast<Float>(slv.m), alpha); if (kappa > kappa_max) break; if (is_timelimit_reached()) break; } } } if (m_best.remaining_constraints == 0) r.status = result_status::success; if (!m_best.x.empty()) { r.solutions.resize(1); convert(m_best, r.solutions[0], variables); } return r; }
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x1e8, %rsp # imm = 0x1E8 vmovsd %xmm0, 0x80(%rsp) vxorpd %xmm0, %xmm0, %xmm0 movq %r8, %r13 movl %ecx, %ebx movq %rdx, %r15 movq %rsi, %r14 movq %rdi, 0x58(%rsp) vmovupd %zmm0, 0x58(%rdi) vmovupd %zmm0, 0x40(%rdi) vmovupd %zmm0, (%rdi) movq $0x7fffffff, 0x98(%rdi) # imm = 0x7FFFFFFF movl $0x2, 0xa0(%rdi) leaq 0x68(%rsp), %rdi movl %ecx, %esi vzeroupper callq 0x368ae movq 0x10(%r14), %r12 movq 0x18(%r14), %rcx leaq 0x110(%rsp), %rdi movq %r12, %rsi movq %r13, %rdx movl %ebx, %r8d callq 0x98bf0 fldl 0x20(%r12) fstpt 0xd0(%rsp) vxorpd %xmm1, %xmm1, %xmm1 fldl 0x28(%r12) fstpt 0xdc(%rsp) fldl 0x30(%r12) fstpt 0xc4(%rsp) fldl 0x8(%r12) vmovsd 0x10(%r12), %xmm0 fld %st(0) fstpt 0x4c(%rsp) vucomisd %xmm0, %xmm1 jbe 0x298f2c movq 0x10(%r14), %rdi leaq 0x110(%rsp), %rsi movl %ebx, %edx fstpt (%rsp) callq 0x98cca fstpt 0x94(%rsp) jmp 0x298f45 fstp %st(0) vmovsd %xmm0, 0xf0(%rsp) fldl 0xf0(%rsp) fstpt 0x94(%rsp) fldl 0x38(%r12) fstpt 0x104(%rsp) fldl 0x40(%r12) fstpt 0xf8(%rsp) vcvttsd2si 0xa0(%r12), %rax movq 0x18(%r14), %rsi movq %rax, 0x88(%rsp) movq 0x8(%r15), %rax subq (%r15), %rax pushq $0x28 popq %rcx cqto idivq %rcx leaq 0x168(%rsp), %rdi leaq 0x110(%rsp), %r8 movl %eax, %edx movl %ebx, %ecx movq %r15, %r9 callq 0x2937e6 movl 0xcc(%r12), %esi leaq 0x130(%rsp), %rdi movl %ebx, %edx callq 0x38cd8 movq 0x10(%r14), %rax movl %ebx, 0x90(%rsp) vmovsd 0x48(%rax), %xmm0 vmovsd %xmm0, 0x40(%rsp) vxorpd %xmm0, %xmm0, %xmm0 vmovapd %xmm0, 0xb0(%rsp) movl 0xd4(%r12), %eax testl %eax, %eax je 0x299027 cmpl $0x1, %eax je 0x299001 cmpl $0x2, %eax jne 0x299038 movq 0x18(%r14), %rdx leaq 0xb0(%rsp), %rdi leaq 0x68(%rsp), %rsi jmp 0x299012 movq 0x18(%r14), %rdx leaq 0x68(%rsp), %rdi leaq 0xb0(%rsp), %rsi vmovsd 0x1bb01e(%rip), %xmm0 # 0x454038 movq %r13, %rcx movq %r15, %r8 callq 0x96db2 jmp 0x299038 leaq 0x68(%rsp), %rdi movq %r13, %rsi movl %ebx, %edx xorl %ecx, %ecx callq 0x96d3f leaq 0x68(%rsp), %r15 movq %r13, 0x78(%rsp) xorl %ebx, %ebx movl (%r15), %ebp cmpl %ebx, %ebp je 0x29906e movq 0x18(%r14), %rdi callq 0x36ac3 vmovsd 0x40(%rsp), %xmm1 vucomisd %xmm0, %xmm1 jbe 0x29906a movq %r15, %rdi movl %ebx, %esi callq 0x3626e incl %ebx jmp 0x299047 leaq 0xb8(%rsp), %rdi callq 0x357fe vmovsd 0x18(%r12), %xmm0 vmovsd %xmm0, 0x40(%rsp) callq 0xb5a0 movq %rax, (%r14) callq 0xb5a0 movq %rax, 0x8(%r14) leaq 0x130(%rsp), %rdi leaq 0x168(%rsp), %rsi leaq 0x68(%rsp), %rdx callq 0x293990 vmovsd 0x40(%rsp), %xmm0 movl $0x7fffffff, %ebp # imm = 0x7FFFFFFF leaq 0x68(%rsp), %rbx xorl %r15d, %r15d vmovsd %xmm0, 0xe8(%rsp) fldl 0xe8(%rsp) fldt 0x4c(%rsp) cmpq 0xb0(%r12), %r15 je 0x2991c9 movq 0x18(%r14), %rcx fstpt 0x20(%rsp) leaq 0x130(%rsp), %rdi leaq 0x168(%rsp), %rsi movq %rbx, %rdx fldt 0x94(%rsp) fstpt 0x10(%rsp) fld %st(0) fstpt 0x40(%rsp) fstpt (%rsp) callq 0x2939d2 movl %eax, %r13d testl %eax, %eax je 0x299270 cmpl %ebp, %r13d jge 0x299140 movq %r14, %rdi movq %rbx, %rsi movl %r13d, %edx movq %r15, %rcx callq 0x299722 movl %r13d, %ebp cmpq 0x88(%rsp), %r15 jle 0x299192 movl %r13d, 0xa8(%rsp) fildl 0xa8(%rsp) movl 0x1e0(%rsp), %eax movl %eax, 0xac(%rsp) fidivl 0xac(%rsp) fldt 0xc4(%rsp) fstpt 0x10(%rsp) fstpt (%rsp) callq 0xb110 fldt 0xd0(%rsp) fmulp %st, %st(1) fldt 0x40(%rsp) faddp %st, %st(1) jmp 0x299196 fldt 0x40(%rsp) fldt 0xdc(%rsp) fxch %st(1) fucomi %st(1), %st fstp %st(1) fldz ja 0x2991c9 fstp %st(0) movq %r14, %rdi fstpt 0x40(%rsp) callq 0x29975a fldt 0x40(%rsp) fldt 0x4c(%rsp) testb %al, %al jne 0x2991c9 incq %r15 jmp 0x2990da fstp %st(0) fstp %st(0) movq 0x58(%rsp), %rax movl $0x5, 0xa0(%rax) fldz fldz fstp %st(1) fstp %st(0) cmpl $0x0, 0x50(%r14) jne 0x2991f7 movq 0x58(%rsp), %rax andl $0x0, 0xa0(%rax) cmpq $0x0, 0x28(%r14) je 0x299228 movq 0x58(%rsp), %rax leaq 0x68(%rax), %rbx pushq $0x1 popq %rsi movq %rbx, %rdi callq 0x35292 movq (%rbx), %rsi addq $0x20, %r14 movl 0x90(%rsp), %edx movq %r14, %rdi callq 0x35781 leaq 0x130(%rsp), %rdi callq 0x393ac leaq 0x168(%rsp), %rdi callq 0x293ea6 leaq 0x110(%rsp), %rdi callq 0x968b6 leaq 0x70(%rsp), %rdi callq 0x357fe movq 0x58(%rsp), %rax addq $0x1e8, %rsp # imm = 0x1E8 popq %rbx popq %r12 popq %r13 popq %r14 popq %r15 popq %rbp retq vmovsd 0x80(%rsp), %xmm0 movq 0x78(%rsp), %rdi leaq 0x68(%rsp), %rbx movq %rbx, %rsi callq 0x9706e movq %r14, %rdi movq %rbx, %rsi movq %r15, %rdx callq 0x2996d8 movl $0x0, 0x64(%rsp) leaq 0x168(%rsp), %r15 leaq 0x68(%rsp), %rbp xorl %ebx, %ebx fldt 0x4c(%rsp) fldt 0x40(%rsp) cmpl 0xbc(%r12), %ebx jge 0x2991e0 movq 0x18(%r14), %rcx fldt 0x104(%rsp) fld %st(1) fstpt 0x40(%rsp) fmulp %st, %st(1) fldt 0xf8(%rsp) fstpt 0x30(%rsp) fxch %st(1) fstpt 0x20(%rsp) leaq 0x130(%rsp), %rdi movq %r15, %rsi movq %rbp, %rdx fldt 0x94(%rsp) fstpt 0x10(%rsp) fstpt (%rsp) callq 0x293c02 testl %eax, %eax jne 0x299340 vmovsd 0x80(%rsp), %xmm0 movq 0x78(%rsp), %rdi movq %rbp, %rsi callq 0x9706e movl 0xc0(%r12), %eax imull %ebx, %eax notl %eax movslq %eax, %rdx movq %r14, %rdi movq %rbp, %rsi callq 0x2996d8 movq %r14, %rdi callq 0x29975a fldt 0x4c(%rsp) fldt 0x40(%rsp) testb %al, %al jne 0x2991e0 movl %ebx, 0x88(%rsp) pushq $0x1 popq %rax movl %eax, %ebx leal -0x1(%rbx), %r13d cmpl 0xc0(%r12), %r13d jge 0x299486 movq 0x18(%r14), %rcx fxch %st(1) fstpt 0x20(%rsp) leaq 0x130(%rsp), %rdi movq %r15, %rsi movq %rbp, %rdx fldt 0x94(%rsp) fstpt 0x10(%rsp) fld %st(0) fstpt 0x40(%rsp) fstpt (%rsp) callq 0x2939d2 testl %eax, %eax je 0x299439 vcvtsi2sd %r13d, %xmm2, %xmm0 vucomisd 0xa0(%r12), %xmm0 jbe 0x299407 movl %eax, 0xa0(%rsp) fildl 0xa0(%rsp) movl 0x1e0(%rsp), %eax movl %eax, 0xa4(%rsp) fidivl 0xa4(%rsp) fldt 0xc4(%rsp) fstpt 0x10(%rsp) fstpt (%rsp) callq 0xb110 fldt 0xd0(%rsp) fmulp %st, %st(1) fldt 0x40(%rsp) faddp %st, %st(1) jmp 0x29940b fldt 0x40(%rsp) fldt 0xdc(%rsp) fxch %st(1) fucomi %st(1), %st fstp %st(1) ja 0x299476 movq %r14, %rdi fstpt 0x40(%rsp) callq 0x29975a testb %al, %al jne 0x29947e fldt 0x4c(%rsp) fldt 0x40(%rsp) incl %ebx jmp 0x299364 vmovsd 0x80(%rsp), %xmm0 movq 0x78(%rsp), %rdi movq %rbp, %rsi callq 0x9706e movl 0xc0(%r12), %eax imull 0x64(%rsp), %eax subl %ebx, %eax movslq %eax, %rdx movq %r14, %rdi movq %rbp, %rsi callq 0x2996d8 fldt 0x4c(%rsp) fldt 0x40(%rsp) jmp 0x299486 fldt 0x4c(%rsp) fxch %st(1) jmp 0x299486 fldt 0x4c(%rsp) fldt 0x40(%rsp) movl 0x88(%rsp), %ebx decl 0x64(%rsp) incl %ebx jmp 0x2992b8 jmp 0x2994ce jmp 0x2994a7 fstp %st(1) fstp %st(0) jmp 0x2994ce movq %rax, %rbx jmp 0x2994de movq %rax, %rbx jmp 0x2994eb movq %rax, %rbx jmp 0x2994f8 movq %rax, %rbx jmp 0x299502 jmp 0x2994ce jmp 0x2994ce jmp 0x2994ce leaq 0xb8(%rsp), %rdi movq %rax, %rbx callq 0x357fe jmp 0x2994d1 movq %rax, %rbx leaq 0x130(%rsp), %rdi callq 0x393ac leaq 0x168(%rsp), %rdi callq 0x293ea6 leaq 0x110(%rsp), %rdi callq 0x968b6 leaq 0x70(%rsp), %rdi callq 0x357fe movq 0x58(%rsp), %rdi callq 0xea60 movq %rbx, %rdi callq 0xb3d0
/quesnel[P]baryonyx/lib/src/itm-solver-common.hpp
baryonyx::result baryonyx::itm::optimize_problem<baryonyx::itm::solver_inequalities_01coeff<long double, baryonyx::itm::minimize_tag, baryonyx::itm::default_cost_type<long double>, true>, long double, baryonyx::itm::minimize_tag, baryonyx::itm::default_cost_type<long double>>(baryonyx::context const&, baryonyx::problem const&)
inline result optimize_problem(const context& ctx, const problem& pb) { result r; if (ctx.start) ctx.start(ctx.parameters); auto constraints{ make_merged_constraints(ctx, pb) }; if (constraints.empty() || pb.vars.values.empty()) { r.status = result_status::success; r.solutions.resize(1); r.solutions.back().value = pb.objective.value; r.strings = pb.strings; r.affected_vars = pb.affected_vars; r.variable_name = pb.vars.names; return r; } random_engine rng(init_random_generator_seed(ctx)); auto variables = length(pb.vars.values); auto cost = Cost(pb.objective, variables); auto cost_constant = pb.objective.value; const auto thread = get_thread_number(ctx); std::vector<optimize_functor<Solver, Float, Mode, Cost>> functors; std::vector<std::thread> pool(thread); best_solution_recorder<Cost, Float, Mode> best_recorder( rng, thread, cost, cost_constant, variables, constraints, ctx.parameters.init_population_size); auto seeds = generate_seed(rng, thread); std::atomic_bool stop_task; stop_task.store(false); for (unsigned i = 0u; i != thread; ++i) functors.emplace_back(ctx, i, seeds[i]); for (unsigned i = 0u; i != thread; ++i) pool[i] = std::thread(std::ref(functors[i]), std::ref(stop_task), std::ref(best_recorder), std::cref(constraints), variables, std::cref(cost), cost_constant); const auto start = std::chrono::steady_clock::now(); auto end = start; do { std::this_thread::sleep_for(std::chrono::seconds{ 1L }); if (ctx.update) { auto call_number = 0L; for (auto i = 0u; i != thread; ++i) call_number += functors[i].m_call_number; int constraints_remaining; long int loop; double value; double duration; best_recorder.get_best( constraints_remaining, value, duration, loop); ctx.update( constraints_remaining, value, loop, duration, call_number); } end = std::chrono::steady_clock::now(); } while (!is_time_limit(ctx.parameters.time_limit, start, end)); stop_task.store(true); for (auto& t : pool) t.join(); r.strings = pb.strings; r.affected_vars = pb.affected_vars; r.variable_name = pb.vars.names; r.variables = variables; r.constraints = length(constraints); const auto& first = best_recorder.get_best(0); if (!first.is_solution()) r.status = result_status::time_limit_reached; else r.status = result_status::success; r.duration = first.duration; r.loop = first.loop; r.remaining_constraints = first.remaining_constraints; switch (ctx.parameters.storage) { case solver_parameters::storage_type::one: { r.solutions.resize(1); convert(first, r.solutions[0], variables); } break; case solver_parameters::storage_type::bound: { r.solutions.resize(2); convert(first, r.solutions[0], variables); convert(best_recorder.get_worst(), r.solutions[1], variables); } break; case solver_parameters::storage_type::five: { r.solutions.resize(5); for (int i = 0; i != 5; ++i) convert(best_recorder.get_best(i), r.solutions[i], variables); } break; } best_recorder.show_population(ctx); if (ctx.finish) ctx.finish(r); return r; }
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x188, %rsp # imm = 0x188 vxorpd %xmm0, %xmm0, %xmm0 movq %rdx, %r13 movq %rsi, %r15 movq %rdi, %r12 vmovupd %zmm0, 0x58(%rdi) vmovupd %zmm0, 0x40(%rdi) vmovupd %zmm0, (%rdi) movq $0x7fffffff, 0x98(%rdi) # imm = 0x7FFFFFFF movl $0x2, 0xa0(%rdi) cmpq $0x0, 0x128(%rsi) je 0x29981f leaq 0x118(%r15), %rdi movq %r15, %rsi vzeroupper callq 0x3527e leaq 0x48(%rsp), %rdi movq %r15, %rsi movq %r13, %rdx vzeroupper callq 0x450078 movq 0x48(%rsp), %rax cmpq 0x50(%rsp), %rax je 0x29997b movq 0xa8(%r13), %rax cmpq 0xb0(%r13), %rax je 0x29997b movq %r15, %rdi callq 0x353ba movl $0x7fffffff, %ecx # imm = 0x7FFFFFFF xorl %edx, %edx divq %rcx cmpq $0x1, %rdx adcq $0x0, %rdx movq %rdx, 0x40(%rsp) movq 0xb0(%r13), %rax subq 0xa8(%r13), %rax pushq $0xc popq %rcx cqto leaq 0x10(%r13), %rsi idivq %rcx movl %eax, 0x4(%rsp) leaq 0x60(%rsp), %rdi movl %eax, %edx movq %r12, 0x10(%rsp) callq 0x8adb0 vmovsd 0x40(%r13), %xmm0 vmovsd %xmm0, 0x38(%rsp) movq %r15, %rdi callq 0x35447 andq $0x0, 0x80(%rsp) vxorpd %xmm0, %xmm0, %xmm0 movl %eax, %ebp vmovapd %xmm0, 0x70(%rsp) leaq 0xb0(%rsp), %rdi leaq 0xc8(%rsp), %rdx movq %rbp, %rsi callq 0x35462 vmovsd 0x38(%rsp), %xmm0 movl 0xc4(%r15), %eax movl 0x4(%rsp), %r8d subq $0x8, %rsp movl %ebp, %edx leaq 0xd0(%rsp), %rdi leaq 0x48(%rsp), %rsi leaq 0x68(%rsp), %rcx leaq 0x50(%rsp), %r9 pushq %rax callq 0xa17b4 popq %rax popq %rcx leaq 0x30(%rsp), %rdi leaq 0x40(%rsp), %rsi movl %ebp, %edx callq 0x3551c xorl %eax, %eax movb %al, 0x3(%rsp) xorl %ecx, %ecx leaq 0x70(%rsp), %rbx leaq 0x8(%rsp), %r14 xchgb %cl, 0x3(%rsp) leaq 0x90(%r13), %rcx movq %rcx, 0x98(%rsp) movl %eax, 0x8(%rsp) cmpl %ebp, %eax je 0x2999d6 movl %eax, %ecx shlq $0x3, %rcx addq 0x30(%rsp), %rcx movq %rbx, %rdi movq %r15, %rsi movq %r14, %rdx callq 0x29aaa4 movl 0x8(%rsp), %eax incl %eax jmp 0x299952 andl $0x0, 0xa0(%r12) leaq 0x68(%r12), %rdi pushq $0x1 popq %rsi callq 0x35292 vmovsd 0x40(%r13), %xmm0 movq 0x70(%r12), %rax vmovsd %xmm0, -0x8(%rax) movq %r12, %rdi movq %r13, %rsi callq 0x31d28 leaq 0xc0(%r13), %rsi leaq 0x20(%r12), %rdi callq 0x352c2 addq $0x90, %r13 leaq 0x50(%r12), %rdi movq %r13, %rsi callq 0x352ec jmp 0x299d6b movq %r13, 0x90(%rsp) movq %r15, %r13 leaq 0x8(%rsp), %r15 leaq 0xa8(%rsp), %r14 leaq 0x4(%rsp), %rbx xorl %r12d, %r12d cmpq %rbp, %r12 je 0x299a97 imulq $0xc0, %r12, %rax addq 0x70(%rsp), %rax leaq 0x3(%rsp), %r10 leaq 0xc8(%rsp), %r11 movq %r10, 0x20(%rsp) leaq 0x48(%rsp), %r10 movq %r11, 0x18(%rsp) movq %r10, 0xa8(%rsp) movq %rax, 0x28(%rsp) leaq 0x60(%rsp), %rax movq %rax, 0xa0(%rsp) leaq 0x20(%rsp), %rdx leaq 0x38(%rsp), %rax movq %r15, %rdi leaq 0x28(%rsp), %rsi leaq 0x18(%rsp), %rcx movq %r14, %r8 movq %rbx, %r9 pushq %rax leaq 0xa8(%rsp), %rax pushq %rax callq 0x29aaf0 popq %rax popq %rcx leaq (,%r12,8), %rdi addq 0xb0(%rsp), %rdi movq %r15, %rsi callq 0x356d6 movq %r15, %rdi callq 0x356f0 incq %r12 jmp 0x2999f6 callq 0xb5a0 imulq $0xc0, %rbp, %r15 leaq 0x138(%r13), %r12 leaq 0x8(%rsp), %rbp movq %rax, %rbx movq $0x1, 0x8(%rsp) movq %rbp, %rdi callq 0x356fd cmpq $0x0, 0x148(%r13) je 0x299b2c movq 0x70(%rsp), %rax xorl %ecx, %ecx xorl %r14d, %r14d cmpq %rcx, %r15 je 0x299aed addq 0xb0(%rax,%rcx), %r14 addq $0xc0, %rcx jmp 0x299ad7 leaq 0xd0(%rsp), %rdi leaq 0x18(%rsp), %rsi leaq 0x28(%rsp), %rdx leaq 0x20(%rsp), %rcx movq %rbp, %r8 callq 0xa90e8 vmovsd 0x28(%rsp), %xmm0 vmovsd 0x20(%rsp), %xmm1 movq 0x8(%rsp), %rdx movl 0x18(%rsp), %esi movq %r12, %rdi movq %r14, %rcx callq 0x35736 callq 0xb5a0 vmovsd (%r13), %xmm0 subq %rbx, %rax vxorpd %xmm1, %xmm1, %xmm1 vucomisd %xmm0, %xmm1 vcvtsi2sd %rax, %xmm2, %xmm1 vdivsd 0x1bff31(%rip), %xmm1, %xmm1 # 0x459a80 jae 0x299ab2 vucomisd %xmm0, %xmm1 jbe 0x299ab2 movb $0x1, %al xchgb %al, 0x3(%rsp) movq 0xb0(%rsp), %rbx movq 0xb8(%rsp), %r14 movq 0x10(%rsp), %r12 cmpq %r14, %rbx je 0x299b8d movq %rbx, %rdi callq 0xb1a0 addq $0x8, %rbx jmp 0x299b7a movq 0x90(%rsp), %rbx movq %r12, %rdi movq %rbx, %rsi callq 0x31d28 addq $0xc0, %rbx leaq 0x20(%r12), %rdi movq %rbx, %rsi callq 0x352c2 leaq 0x50(%r12), %rdi movq 0x98(%rsp), %rsi callq 0x352ec movl 0x4(%rsp), %eax movl %eax, 0x90(%r12) movq 0x50(%rsp), %rax subq 0x48(%rsp), %rax pushq $0x28 popq %rcx cqto xorl %esi, %esi idivq %rcx movq 0x108(%rsp), %rdi movq 0x120(%rsp), %rcx movl %eax, 0x94(%r12) movslq (%rdi), %rax imulq $0x38, %rax, %rax movl 0x30(%rcx,%rax), %edx leaq (%rcx,%rax), %rbx testl %edx, %edx setne %sil leal (%rsi,%rsi,2), %esi movl %esi, 0xa0(%r12) vmovsd 0x18(%rcx,%rax), %xmm0 vmovsd %xmm0, 0x80(%r12) movq 0x28(%rcx,%rax), %rax movq %rax, 0x88(%r12) movl %edx, 0x98(%r12) movl 0xe8(%r13), %eax testl %eax, %eax je 0x299ce8 cmpl $0x1, %eax je 0x299cb0 cmpl $0x2, %eax jne 0x299d07 leaq 0x68(%r12), %rbx pushq $0x5 popq %rsi movq %rbx, %rdi callq 0x35292 xorl %r14d, %r14d xorl %r15d, %r15d cmpq $0x5, %r15 je 0x299d07 movq 0x108(%rsp), %rax movq (%rbx), %rsi movl 0x4(%rsp), %edx movslq (%rax,%r15,4), %rax addq %r14, %rsi imulq $0x38, %rax, %rdi addq 0x120(%rsp), %rdi callq 0x4f485 incq %r15 addq $0x20, %r14 jmp 0x299c76 leaq 0x68(%r12), %r14 pushq $0x2 popq %rsi movq %r14, %rdi callq 0x35292 movl 0x4(%rsp), %edx movq (%r14), %rsi movq %rbx, %rdi callq 0x4f485 leaq 0xd0(%rsp), %rdi callq 0xa9166 movq (%r14), %rsi movq %rax, %rbx addq $0x20, %rsi jmp 0x299cfb leaq 0x68(%r12), %r14 pushq $0x1 popq %rsi movq %r14, %rdi callq 0x35292 movq (%r14), %rsi movl 0x4(%rsp), %edx movq %rbx, %rdi callq 0x4f485 leaq 0xc8(%rsp), %rdi movq %r13, %rsi callq 0xa1918 cmpq $0x0, 0x168(%r13) je 0x299d33 addq $0x158, %r13 # imm = 0x158 movq %r13, %rdi movq %r12, %rsi callq 0x357ea leaq 0x30(%rsp), %rdi callq 0x357fe leaq 0xc8(%rsp), %rdi callq 0xa192c leaq 0xb0(%rsp), %rdi callq 0x35836 leaq 0x70(%rsp), %rdi callq 0x2a033c leaq 0x68(%rsp), %rdi callq 0x8affa leaq 0x48(%rsp), %rdi callq 0x35858 movq %r12, %rax addq $0x188, %rsp # imm = 0x188 popq %rbx popq %r12 popq %r13 popq %r14 popq %r15 popq %rbp retq movq %rax, %rbx jmp 0x299dcc movq %rax, %rbx jmp 0x299dd9 movq %rax, %rbx jmp 0x299de6 jmp 0x299db0 jmp 0x299da1 jmp 0x299dbf jmp 0x299dbf movq %r12, 0x10(%rsp) movq %rax, %rbx jmp 0x299e04 movq %r12, 0x10(%rsp) movq %rax, %rbx jmp 0x299dfa jmp 0x299dbf jmp 0x299dbf jmp 0x299dbf jmp 0x299dbf jmp 0x299dbf movq %rax, %rbx leaq 0x30(%rsp), %rdi callq 0x357fe leaq 0xc8(%rsp), %rdi callq 0xa192c leaq 0xb0(%rsp), %rdi callq 0x35836 leaq 0x70(%rsp), %rdi callq 0x2a033c leaq 0x68(%rsp), %rdi callq 0x8affa leaq 0x48(%rsp), %rdi callq 0x35858 movq 0x10(%rsp), %rdi callq 0xea60 movq %rbx, %rdi callq 0xb3d0
/quesnel[P]baryonyx/lib/src/itm-optimizer-common.hpp
baryonyx::result baryonyx::itm::optimize_problem<baryonyx::itm::solver_inequalities_01coeff<long double, baryonyx::itm::minimize_tag, baryonyx::itm::default_cost_type<long double>, false>, long double, baryonyx::itm::minimize_tag, baryonyx::itm::default_cost_type<long double>>(baryonyx::context const&, baryonyx::problem const&)
inline result optimize_problem(const context& ctx, const problem& pb) { result r; if (ctx.start) ctx.start(ctx.parameters); auto constraints{ make_merged_constraints(ctx, pb) }; if (constraints.empty() || pb.vars.values.empty()) { r.status = result_status::success; r.solutions.resize(1); r.solutions.back().value = pb.objective.value; r.strings = pb.strings; r.affected_vars = pb.affected_vars; r.variable_name = pb.vars.names; return r; } random_engine rng(init_random_generator_seed(ctx)); auto variables = length(pb.vars.values); auto cost = Cost(pb.objective, variables); auto cost_constant = pb.objective.value; const auto thread = get_thread_number(ctx); std::vector<optimize_functor<Solver, Float, Mode, Cost>> functors; std::vector<std::thread> pool(thread); best_solution_recorder<Cost, Float, Mode> best_recorder( rng, thread, cost, cost_constant, variables, constraints, ctx.parameters.init_population_size); auto seeds = generate_seed(rng, thread); std::atomic_bool stop_task; stop_task.store(false); for (unsigned i = 0u; i != thread; ++i) functors.emplace_back(ctx, i, seeds[i]); for (unsigned i = 0u; i != thread; ++i) pool[i] = std::thread(std::ref(functors[i]), std::ref(stop_task), std::ref(best_recorder), std::cref(constraints), variables, std::cref(cost), cost_constant); const auto start = std::chrono::steady_clock::now(); auto end = start; do { std::this_thread::sleep_for(std::chrono::seconds{ 1L }); if (ctx.update) { auto call_number = 0L; for (auto i = 0u; i != thread; ++i) call_number += functors[i].m_call_number; int constraints_remaining; long int loop; double value; double duration; best_recorder.get_best( constraints_remaining, value, duration, loop); ctx.update( constraints_remaining, value, loop, duration, call_number); } end = std::chrono::steady_clock::now(); } while (!is_time_limit(ctx.parameters.time_limit, start, end)); stop_task.store(true); for (auto& t : pool) t.join(); r.strings = pb.strings; r.affected_vars = pb.affected_vars; r.variable_name = pb.vars.names; r.variables = variables; r.constraints = length(constraints); const auto& first = best_recorder.get_best(0); if (!first.is_solution()) r.status = result_status::time_limit_reached; else r.status = result_status::success; r.duration = first.duration; r.loop = first.loop; r.remaining_constraints = first.remaining_constraints; switch (ctx.parameters.storage) { case solver_parameters::storage_type::one: { r.solutions.resize(1); convert(first, r.solutions[0], variables); } break; case solver_parameters::storage_type::bound: { r.solutions.resize(2); convert(first, r.solutions[0], variables); convert(best_recorder.get_worst(), r.solutions[1], variables); } break; case solver_parameters::storage_type::five: { r.solutions.resize(5); for (int i = 0; i != 5; ++i) convert(best_recorder.get_best(i), r.solutions[i], variables); } break; } best_recorder.show_population(ctx); if (ctx.finish) ctx.finish(r); return r; }
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x188, %rsp # imm = 0x188 vxorpd %xmm0, %xmm0, %xmm0 movq %rdx, %r13 movq %rsi, %r15 movq %rdi, %r12 vmovupd %zmm0, 0x58(%rdi) vmovupd %zmm0, 0x40(%rdi) vmovupd %zmm0, (%rdi) movq $0x7fffffff, 0x98(%rdi) # imm = 0x7FFFFFFF movl $0x2, 0xa0(%rdi) cmpq $0x0, 0x128(%rsi) je 0x29a194 leaq 0x118(%r15), %rdi movq %r15, %rsi vzeroupper callq 0x3527e leaq 0x48(%rsp), %rdi movq %r15, %rsi movq %r13, %rdx vzeroupper callq 0x450078 movq 0x48(%rsp), %rax cmpq 0x50(%rsp), %rax je 0x29a2f0 movq 0xa8(%r13), %rax cmpq 0xb0(%r13), %rax je 0x29a2f0 movq %r15, %rdi callq 0x353ba movl $0x7fffffff, %ecx # imm = 0x7FFFFFFF xorl %edx, %edx divq %rcx cmpq $0x1, %rdx adcq $0x0, %rdx movq %rdx, 0x40(%rsp) movq 0xb0(%r13), %rax subq 0xa8(%r13), %rax pushq $0xc popq %rcx cqto leaq 0x10(%r13), %rsi idivq %rcx movl %eax, 0x4(%rsp) leaq 0x60(%rsp), %rdi movl %eax, %edx movq %r12, 0x10(%rsp) callq 0x8adb0 vmovsd 0x40(%r13), %xmm0 vmovsd %xmm0, 0x38(%rsp) movq %r15, %rdi callq 0x35447 andq $0x0, 0x80(%rsp) vxorpd %xmm0, %xmm0, %xmm0 movl %eax, %ebp vmovapd %xmm0, 0x70(%rsp) leaq 0xb0(%rsp), %rdi leaq 0xc8(%rsp), %rdx movq %rbp, %rsi callq 0x35462 vmovsd 0x38(%rsp), %xmm0 movl 0xc4(%r15), %eax movl 0x4(%rsp), %r8d subq $0x8, %rsp movl %ebp, %edx leaq 0xd0(%rsp), %rdi leaq 0x48(%rsp), %rsi leaq 0x68(%rsp), %rcx leaq 0x50(%rsp), %r9 pushq %rax callq 0xa17b4 popq %rax popq %rcx leaq 0x30(%rsp), %rdi leaq 0x40(%rsp), %rsi movl %ebp, %edx callq 0x3551c xorl %eax, %eax movb %al, 0x3(%rsp) xorl %ecx, %ecx leaq 0x70(%rsp), %rbx leaq 0x8(%rsp), %r14 xchgb %cl, 0x3(%rsp) leaq 0x90(%r13), %rcx movq %rcx, 0x98(%rsp) movl %eax, 0x8(%rsp) cmpl %ebp, %eax je 0x29a34b movl %eax, %ecx shlq $0x3, %rcx addq 0x30(%rsp), %rcx movq %rbx, %rdi movq %r15, %rsi movq %r14, %rdx callq 0x2a1b74 movl 0x8(%rsp), %eax incl %eax jmp 0x29a2c7 andl $0x0, 0xa0(%r12) leaq 0x68(%r12), %rdi pushq $0x1 popq %rsi callq 0x35292 vmovsd 0x40(%r13), %xmm0 movq 0x70(%r12), %rax vmovsd %xmm0, -0x8(%rax) movq %r12, %rdi movq %r13, %rsi callq 0x31d28 leaq 0xc0(%r13), %rsi leaq 0x20(%r12), %rdi callq 0x352c2 addq $0x90, %r13 leaq 0x50(%r12), %rdi movq %r13, %rsi callq 0x352ec jmp 0x29a6e0 movq %r13, 0x90(%rsp) movq %r15, %r13 leaq 0x8(%rsp), %r15 leaq 0xa8(%rsp), %r14 leaq 0x4(%rsp), %rbx xorl %r12d, %r12d cmpq %rbp, %r12 je 0x29a40c imulq $0xc0, %r12, %rax addq 0x70(%rsp), %rax leaq 0x3(%rsp), %r10 leaq 0xc8(%rsp), %r11 movq %r10, 0x20(%rsp) leaq 0x48(%rsp), %r10 movq %r11, 0x18(%rsp) movq %r10, 0xa8(%rsp) movq %rax, 0x28(%rsp) leaq 0x60(%rsp), %rax movq %rax, 0xa0(%rsp) leaq 0x20(%rsp), %rdx leaq 0x38(%rsp), %rax movq %r15, %rdi leaq 0x28(%rsp), %rsi leaq 0x18(%rsp), %rcx movq %r14, %r8 movq %rbx, %r9 pushq %rax leaq 0xa8(%rsp), %rax pushq %rax callq 0x2a1bc0 popq %rax popq %rcx leaq (,%r12,8), %rdi addq 0xb0(%rsp), %rdi movq %r15, %rsi callq 0x356d6 movq %r15, %rdi callq 0x356f0 incq %r12 jmp 0x29a36b callq 0xb5a0 imulq $0xc0, %rbp, %r15 leaq 0x138(%r13), %r12 leaq 0x8(%rsp), %rbp movq %rax, %rbx movq $0x1, 0x8(%rsp) movq %rbp, %rdi callq 0x356fd cmpq $0x0, 0x148(%r13) je 0x29a4a1 movq 0x70(%rsp), %rax xorl %ecx, %ecx xorl %r14d, %r14d cmpq %rcx, %r15 je 0x29a462 addq 0xb0(%rax,%rcx), %r14 addq $0xc0, %rcx jmp 0x29a44c leaq 0xd0(%rsp), %rdi leaq 0x18(%rsp), %rsi leaq 0x28(%rsp), %rdx leaq 0x20(%rsp), %rcx movq %rbp, %r8 callq 0xa90e8 vmovsd 0x28(%rsp), %xmm0 vmovsd 0x20(%rsp), %xmm1 movq 0x8(%rsp), %rdx movl 0x18(%rsp), %esi movq %r12, %rdi movq %r14, %rcx callq 0x35736 callq 0xb5a0 vmovsd (%r13), %xmm0 subq %rbx, %rax vxorpd %xmm1, %xmm1, %xmm1 vucomisd %xmm0, %xmm1 vcvtsi2sd %rax, %xmm2, %xmm1 vdivsd 0x1bf5bc(%rip), %xmm1, %xmm1 # 0x459a80 jae 0x29a427 vucomisd %xmm0, %xmm1 jbe 0x29a427 movb $0x1, %al xchgb %al, 0x3(%rsp) movq 0xb0(%rsp), %rbx movq 0xb8(%rsp), %r14 movq 0x10(%rsp), %r12 cmpq %r14, %rbx je 0x29a502 movq %rbx, %rdi callq 0xb1a0 addq $0x8, %rbx jmp 0x29a4ef movq 0x90(%rsp), %rbx movq %r12, %rdi movq %rbx, %rsi callq 0x31d28 addq $0xc0, %rbx leaq 0x20(%r12), %rdi movq %rbx, %rsi callq 0x352c2 leaq 0x50(%r12), %rdi movq 0x98(%rsp), %rsi callq 0x352ec movl 0x4(%rsp), %eax movl %eax, 0x90(%r12) movq 0x50(%rsp), %rax subq 0x48(%rsp), %rax pushq $0x28 popq %rcx cqto xorl %esi, %esi idivq %rcx movq 0x108(%rsp), %rdi movq 0x120(%rsp), %rcx movl %eax, 0x94(%r12) movslq (%rdi), %rax imulq $0x38, %rax, %rax movl 0x30(%rcx,%rax), %edx leaq (%rcx,%rax), %rbx testl %edx, %edx setne %sil leal (%rsi,%rsi,2), %esi movl %esi, 0xa0(%r12) vmovsd 0x18(%rcx,%rax), %xmm0 vmovsd %xmm0, 0x80(%r12) movq 0x28(%rcx,%rax), %rax movq %rax, 0x88(%r12) movl %edx, 0x98(%r12) movl 0xe8(%r13), %eax testl %eax, %eax je 0x29a65d cmpl $0x1, %eax je 0x29a625 cmpl $0x2, %eax jne 0x29a67c leaq 0x68(%r12), %rbx pushq $0x5 popq %rsi movq %rbx, %rdi callq 0x35292 xorl %r14d, %r14d xorl %r15d, %r15d cmpq $0x5, %r15 je 0x29a67c movq 0x108(%rsp), %rax movq (%rbx), %rsi movl 0x4(%rsp), %edx movslq (%rax,%r15,4), %rax addq %r14, %rsi imulq $0x38, %rax, %rdi addq 0x120(%rsp), %rdi callq 0x4f485 incq %r15 addq $0x20, %r14 jmp 0x29a5eb leaq 0x68(%r12), %r14 pushq $0x2 popq %rsi movq %r14, %rdi callq 0x35292 movl 0x4(%rsp), %edx movq (%r14), %rsi movq %rbx, %rdi callq 0x4f485 leaq 0xd0(%rsp), %rdi callq 0xa9166 movq (%r14), %rsi movq %rax, %rbx addq $0x20, %rsi jmp 0x29a670 leaq 0x68(%r12), %r14 pushq $0x1 popq %rsi movq %r14, %rdi callq 0x35292 movq (%r14), %rsi movl 0x4(%rsp), %edx movq %rbx, %rdi callq 0x4f485 leaq 0xc8(%rsp), %rdi movq %r13, %rsi callq 0xa1918 cmpq $0x0, 0x168(%r13) je 0x29a6a8 addq $0x158, %r13 # imm = 0x158 movq %r13, %rdi movq %r12, %rsi callq 0x357ea leaq 0x30(%rsp), %rdi callq 0x357fe leaq 0xc8(%rsp), %rdi callq 0xa192c leaq 0xb0(%rsp), %rdi callq 0x35836 leaq 0x70(%rsp), %rdi callq 0x2a6c7c leaq 0x68(%rsp), %rdi callq 0x8affa leaq 0x48(%rsp), %rdi callq 0x35858 movq %r12, %rax addq $0x188, %rsp # imm = 0x188 popq %rbx popq %r12 popq %r13 popq %r14 popq %r15 popq %rbp retq movq %rax, %rbx jmp 0x29a741 movq %rax, %rbx jmp 0x29a74e movq %rax, %rbx jmp 0x29a75b jmp 0x29a725 jmp 0x29a716 jmp 0x29a734 jmp 0x29a734 movq %r12, 0x10(%rsp) movq %rax, %rbx jmp 0x29a779 movq %r12, 0x10(%rsp) movq %rax, %rbx jmp 0x29a76f jmp 0x29a734 jmp 0x29a734 jmp 0x29a734 jmp 0x29a734 jmp 0x29a734 movq %rax, %rbx leaq 0x30(%rsp), %rdi callq 0x357fe leaq 0xc8(%rsp), %rdi callq 0xa192c leaq 0xb0(%rsp), %rdi callq 0x35836 leaq 0x70(%rsp), %rdi callq 0x2a6c7c leaq 0x68(%rsp), %rdi callq 0x8affa leaq 0x48(%rsp), %rdi callq 0x35858 movq 0x10(%rsp), %rdi callq 0xea60 movq %rbx, %rdi callq 0xb3d0
/quesnel[P]baryonyx/lib/src/itm-optimizer-common.hpp
bool baryonyx::itm::solver_inequalities_01coeff<long double, baryonyx::itm::minimize_tag, baryonyx::itm::default_cost_type<long double>, true>::compute_update_row<baryonyx::bit_array, std::reverse_iterator<__gnu_cxx::__normal_iterator<int const*, std::vector<int, std::allocator<int>>>>>(baryonyx::bit_array&, std::reverse_iterator<__gnu_cxx::__normal_iterator<int const*, std::vector<int, std::allocator<int>>>>, std::reverse_iterator<__gnu_cxx::__normal_iterator<int const*, std::vector<int, std::allocator<int>>>>, long double, long double, long double)
bool compute_update_row(Xtype& x, Iterator first, Iterator last, Float kappa, Float delta, Float theta) { auto at_least_one_pi_changed{ false }; logger::log("update-row {} {} {}\n", kappa, delta, theta); for (; first != last; ++first) { auto k = constraint(first); const auto it = ap.row(k); decrease_preference(std::get<0>(it), std::get<1>(it), theta); const auto r_size = compute_reduced_costs(std::get<0>(it), std::get<1>(it), x); calculator_sort<Mode>(R.get(), R.get() + r_size, rng); int selected = select_variables(r_size, b[k].min, b[k].max); logger::log("constraints {}: {} <= ", k, b[k].min); for (int i = 0; i < r_size; ++i) logger::log("({} {}) ", R[i].value, R[i].id); logger::log("<= {} => Selected: {}\n", b[k].max, selected); auto pi_change = affect( *this, x, std::get<0>(it), k, selected, r_size, kappa, delta); at_least_one_pi_changed = at_least_one_pi_changed || pi_change; } return at_least_one_pi_changed; }
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0xa8, %rsp movq %rcx, 0x60(%rsp) movq %rdx, %rbx leaq 0x1c050e(%rip), %rdx # 0x45c344 movq %rsi, %r14 movq %rdi, %r12 fldt 0x100(%rsp) fldt 0xf0(%rsp) fldt 0xe0(%rsp) fstpt 0x90(%rsp) fstpt 0x80(%rsp) fstpt 0x70(%rsp) pushq $0x14 popq %rsi leaq 0x90(%rsp), %rcx leaq 0x80(%rsp), %r8 leaq 0x70(%rsp), %r9 callq 0x126c32 leaq 0x10(%r12), %rax movq %rbx, 0x58(%rsp) xorl %edx, %edx movq %rax, 0x48(%rsp) movq (%rbx), %rax pushq $0x16 popq %r13 leaq 0x1c789f(%rip), %rbp # 0x46373e pushq $0x8 popq %rbx movq %r14, 0x50(%rsp) movq 0x60(%rsp), %rcx cmpq (%rcx), %rax je 0x29c01a movq %rdx, 0x68(%rsp) movl -0x4(%rax), %edx movq 0x48(%rsp), %rsi leaq 0x38(%rsp), %rdi movl %edx, 0x2c(%rsp) callq 0x3de66 movq 0x38(%rsp), %rdx movq 0x40(%rsp), %rsi fldt 0x70(%rsp) movq %r12, %rdi fstpt (%rsp) callq 0x29c746 movq 0x38(%rsp), %rdx movq 0x40(%rsp), %rsi movq %r12, %rdi movq %r14, %rcx callq 0x29c76c movq 0x58(%r12), %rdi movq 0x8(%r12), %rdx movslq %eax, %r14 movl %eax, 0x34(%rsp) movq %r14, %rsi shlq $0x5, %rsi addq %rdi, %rsi callq 0x29c832 movslq 0x2c(%rsp), %rax movq 0x60(%r12), %rcx movq %r12, %rdi movl %r14d, %esi movl (%rcx,%rax,8), %edx movl 0x4(%rcx,%rax,8), %ecx callq 0x29c866 movslq 0x2c(%rsp), %r8 movl %eax, 0x30(%rsp) leaq 0x1c50fb(%rip), %rdx # 0x461045 leaq 0x2c(%rsp), %rcx movq %r12, %rdi movq %r13, %r15 movq %r13, %rsi shlq $0x3, %r8 addq 0x60(%r12), %r8 callq 0xb91c2 movl %r14d, %eax sarl $0x1f, %eax xorl %r13d, %r13d andnl %r14d, %eax, %r14d shlq $0x5, %r14 cmpq %r13, %r14 je 0x29bf9e movq 0x58(%r12), %rcx movq %r12, %rdi movq %rbx, %rsi movq %rbp, %rdx leaq 0x10(%rcx,%r13), %r8 addq %r13, %rcx callq 0x126d84 addq $0x20, %r13 jmp 0x29bf78 movslq 0x2c(%rsp), %rax movq 0x60(%r12), %rcx leaq 0x1c50ad(%rip), %rdx # 0x46105c leaq 0x30(%rsp), %r8 movq %r12, %rdi movq %r15, %r13 movq %r15, %rsi leaq 0x4(%rcx,%rax,8), %rcx callq 0xb91c2 movq 0x40(%rsp), %rdx movl 0x2c(%rsp), %ecx movl 0x30(%rsp), %r8d fldt 0x90(%rsp) fldt 0x80(%rsp) fstpt 0x10(%rsp) fstpt (%rsp) movq %r12, %rdi movq 0x50(%rsp), %r14 movl 0x34(%rsp), %r9d movq %r14, %rsi callq 0x29c8d7 movq 0x68(%rsp), %rdx movq 0x58(%rsp), %rcx orb %al, %dl movq (%rcx), %rax addq $-0x4, %rax movq %rax, (%rcx) jmp 0x29bea7 andb $0x1, %dl movl %edx, %eax addq $0xa8, %rsp popq %rbx popq %r12 popq %r13 popq %r14 popq %r15 popq %rbp retq nop
/quesnel[P]baryonyx/lib/src/itm-solver-inequalities-01.cpp
bool baryonyx::itm::solver_inequalities_01coeff<long double, baryonyx::itm::minimize_tag, baryonyx::itm::default_cost_type<long double>, true>::compute_update_row<baryonyx::bit_array, __gnu_cxx::__normal_iterator<int*, std::vector<int, std::allocator<int>>>>(baryonyx::bit_array&, __gnu_cxx::__normal_iterator<int*, std::vector<int, std::allocator<int>>>, __gnu_cxx::__normal_iterator<int*, std::vector<int, std::allocator<int>>>, long double, long double, long double)
bool compute_update_row(Xtype& x, Iterator first, Iterator last, Float kappa, Float delta, Float theta) { auto at_least_one_pi_changed{ false }; logger::log("update-row {} {} {}\n", kappa, delta, theta); for (; first != last; ++first) { auto k = constraint(first); const auto it = ap.row(k); decrease_preference(std::get<0>(it), std::get<1>(it), theta); const auto r_size = compute_reduced_costs(std::get<0>(it), std::get<1>(it), x); calculator_sort<Mode>(R.get(), R.get() + r_size, rng); int selected = select_variables(r_size, b[k].min, b[k].max); logger::log("constraints {}: {} <= ", k, b[k].min); for (int i = 0; i < r_size; ++i) logger::log("({} {}) ", R[i].value, R[i].id); logger::log("<= {} => Selected: {}\n", b[k].max, selected); auto pi_change = affect( *this, x, std::get<0>(it), k, selected, r_size, kappa, delta); at_least_one_pi_changed = at_least_one_pi_changed || pi_change; } return at_least_one_pi_changed; }
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x98, %rsp movq %rcx, 0x48(%rsp) movq %rdx, %r14 movq %rsi, %r13 movq %rdi, %r12 leaq 0x1c02ec(%rip), %rdx # 0x45c344 fldt 0xf0(%rsp) fldt 0xe0(%rsp) fldt 0xd0(%rsp) fstpt 0x80(%rsp) fstpt 0x70(%rsp) fstpt 0x60(%rsp) pushq $0x14 popq %rsi leaq 0x80(%rsp), %rcx leaq 0x70(%rsp), %r8 leaq 0x60(%rsp), %r9 callq 0x126c32 leaq 0x10(%r12), %rax leaq 0x1c769c(%rip), %rbp # 0x46373e movq %rax, 0x38(%rsp) xorl %eax, %eax pushq $0x8 popq %rbx movq %r13, 0x40(%rsp) cmpq 0x48(%rsp), %r14 je 0x29c217 movl (%r14), %edx movq 0x38(%rsp), %rsi leaq 0x28(%rsp), %rdi movq %rax, 0x50(%rsp) movq %r14, 0x58(%rsp) movl %edx, 0x20(%rsp) callq 0x3de66 movq 0x28(%rsp), %rdx movq 0x30(%rsp), %rsi fldt 0x60(%rsp) movq %r12, %rdi fstpt (%rsp) callq 0x29c746 movq 0x28(%rsp), %rdx movq 0x30(%rsp), %rsi movq %r12, %rdi movq %r13, %rcx callq 0x29c76c movq 0x58(%r12), %rdi movq 0x8(%r12), %rdx movslq %eax, %r13 movl %eax, %r15d movq %r13, %rsi shlq $0x5, %rsi addq %rdi, %rsi callq 0x29c832 movslq 0x20(%rsp), %rax movq 0x60(%r12), %rcx movq %r12, %rdi movl %r13d, %esi movl (%rcx,%rax,8), %edx movl 0x4(%rcx,%rax,8), %ecx callq 0x29c866 movslq 0x20(%rsp), %r8 movl %eax, 0x24(%rsp) movq %r12, %rdi shlq $0x3, %r8 addq 0x60(%r12), %r8 pushq $0x16 popq %rsi leaq 0x1c4ee1(%rip), %rdx # 0x461045 leaq 0x20(%rsp), %rcx callq 0xb91c2 movl %r13d, %eax sarl $0x1f, %eax xorl %r14d, %r14d andnl %r13d, %eax, %r13d shlq $0x5, %r13 cmpq %r14, %r13 je 0x29c1a6 movq 0x58(%r12), %rcx movq %r12, %rdi movq %rbx, %rsi movq %rbp, %rdx leaq 0x10(%rcx,%r14), %r8 addq %r14, %rcx callq 0x126d84 addq $0x20, %r14 jmp 0x29c180 movslq 0x20(%rsp), %rax movq 0x60(%r12), %rcx movq %r12, %rdi leaq 0x4(%rcx,%rax,8), %rcx pushq $0x16 popq %rsi leaq 0x1c4e9a(%rip), %rdx # 0x46105c leaq 0x24(%rsp), %r8 callq 0xb91c2 movq 0x30(%rsp), %rdx movl 0x20(%rsp), %ecx movl 0x24(%rsp), %r8d fldt 0x80(%rsp) fldt 0x70(%rsp) fstpt 0x10(%rsp) fstpt (%rsp) movq %r12, %rdi movl %r15d, %r9d movq 0x40(%rsp), %r13 movq %r13, %rsi callq 0x29c8d7 movq 0x50(%rsp), %rcx movq 0x58(%rsp), %r14 orb %al, %cl addq $0x4, %r14 movq %rcx, %rax jmp 0x29c0b1 andb $0x1, %al addq $0x98, %rsp popq %rbx popq %r12 popq %r13 popq %r14 popq %r15 popq %rbp retq nop
/quesnel[P]baryonyx/lib/src/itm-solver-inequalities-01.cpp
void baryonyx::itm::random_shuffle_unique<baryonyx::itm::solver_inequalities_01coeff<long double, baryonyx::itm::minimize_tag, baryonyx::itm::default_cost_type<long double>, false>::rc_data*>(baryonyx::itm::solver_inequalities_01coeff<long double, baryonyx::itm::minimize_tag, baryonyx::itm::default_cost_type<long double>, false>::rc_data*, baryonyx::itm::solver_inequalities_01coeff<long double, baryonyx::itm::minimize_tag, baryonyx::itm::default_cost_type<long double>, false>::rc_data*, std::linear_congruential_engine<unsigned long, 16807ul, 0ul, 2147483647ul>&)
inline void random_shuffle_unique(iteratorT begin, iteratorT end, random_engine& rng) noexcept { auto ret = begin++; for (; begin != end; ++begin) { if (ret->value != begin->value) { std::shuffle(ret, begin, rng); ret = begin; } } std::shuffle(ret, begin, rng); }
pushq %r15 pushq %r14 pushq %rbx movq %rdx, %rbx movq %rsi, %r14 leaq 0x20(%rdi), %r15 cmpq %r14, %r15 je 0x2a3852 fldt (%rdi) fldt (%r15) addq $0x20, %r15 fxch %st(1) fucompi %st(1), %st fstp %st(0) jne 0x2a383e jnp 0x2a3826 addq $-0x20, %r15 movq %r15, %rsi movq %rbx, %rdx callq 0x2a3dd3 movq %r15, %rdi jmp 0x2a3822 movq %r15, %rsi movq %rbx, %rdx callq 0x2a3dd3 popq %rbx popq %r14 popq %r15 retq jmp 0x2a3865 movq %rax, %rdi callq 0xeb9f
/quesnel[P]baryonyx/lib/src/itm-common.hpp
baryonyx::itm::solver_functor<baryonyx::itm::solver_inequalities_01coeff<long double, baryonyx::itm::minimize_tag, baryonyx::itm::default_cost_type<long double>, false>, long double, baryonyx::itm::minimize_tag, baryonyx::itm::default_cost_type<long double>, baryonyx::pnm_observer>::is_timelimit_reached()
bool is_timelimit_reached() { m_end = std::chrono::steady_clock::now(); return is_time_limit(m_ctx.parameters.time_limit, m_begin, m_end); }
pushq %rbx movq %rdi, %rbx callq 0xb5a0 movq %rax, 0x8(%rbx) vxorpd %xmm2, %xmm2, %xmm2 movq 0x10(%rbx), %rcx subq (%rbx), %rax vmovsd (%rcx), %xmm0 vcvtsi2sd %rax, %xmm1, %xmm1 vdivsd 0x1b17cf(%rip), %xmm1, %xmm1 # 0x459a80 vcmpnlepd %xmm2, %xmm0, %k1 vcmpltpd %xmm1, %xmm0, %k0 {%k1} kmovd %k0, %eax andb $0x1, %al popq %rbx retq nop
/quesnel[P]baryonyx/lib/src/itm-solver-common.hpp
bool baryonyx::itm::solver_inequalities_01coeff<long double, baryonyx::itm::minimize_tag, baryonyx::itm::quadratic_cost_type<long double>, true>::compute_update_row<baryonyx::bit_array, __gnu_cxx::__normal_iterator<int*, std::vector<int, std::allocator<int>>>>(baryonyx::bit_array&, __gnu_cxx::__normal_iterator<int*, std::vector<int, std::allocator<int>>>, __gnu_cxx::__normal_iterator<int*, std::vector<int, std::allocator<int>>>, long double, long double, long double)
bool compute_update_row(Xtype& x, Iterator first, Iterator last, Float kappa, Float delta, Float theta) { auto at_least_one_pi_changed{ false }; logger::log("update-row {} {} {}\n", kappa, delta, theta); for (; first != last; ++first) { auto k = constraint(first); const auto it = ap.row(k); decrease_preference(std::get<0>(it), std::get<1>(it), theta); const auto r_size = compute_reduced_costs(std::get<0>(it), std::get<1>(it), x); calculator_sort<Mode>(R.get(), R.get() + r_size, rng); int selected = select_variables(r_size, b[k].min, b[k].max); logger::log("constraints {}: {} <= ", k, b[k].min); for (int i = 0; i < r_size; ++i) logger::log("({} {}) ", R[i].value, R[i].id); logger::log("<= {} => Selected: {}\n", b[k].max, selected); auto pi_change = affect( *this, x, std::get<0>(it), k, selected, r_size, kappa, delta); at_least_one_pi_changed = at_least_one_pi_changed || pi_change; } return at_least_one_pi_changed; }
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x98, %rsp movq %rcx, 0x48(%rsp) movq %rdx, %r14 movq %rsi, %r13 movq %rdi, %r12 leaq 0x1b15da(%rip), %rdx # 0x45c344 fldt 0xf0(%rsp) fldt 0xe0(%rsp) fldt 0xd0(%rsp) fstpt 0x80(%rsp) fstpt 0x70(%rsp) fstpt 0x60(%rsp) pushq $0x14 popq %rsi leaq 0x80(%rsp), %rcx leaq 0x70(%rsp), %r8 leaq 0x60(%rsp), %r9 callq 0x126c32 leaq 0x10(%r12), %rax leaq 0x1b898a(%rip), %rbp # 0x46373e movq %rax, 0x38(%rsp) xorl %eax, %eax pushq $0x8 popq %rbx movq %r13, 0x40(%rsp) cmpq 0x48(%rsp), %r14 je 0x2aaf29 movl (%r14), %edx movq 0x38(%rsp), %rsi leaq 0x28(%rsp), %rdi movq %rax, 0x50(%rsp) movq %r14, 0x58(%rsp) movl %edx, 0x20(%rsp) callq 0x3de66 movq 0x28(%rsp), %rdx movq 0x30(%rsp), %rsi fldt 0x60(%rsp) movq %r12, %rdi fstpt (%rsp) callq 0x2ab458 movq 0x28(%rsp), %rdx movq 0x30(%rsp), %rsi movq %r12, %rdi movq %r13, %rcx callq 0x2ab47e movq 0x58(%r12), %rdi movq 0x8(%r12), %rdx movslq %eax, %r13 movl %eax, %r15d movq %r13, %rsi shlq $0x5, %rsi addq %rdi, %rsi callq 0x2ab554 movslq 0x20(%rsp), %rax movq 0x60(%r12), %rcx movq %r12, %rdi movl %r13d, %esi movl (%rcx,%rax,8), %edx movl 0x4(%rcx,%rax,8), %ecx callq 0x2ab588 movslq 0x20(%rsp), %r8 movl %eax, 0x24(%rsp) movq %r12, %rdi shlq $0x3, %r8 addq 0x60(%r12), %r8 pushq $0x16 popq %rsi leaq 0x1b61cf(%rip), %rdx # 0x461045 leaq 0x20(%rsp), %rcx callq 0xb91c2 movl %r13d, %eax sarl $0x1f, %eax xorl %r14d, %r14d andnl %r13d, %eax, %r13d shlq $0x5, %r13 cmpq %r14, %r13 je 0x2aaeb8 movq 0x58(%r12), %rcx movq %r12, %rdi movq %rbx, %rsi movq %rbp, %rdx leaq 0x10(%rcx,%r14), %r8 addq %r14, %rcx callq 0x126d84 addq $0x20, %r14 jmp 0x2aae92 movslq 0x20(%rsp), %rax movq 0x60(%r12), %rcx movq %r12, %rdi leaq 0x4(%rcx,%rax,8), %rcx pushq $0x16 popq %rsi leaq 0x1b6188(%rip), %rdx # 0x46105c leaq 0x24(%rsp), %r8 callq 0xb91c2 movq 0x30(%rsp), %rdx movl 0x20(%rsp), %ecx movl 0x24(%rsp), %r8d fldt 0x80(%rsp) fldt 0x70(%rsp) fstpt 0x10(%rsp) fstpt (%rsp) movq %r12, %rdi movl %r15d, %r9d movq 0x40(%rsp), %r13 movq %r13, %rsi callq 0x2ab5f9 movq 0x50(%rsp), %rcx movq 0x58(%rsp), %r14 orb %al, %cl addq $0x4, %r14 movq %rcx, %rax jmp 0x2aadc3 andb $0x1, %al addq $0x98, %rsp popq %rbx popq %r12 popq %r13 popq %r14 popq %r15 popq %rbp retq nop
/quesnel[P]baryonyx/lib/src/itm-solver-inequalities-01.cpp
int baryonyx::itm::compute_order::local_compute_violated_constraints<baryonyx::itm::solver_inequalities_01coeff<long double, baryonyx::itm::minimize_tag, baryonyx::itm::quadratic_cost_type<long double>, true>, baryonyx::bit_array>(baryonyx::itm::solver_inequalities_01coeff<long double, baryonyx::itm::minimize_tag, baryonyx::itm::quadratic_cost_type<long double>, true> const&, baryonyx::bit_array const&)
int local_compute_violated_constraints(const Solver& slv, const Xtype& x) { int remaining = 0; for (int k = 0; k != slv.m; ++k) if (!is_valid_constraint(slv, k, x)) ++remaining; return remaining; }
pushq %rbp pushq %r15 pushq %r14 pushq %rbx pushq %rax movq %rdx, %rbx movq %rsi, %r14 xorl %ebp, %ebp xorl %r15d, %r15d cmpl 0x78(%r14), %r15d je 0x2ab44a movq %r14, %rdi movl %r15d, %esi movq %rbx, %rdx callq 0x2aaab3 xorb $0x1, %al incl %r15d movzbl %al, %eax addl %eax, %ebp jmp 0x2ab42a movl %ebp, %eax addq $0x8, %rsp popq %rbx popq %r14 popq %r15 popq %rbp retq nop
/quesnel[P]baryonyx/lib/src/itm-common.hpp
bool baryonyx::itm::solver_inequalities_01coeff<long double, baryonyx::itm::minimize_tag, baryonyx::itm::quadratic_cost_type<long double>, true>::push_and_compute_update_row<baryonyx::bit_array, __gnu_cxx::__normal_iterator<int*, std::vector<int, std::allocator<int>>>>(baryonyx::bit_array&, __gnu_cxx::__normal_iterator<int*, std::vector<int, std::allocator<int>>>, __gnu_cxx::__normal_iterator<int*, std::vector<int, std::allocator<int>>>, long double, long double, long double, long double)
bool push_and_compute_update_row(Xtype& x, Iterator first, Iterator last, Float kappa, Float delta, Float theta, Float objective_amplifier) { auto at_least_one_pi_changed{ false }; logger::log("push-update-row {} {} {}\n", kappa, delta, theta); for (; first != last; ++first) { auto k = constraint(first); bx_expects(k < m); const auto it = ap.row(k); decrease_preference(std::get<0>(it), std::get<1>(it), theta); const auto r_size = compute_reduced_costs(std::get<0>(it), std::get<1>(it), x); // Before sort and select variables, we apply the push method: for // each reduces cost, we had the cost multiply with an objective // amplifier. for (int i = 0; i != r_size; ++i) R[i].value += objective_amplifier * c(R[i].id, x); calculator_sort<Mode>(R.get(), R.get() + r_size, rng); int selected = select_variables(r_size, b[k].min, b[k].max); logger::log("constraints {}: {} <= ", k, b[k].min); for (int i = 0; i < r_size; ++i) logger::log("({} {}) ", R[i].value, R[i].id); logger::log("<= {} => Selected: {}\n", b[k].max, selected); auto pi_change = affect( *this, x, std::get<0>(it), k, selected, r_size, kappa, delta); at_least_one_pi_changed = at_least_one_pi_changed || pi_change; } return at_least_one_pi_changed; }
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0xa8, %rsp movq %rcx, 0x48(%rsp) movq %rdx, %r14 movq %rsi, %r15 movq %rdi, %r12 leaq 0x1aed87(%rip), %rdx # 0x45c33f fldt 0x110(%rsp) fstpt 0x64(%rsp) fldt 0x100(%rsp) fldt 0xf0(%rsp) fldt 0xe0(%rsp) fstpt 0x90(%rsp) fstpt 0x80(%rsp) fstpt 0x70(%rsp) pushq $0x19 popq %rsi leaq 0x90(%rsp), %rcx leaq 0x80(%rsp), %r8 leaq 0x70(%rsp), %r9 callq 0x126c32 leaq 0x10(%r12), %rax leaq 0x1b612b(%rip), %rbp # 0x46373e movq %rax, 0x40(%rsp) xorl %eax, %eax pushq $0x8 popq %rbx cmpq 0x48(%rsp), %r14 je 0x2ad7d2 movl (%r14), %edx movl %edx, 0x24(%rsp) cmpl 0x78(%r12), %edx jge 0x2ad7e6 movq 0x40(%rsp), %rsi leaq 0x30(%rsp), %rdi movq %rax, 0x50(%rsp) movq %r14, 0x58(%rsp) callq 0x3de66 movq 0x30(%rsp), %rdx movq 0x38(%rsp), %rsi fldt 0x70(%rsp) movq %r12, %rdi fstpt (%rsp) callq 0x2ab458 movq 0x30(%rsp), %rdx movq 0x38(%rsp), %rsi movq %r12, %rdi movq %r15, %rcx callq 0x2ab47e movl %eax, %r14d shlq $0x5, %r14 xorl %r13d, %r13d movl %eax, 0x28(%rsp) cmpq %r13, %r14 je 0x2ad6c6 movq 0x58(%r12), %rax movq 0x70(%r12), %rdi movq %r15, %rdx movl 0x10(%rax,%r13), %esi callq 0x9c23c movq 0x58(%r12), %rax fldt (%rax,%r13) fldt 0x64(%rsp) fmulp %st, %st(2) faddp %st, %st(1) fstpt (%rax,%r13) addq $0x20, %r13 jmp 0x2ad68f movl 0x28(%rsp), %r14d movq 0x58(%r12), %rdi movq 0x8(%r12), %rdx movslq %r14d, %rsi shlq $0x5, %rsi addq %rdi, %rsi callq 0x2ab554 movslq 0x24(%rsp), %rax movq 0x60(%r12), %rcx movq %r12, %rdi movl %r14d, %esi movl (%rcx,%rax,8), %edx movl 0x4(%rcx,%rax,8), %ecx callq 0x2ab588 movslq 0x24(%rsp), %r8 movl %eax, 0x2c(%rsp) movq %r12, %rdi shlq $0x3, %r8 addq 0x60(%r12), %r8 pushq $0x16 popq %rsi leaq 0x1b3926(%rip), %rdx # 0x461045 leaq 0x24(%rsp), %rcx callq 0xb91c2 movl %r14d, %eax sarl $0x1f, %eax xorl %r13d, %r13d andnl %r14d, %eax, %r14d shlq $0x5, %r14 cmpq %r13, %r14 je 0x2ad761 movq 0x58(%r12), %rcx movq %r12, %rdi movq %rbx, %rsi movq %rbp, %rdx leaq 0x10(%rcx,%r13), %r8 addq %r13, %rcx callq 0x126d84 addq $0x20, %r13 jmp 0x2ad73b movslq 0x24(%rsp), %rax movq 0x60(%r12), %rcx movq %r12, %rdi leaq 0x4(%rcx,%rax,8), %rcx pushq $0x16 popq %rsi leaq 0x1b38df(%rip), %rdx # 0x46105c leaq 0x2c(%rsp), %r8 callq 0xb91c2 movq 0x38(%rsp), %rdx movl 0x24(%rsp), %ecx movl 0x2c(%rsp), %r8d fldt 0x90(%rsp) fldt 0x80(%rsp) fstpt 0x10(%rsp) fstpt (%rsp) movq %r12, %rdi movq %r15, %rsi movl 0x28(%rsp), %r9d callq 0x2ab5f9 movq 0x50(%rsp), %rcx movq 0x58(%rsp), %r14 orb %al, %cl addq $0x4, %r14 movq %rcx, %rax jmp 0x2ad61d andb $0x1, %al addq $0xa8, %rsp popq %rbx popq %r12 popq %r13 popq %r14 popq %r15 popq %rbp retq leaq 0x1aa9f1(%rip), %rdi # 0x4581de leaq 0x1acc86(%rip), %rsi # 0x45a47a leaq 0x1b37da(%rip), %rdx # 0x460fd5 leaq 0x1b3871(%rip), %rcx # 0x461073 callq 0x2813f nop
/quesnel[P]baryonyx/lib/src/itm-solver-inequalities-01.cpp
baryonyx::itm::solver_functor<baryonyx::itm::solver_inequalities_01coeff<long double, baryonyx::itm::minimize_tag, baryonyx::itm::quadratic_cost_type<long double>, true>, long double, baryonyx::itm::minimize_tag, baryonyx::itm::quadratic_cost_type<long double>, baryonyx::pnm_observer>::store_if_better(baryonyx::bit_array const&, double, long)
void store_if_better(const bit_array& x, double current, long int i) { if (is_better_solution<Mode>(current, m_best.value)) { m_best.x = x; m_best.duration = duration(); m_best.loop = i; m_best.remaining_constraints = 0; m_best.value = current; } }
vmovsd 0x30(%rdi), %xmm1 vucomisd %xmm0, %xmm1 jbe 0x2b0640 pushq %r14 pushq %rbx pushq %rax movq %rdi, %r14 addq $0x20, %rdi movq %rdx, %rbx vmovsd %xmm0, (%rsp) callq 0x36a44 movq %r14, %rdi callq 0x2b06ba vmovsd (%rsp), %xmm1 vmovsd %xmm0, 0x38(%r14) movq %rbx, 0x48(%r14) andl $0x0, 0x50(%r14) vmovsd %xmm1, 0x30(%r14) addq $0x8, %rsp popq %rbx popq %r14 retq nop
/quesnel[P]baryonyx/lib/src/itm-solver-common.hpp
baryonyx::itm::solver_functor<baryonyx::itm::solver_inequalities_01coeff<long double, baryonyx::itm::minimize_tag, baryonyx::itm::quadratic_cost_type<long double>, true>, long double, baryonyx::itm::minimize_tag, baryonyx::itm::quadratic_cost_type<long double>, baryonyx::pnm_observer>::is_timelimit_reached()
bool is_timelimit_reached() { m_end = std::chrono::steady_clock::now(); return is_time_limit(m_ctx.parameters.time_limit, m_begin, m_end); }
pushq %rbx movq %rdi, %rbx callq 0xb5a0 movq %rax, 0x8(%rbx) vxorpd %xmm2, %xmm2, %xmm2 movq 0x10(%rbx), %rcx subq (%rbx), %rax vmovsd (%rcx), %xmm0 vcvtsi2sd %rax, %xmm1, %xmm1 vdivsd 0x1a93dd(%rip), %xmm1, %xmm1 # 0x459a80 vcmpnlepd %xmm2, %xmm0, %k1 vcmpltpd %xmm1, %xmm0, %k0 {%k1} kmovd %k0, %eax andb $0x1, %al popq %rbx retq nop
/quesnel[P]baryonyx/lib/src/itm-solver-common.hpp
baryonyx::itm::optimize_functor<baryonyx::itm::solver_inequalities_01coeff<long double, baryonyx::itm::minimize_tag, baryonyx::itm::quadratic_cost_type<long double>, false>, long double, baryonyx::itm::minimize_tag, baryonyx::itm::quadratic_cost_type<long double>>::operator()(std::atomic<bool> const&, baryonyx::itm::best_solution_recorder<baryonyx::itm::quadratic_cost_type<long double>, long double, baryonyx::itm::minimize_tag>&, std::vector<baryonyx::itm::merged_constraint, std::allocator<baryonyx::itm::merged_constraint>> const&, int, baryonyx::itm::quadratic_cost_type<long double> const&, double)
void operator()(const std::atomic_bool& stop_task, best_solution_recorder<Cost, Float, Mode>& best_recorder, const std::vector<merged_constraint>& constraints, int variables, const Cost& original_costs, double cost_constant) { bit_array x(variables); auto& p = m_ctx.parameters; auto norm_costs = normalize_costs<Float, Cost>( m_ctx, original_costs, m_local_ctx.rng, variables); const auto kappa_step = static_cast<Float>(p.kappa_step); const auto kappa_max = static_cast<Float>(p.kappa_max); const auto alpha = static_cast<Float>(p.alpha); const auto theta = static_cast<Float>(p.theta); const auto delta = p.delta < 0 ? compute_delta<Float, Cost>(m_ctx, norm_costs, theta, variables) : static_cast<Float>(p.delta); const auto pushing_k_factor = static_cast<Float>(p.pushing_k_factor); const auto pushing_objective_amplifier = static_cast<Float>(p.pushing_objective_amplifier); const auto w_limit = static_cast<long int>(p.w); Solver slv(m_local_ctx.rng, length(constraints), variables, norm_costs, constraints); compute_order compute(p.order, variables); bool is_a_solution = false; while (!stop_task.load()) { ++m_call_number; const auto kappa_start = static_cast<Float>(best_recorder.reinit( m_local_ctx, is_a_solution, p.kappa_min, p.kappa_max, x)); auto kappa = kappa_start; compute.init(slv, x); auto best_remaining = INT_MAX; is_a_solution = false; for (long int i = 0; !stop_task.load() && i != p.limit; ++i) { auto remaining = compute.run(slv, x, m_local_ctx.rng, kappa, delta, theta); if (remaining == 0) { best_recorder.try_update( m_local_ctx, x, original_costs.results(x, cost_constant), i); best_remaining = 0; is_a_solution = true; break; } else { best_remaining = std::min(remaining, best_remaining); } if (i > w_limit) kappa += kappa_step * std::pow(static_cast<Float>(remaining) / static_cast<Float>(slv.m), alpha); if (kappa > kappa_max) break; } if (best_remaining > 0) { best_recorder.try_advance( m_local_ctx, x, best_remaining, p.limit); continue; } for (int push = 0; !stop_task.load() && push < p.pushes_limit; ++push) { auto remaining = compute.push_and_run(slv, x, m_local_ctx.rng, pushing_k_factor, delta, theta, pushing_objective_amplifier); if (remaining == 0) { best_recorder.try_update( m_local_ctx, x, original_costs.results(x, cost_constant), -push * p.pushing_iteration_limit - 1); is_a_solution = true; } kappa = kappa_start; for (int iter = 0; !stop_task.load() && iter < p.pushing_iteration_limit; ++iter) { remaining = compute.run( slv, x, m_local_ctx.rng, kappa, delta, theta); if (remaining == 0) { best_recorder.try_update( m_local_ctx, x, original_costs.results(x, cost_constant), -push * p.pushing_iteration_limit - iter - 1); is_a_solution = true; break; } if (iter > p.w) kappa += kappa_step * std::pow(static_cast<Float>(remaining) / static_cast<Float>(slv.m), alpha); if (kappa > kappa_max) break; } } } }
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x1f8, %rsp # imm = 0x1F8 movq %rdi, %rbp leaq 0x60(%rsp), %rdi movq %rsi, %r15 movl %r8d, %esi vmovsd %xmm0, 0x90(%rsp) movq %r9, %r12 movl %r8d, %ebx movq %rcx, %r14 movq %rdx, 0x70(%rsp) callq 0x368ae movq (%rbp), %r13 movq %rbp, 0xb8(%rsp) addq $0x8, %rbp leaq 0x120(%rsp), %rdi movq %r13, %rsi movq %r12, 0x88(%rsp) movq %r12, %rdx movq %rbp, %rcx movl %ebx, %r8d callq 0x98bf0 fldl 0x20(%r13) fstpt 0xd8(%rsp) vxorpd %xmm1, %xmm1, %xmm1 fldl 0x28(%r13) fstpt 0xe4(%rsp) fldl 0x30(%r13) fstpt 0xcc(%rsp) fldl 0x8(%r13) vmovsd 0x10(%r13), %xmm0 fld %st(0) fstpt 0x4c(%rsp) vucomisd %xmm0, %xmm1 jbe 0x2b0d87 movq 0xb8(%rsp), %rax movq (%rax), %rdi leaq 0x120(%rsp), %rsi movl %ebx, %edx fstpt (%rsp) callq 0x98cca fstpt 0x9c(%rsp) jmp 0x2b0da0 fstp %st(0) vmovsd %xmm0, 0x100(%rsp) fldl 0x100(%rsp) fstpt 0x9c(%rsp) fldl 0x38(%r13) fstpt 0x114(%rsp) fldl 0x40(%r13) fstpt 0x108(%rsp) vcvttsd2si 0xa0(%r13), %rax movq %rax, 0xf0(%rsp) movq 0x8(%r14), %rax subq (%r14), %rax pushq $0x28 popq %rcx cqto idivq %rcx leaq 0x178(%rsp), %rdi leaq 0x120(%rsp), %r8 movq %rbp, %rsi movl %eax, %edx movl %ebx, %ecx movq %r14, %r9 callq 0x2b126c movl 0xcc(%r13), %esi leaq 0x140(%rsp), %rdi movl %ebx, %edx callq 0x38cd8 movq $0x0, 0x58(%rsp) movb (%r15), %al testb $0x1, %al jne 0x2b11d4 movq 0xb8(%rsp), %rax incq 0xb0(%rax) vmovsd 0x18(%r13), %xmm0 vmovsd 0x28(%r13), %xmm1 movzbl 0x58(%rsp), %edx movq 0x70(%rsp), %rdi leaq 0x60(%rsp), %rbx movq %rbp, %rsi movq %rbx, %rcx andl $0x1, %edx callq 0xad93c vmovsd %xmm0, 0xf8(%rsp) fldl 0xf8(%rsp) fstpt 0xc0(%rsp) leaq 0x140(%rsp), %rdi leaq 0x178(%rsp), %rsi movq %rbx, %rdx callq 0x2b1416 fldt 0xc0(%rsp) fldt 0x4c(%rsp) movl $0x7fffffff, %ebx # imm = 0x7FFFFFFF xorl %r14d, %r14d movb (%r15), %al testb $0x1, %al jne 0x2b0f69 cmpq 0xb0(%r13), %r14 je 0x2b0f69 fstpt 0x20(%rsp) leaq 0x140(%rsp), %rdi leaq 0x178(%rsp), %rsi leaq 0x60(%rsp), %rdx movq %rbp, %rcx fldt 0x9c(%rsp) fstpt 0x10(%rsp) fld %st(0) fstpt 0x7c(%rsp) fstpt (%rsp) callq 0x2b1458 testl %eax, %eax je 0x2b0f83 cmpl %eax, %ebx cmovgel %eax, %ebx cmpq 0xf0(%rsp), %r14 jle 0x2b0f48 movl %eax, 0xb0(%rsp) fildl 0xb0(%rsp) movl 0x1f0(%rsp), %eax movl %eax, 0xb4(%rsp) fidivl 0xb4(%rsp) fldt 0xcc(%rsp) fstpt 0x10(%rsp) fstpt (%rsp) callq 0xb110 fldt 0xd8(%rsp) fmulp %st, %st(1) fldt 0x7c(%rsp) faddp %st, %st(1) fldt 0x4c(%rsp) jmp 0x2b0f52 fldt 0x4c(%rsp) fldt 0x7c(%rsp) fxch %st(1) fldt 0xe4(%rsp) fxch %st(2) fucomi %st(2), %st fstp %st(2) ja 0x2b0f69 incq %r14 jmp 0x2b0e99 fstp %st(1) testl %ebx, %ebx jg 0x2b11b2 movq $0x0, 0x58(%rsp) leaq 0x60(%rsp), %rbx jmp 0x2b0fbf vmovsd 0x90(%rsp), %xmm0 movq 0x88(%rsp), %rdi leaq 0x60(%rsp), %rbx movq %rbx, %rsi callq 0x9706e movq 0x70(%rsp), %rdi movq %rbp, %rsi movq %rbx, %rdx movq %r14, %rcx callq 0xadcae movb $0x1, %al movq %rax, 0x58(%rsp) fldt 0x4c(%rsp) xorl %r14d, %r14d movl $0x0, 0x78(%rsp) movb (%r15), %al testb $0x1, %al jne 0x2b11ab cmpl 0xbc(%r13), %r14d jge 0x2b11ab fldt 0x108(%rsp) fstpt 0x30(%rsp) fstpt 0x20(%rsp) leaq 0x140(%rsp), %rdi leaq 0x178(%rsp), %rsi movq %rbx, %rdx movq %rbp, %rcx fldt 0x9c(%rsp) fstpt 0x10(%rsp) fldt 0x114(%rsp) fstpt (%rsp) callq 0x2b1688 testl %eax, %eax jne 0x2b1065 vmovsd 0x90(%rsp), %xmm0 movq 0x88(%rsp), %rdi movq %rbx, %rsi callq 0x9706e movl 0xc0(%r13), %eax movq 0x70(%rsp), %rdi movq %rbp, %rsi movq %rbx, %rdx imull %r14d, %eax notl %eax movslq %eax, %rcx callq 0xadcae movb $0x1, %al movq %rax, 0x58(%rsp) pushq $0x1 popq %rax fldt 0xc0(%rsp) fldt 0x4c(%rsp) movl %eax, %ebx movb (%r15), %al testb $0x1, %al jne 0x2b1149 leal -0x1(%rbx), %r12d cmpl 0xc0(%r13), %r12d jge 0x2b1149 fstpt 0x20(%rsp) leaq 0x140(%rsp), %rdi leaq 0x178(%rsp), %rsi leaq 0x60(%rsp), %rdx movq %rbp, %rcx fldt 0x9c(%rsp) fstpt 0x10(%rsp) fld %st(0) fstpt 0x7c(%rsp) fstpt (%rsp) callq 0x2b1458 testl %eax, %eax je 0x2b1152 vcvtsi2sd %r12d, %xmm2, %xmm0 vucomisd 0xa0(%r13), %xmm0 jbe 0x2b1129 movl %eax, 0xa8(%rsp) fildl 0xa8(%rsp) movl 0x1f0(%rsp), %eax movl %eax, 0xac(%rsp) fidivl 0xac(%rsp) fldt 0xcc(%rsp) fstpt 0x10(%rsp) fstpt (%rsp) callq 0xb110 fldt 0xd8(%rsp) fmulp %st, %st(1) fldt 0x7c(%rsp) faddp %st, %st(1) fldt 0x4c(%rsp) jmp 0x2b1133 fldt 0x4c(%rsp) fldt 0x7c(%rsp) fxch %st(1) fldt 0xe4(%rsp) fxch %st(2) fucomi %st(2), %st fstp %st(2) ja 0x2b1149 incl %ebx jmp 0x2b1075 fstp %st(1) leaq 0x60(%rsp), %rbx jmp 0x2b119f vmovsd 0x90(%rsp), %xmm0 movq 0x88(%rsp), %rdi leaq 0x60(%rsp), %r12 movq %r12, %rsi callq 0x9706e movl 0xc0(%r13), %eax movq 0x70(%rsp), %rdi movq %rbp, %rsi movq %r12, %rdx imull 0x78(%rsp), %eax subl %ebx, %eax movq %r12, %rbx movslq %eax, %rcx callq 0xadcae movb $0x1, %al movq %rax, 0x58(%rsp) fldt 0x4c(%rsp) decl 0x78(%rsp) incl %r14d jmp 0x2b0fca fstp %st(0) jmp 0x2b0e14 fstp %st(0) movq 0xb0(%r13), %r8 movq 0x70(%rsp), %rdi leaq 0x60(%rsp), %rdx movq %rbp, %rsi movl %ebx, %ecx callq 0xadd36 jmp 0x2b0e0b leaq 0x140(%rsp), %rdi callq 0x393ac leaq 0x178(%rsp), %rdi callq 0x2b192c leaq 0x120(%rsp), %rdi callq 0x968b6 leaq 0x68(%rsp), %rdi callq 0x357fe addq $0x1f8, %rsp # imm = 0x1F8 popq %rbx popq %r12 popq %r13 popq %r14 popq %r15 popq %rbp retq jmp 0x2b121e movq %rax, %rbx jmp 0x2b1240 movq %rax, %rbx jmp 0x2b124d movq %rax, %rbx jmp 0x2b125a jmp 0x2b1230 jmp 0x2b1230 jmp 0x2b1230 jmp 0x2b1230 leaq 0x140(%rsp), %rdi movq %rax, %rbx callq 0x393ac leaq 0x178(%rsp), %rdi callq 0x2b192c leaq 0x120(%rsp), %rdi callq 0x968b6 leaq 0x68(%rsp), %rdi callq 0x357fe movq %rbx, %rdi callq 0xb3d0
/quesnel[P]baryonyx/lib/src/itm-optimizer-common.hpp
baryonyx::itm::solver_inequalities_01coeff<long double, baryonyx::itm::minimize_tag, baryonyx::itm::quadratic_cost_type<long double>, false>::decrease_preference(baryonyx::sparse_matrix<int>::row_value*, baryonyx::sparse_matrix<int>::row_value*, long double)
void decrease_preference(sparse_matrix<int>::row_iterator begin, sparse_matrix<int>::row_iterator end, Float theta) noexcept { for (; begin != end; ++begin) P[begin->value] *= theta; }
fldt 0x8(%rsp) movq 0x50(%rdi), %rax cmpq %rdx, %rsi je 0x2b218a movslq (%rsi), %rcx addq $0x8, %rsi shlq $0x4, %rcx fldt (%rax,%rcx) fmul %st(1), %st fstpt (%rax,%rcx) jmp 0x2b2170 fstp %st(0) retq nop
/quesnel[P]baryonyx/lib/src/itm-solver-inequalities-01.cpp
baryonyx::itm::solver_functor<baryonyx::itm::solver_inequalities_01coeff<long double, baryonyx::itm::minimize_tag, baryonyx::itm::quadratic_cost_type<long double>, false>, long double, baryonyx::itm::minimize_tag, baryonyx::itm::quadratic_cost_type<long double>, baryonyx::file_observer>::operator()(std::vector<baryonyx::itm::merged_constraint, std::allocator<baryonyx::itm::merged_constraint>> const&, int, baryonyx::itm::quadratic_cost_type<long double> const&, double)
result operator()(const std::vector<merged_constraint>& constraints, int variables, const Cost& original_costs, double cost_constant) { result r; bit_array x(variables); int best_remaining = INT_MAX; auto& p = m_ctx.parameters; auto norm_costs = normalize_costs<Float, Cost>( m_ctx, original_costs, m_rng, variables); const auto kappa_step = static_cast<Float>(p.kappa_step); const auto kappa_max = static_cast<Float>(p.kappa_max); const auto alpha = static_cast<Float>(p.alpha); const auto theta = static_cast<Float>(p.theta); const auto delta = p.delta < 0 ? compute_delta<Float, Cost>(m_ctx, norm_costs, theta, variables) : static_cast<Float>(p.delta); const auto pushing_k_factor = static_cast<Float>(p.pushing_k_factor); const auto pushing_objective_amplifier = static_cast<Float>(p.pushing_objective_amplifier); const long int w_limit = static_cast<long int>(p.w); Solver slv( m_rng, length(constraints), variables, norm_costs, constraints); compute_order compute(p.order, variables); { std::bernoulli_distribution choose_mutation( m_ctx.parameters.init_policy_random); bit_array empty_x; switch (p.init_policy) { case solver_parameters::init_policy_type::pessimistic_solve: init_with_pre_solve<Cost, Mode>( x, empty_x, m_rng, original_costs, constraints, 1.0); break; case solver_parameters::init_policy_type::optimistic_solve: init_with_pre_solve<Cost, Mode>( empty_x, x, m_rng, original_costs, constraints, 1.0); break; case solver_parameters::init_policy_type::bastert: init_with_bastert<Cost, Mode>(x, original_costs, variables, 0); break; } for (int i = 0, e = x.size(); i != e; ++i) if (choose_mutation(m_rng)) x.invert(i); } bool start_push = false; auto kappa = static_cast<Float>(p.kappa_min); Observer obs("img", slv.m, slv.n, p.limit); m_begin = std::chrono::steady_clock::now(); m_end = std::chrono::steady_clock::now(); compute.init(slv, x); for (long int i = 0; i != p.limit; ++i) { auto remaining = compute.run(slv, x, m_rng, kappa, delta, theta); obs.make_observation(slv.ap, slv.P.get(), slv.pi.get()); if (remaining == 0) { store_if_better( x, original_costs.results(x, cost_constant), i); best_remaining = remaining; start_push = true; break; } if (remaining < best_remaining) { store_if_better(x, remaining, i); best_remaining = remaining; } if (i > w_limit) kappa += kappa_step * std::pow(static_cast<Float>(remaining) / static_cast<Float>(slv.m), alpha); if (kappa > kappa_max) { r.status = result_status::kappa_max_reached; break; } if (is_timelimit_reached()) { r.status = result_status::time_limit_reached; break; } } if (!start_push) { r.status = result_status::limit_reached; } else { for (int push = 0; push < p.pushes_limit; ++push) { auto remaining = compute.push_and_run(slv, x, m_rng, pushing_k_factor * kappa, delta, theta, pushing_objective_amplifier); if (remaining == 0) store_if_better(x, original_costs.results(x, cost_constant), -push * p.pushing_iteration_limit - 1); if (is_timelimit_reached()) break; for (int iter = 0; iter < p.pushing_iteration_limit; ++iter) { remaining = compute.run(slv, x, m_rng, kappa, delta, theta); if (remaining == 0) { store_if_better( x, original_costs.results(x, cost_constant), -push * p.pushing_iteration_limit - iter - 1); break; } if (iter > p.w) kappa += kappa_step * std::pow(static_cast<Float>(remaining) / static_cast<Float>(slv.m), alpha); if (kappa > kappa_max) break; if (is_timelimit_reached()) break; } } } if (m_best.remaining_constraints == 0) r.status = result_status::success; if (!m_best.x.empty()) { r.solutions.resize(1); convert(m_best, r.solutions[0], variables); } return r; }
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x438, %rsp # imm = 0x438 vmovsd %xmm0, 0x88(%rsp) vxorpd %xmm0, %xmm0, %xmm0 movq %r8, %r13 movl %ecx, %ebx movq %rdx, %r15 movq %rsi, %r14 movq %rdi, 0x60(%rsp) vmovupd %zmm0, 0x58(%rdi) vmovupd %zmm0, 0x40(%rdi) vmovupd %zmm0, (%rdi) movq $0x7fffffff, 0x98(%rdi) # imm = 0x7FFFFFFF movl $0x2, 0xa0(%rdi) leaq 0x70(%rsp), %rdi movl %ecx, %esi vzeroupper callq 0x368ae movq 0x10(%r14), %r12 movq 0x18(%r14), %rcx leaq 0x108(%rsp), %rdi movq %r12, %rsi movq %r13, %rdx movl %ebx, %r8d callq 0x98bf0 fldl 0x20(%r12) fstpt 0xc8(%rsp) vxorpd %xmm1, %xmm1, %xmm1 fldl 0x28(%r12) fstpt 0xd4(%rsp) fldl 0x30(%r12) fstpt 0xbc(%rsp) fldl 0x8(%r12) vmovsd 0x10(%r12), %xmm0 fld %st(0) fstpt 0x54(%rsp) vucomisd %xmm0, %xmm1 jbe 0x2b626a movq 0x10(%r14), %rdi leaq 0x108(%rsp), %rsi movl %ebx, %edx fstpt (%rsp) callq 0x98cca fstpt 0xa0(%rsp) jmp 0x2b6283 fstp %st(0) vmovsd %xmm0, 0xe8(%rsp) fldl 0xe8(%rsp) fstpt 0xa0(%rsp) fldl 0x38(%r12) fstpt 0xfc(%rsp) fldl 0x40(%r12) fstpt 0xf0(%rsp) vcvttsd2si 0xa0(%r12), %rax movq 0x18(%r14), %rsi movq %rax, 0x90(%rsp) movq 0x8(%r15), %rax subq (%r15), %rax pushq $0x28 popq %rcx cqto idivq %rcx leaq 0x160(%rsp), %rdi leaq 0x108(%rsp), %r8 movl %eax, %edx movl %ebx, %ecx movq %r15, %r9 callq 0x2b126c movl 0xcc(%r12), %esi leaq 0x128(%rsp), %rdi movl %ebx, %edx callq 0x38cd8 movq 0x10(%r14), %rax movl %ebx, 0x9c(%rsp) vmovsd 0x48(%rax), %xmm0 vmovsd %xmm0, 0x48(%rsp) vxorpd %xmm0, %xmm0, %xmm0 vmovapd %xmm0, 0x1e0(%rsp) movl 0xd4(%r12), %eax testl %eax, %eax je 0x2b6365 cmpl $0x1, %eax je 0x2b633f cmpl $0x2, %eax jne 0x2b6376 movq 0x18(%r14), %rdx leaq 0x1e0(%rsp), %rdi leaq 0x70(%rsp), %rsi jmp 0x2b6350 movq 0x18(%r14), %rdx leaq 0x70(%rsp), %rdi leaq 0x1e0(%rsp), %rsi vmovsd 0x19dce0(%rip), %xmm0 # 0x454038 movq %r13, %rcx movq %r15, %r8 callq 0xab9ee jmp 0x2b6376 leaq 0x70(%rsp), %rdi movq %r13, %rsi movl %ebx, %edx xorl %ecx, %ecx callq 0xab979 leaq 0x70(%rsp), %r15 movq %r13, 0x80(%rsp) xorl %ebx, %ebx movl (%r15), %ebp cmpl %ebx, %ebp je 0x2b63af movq 0x18(%r14), %rdi callq 0x36ac3 vmovsd 0x48(%rsp), %xmm1 vucomisd %xmm0, %xmm1 jbe 0x2b63ab movq %r15, %rdi movl %ebx, %esi callq 0x3626e incl %ebx jmp 0x2b6388 leaq 0x1e8(%rsp), %rdi callq 0x357fe vmovsd 0x18(%r12), %xmm0 movl 0x1d8(%rsp), %ecx movl 0x1dc(%rsp), %r8d movq 0xb0(%r12), %r9 vmovsd %xmm0, 0x48(%rsp) leaq 0x1a4190(%rip), %rdx # 0x45a577 leaq 0x1e0(%rsp), %rdi pushq $0x3 popq %rsi callq 0x4346e callq 0xb5a0 movq %rax, (%r14) callq 0xb5a0 movq %rax, 0x8(%r14) leaq 0x128(%rsp), %rdi leaq 0x160(%rsp), %rsi leaq 0x70(%rsp), %rdx callq 0x2b1416 vmovsd 0x48(%rsp), %xmm0 movl $0x7fffffff, %ebp # imm = 0x7FFFFFFF leaq 0x70(%rsp), %r13 xorl %r15d, %r15d vmovsd %xmm0, 0xe0(%rsp) fldl 0xe0(%rsp) fldt 0x54(%rsp) cmpq 0xb0(%r12), %r15 je 0x2b6558 movq 0x18(%r14), %rcx fstpt 0x20(%rsp) leaq 0x128(%rsp), %rdi leaq 0x160(%rsp), %rsi movq %r13, %rdx fldt 0xa0(%rsp) fstpt 0x10(%rsp) fld %st(0) fstpt 0x48(%rsp) fstpt (%rsp) callq 0x2b1458 movq 0x1b0(%rsp), %rdx movq 0x1c8(%rsp), %rcx movl %eax, %ebx leaq 0x1e0(%rsp), %rdi leaq 0x170(%rsp), %rsi callq 0x95692 testl %ebx, %ebx je 0x2b660c cmpl %ebp, %ebx jge 0x2b64d0 movq %r14, %rdi movq %r13, %rsi movl %ebx, %edx movq %r15, %rcx callq 0x2b70ce movl %ebx, %ebp cmpq 0x90(%rsp), %r15 jle 0x2b6521 movl %ebx, 0xb4(%rsp) fildl 0xb4(%rsp) movl 0x1d8(%rsp), %eax movl %eax, 0xb8(%rsp) fidivl 0xb8(%rsp) fldt 0xbc(%rsp) fstpt 0x10(%rsp) fstpt (%rsp) callq 0xb110 fldt 0xc8(%rsp) fmulp %st, %st(1) fldt 0x48(%rsp) faddp %st, %st(1) jmp 0x2b6525 fldt 0x48(%rsp) fldt 0xd4(%rsp) fxch %st(1) fucomi %st(1), %st fstp %st(1) fldz ja 0x2b6558 fstp %st(0) movq %r14, %rdi fstpt 0x48(%rsp) callq 0x2b7106 fldt 0x48(%rsp) fldt 0x54(%rsp) testb %al, %al jne 0x2b6558 incq %r15 jmp 0x2b6449 fstp %st(0) fstp %st(0) movq 0x60(%rsp), %rax movl $0x5, 0xa0(%rax) fldz fldz fstp %st(1) fstp %st(0) cmpl $0x0, 0x50(%r14) jne 0x2b6586 movq 0x60(%rsp), %rax andl $0x0, 0xa0(%rax) cmpq $0x0, 0x28(%r14) je 0x2b65b7 movq 0x60(%rsp), %rax leaq 0x68(%rax), %rbx pushq $0x1 popq %rsi movq %rbx, %rdi callq 0x35292 movq (%rbx), %rsi addq $0x20, %r14 movl 0x9c(%rsp), %edx movq %r14, %rdi callq 0x4f485 leaq 0x1e0(%rsp), %rdi callq 0x435d0 leaq 0x128(%rsp), %rdi callq 0x393ac leaq 0x160(%rsp), %rdi callq 0x2b192c leaq 0x108(%rsp), %rdi callq 0x968b6 leaq 0x78(%rsp), %rdi callq 0x357fe movq 0x60(%rsp), %rax addq $0x438, %rsp # imm = 0x438 popq %rbx popq %r12 popq %r13 popq %r14 popq %r15 popq %rbp retq vmovsd 0x88(%rsp), %xmm0 movq 0x80(%rsp), %rdi leaq 0x70(%rsp), %rbx movq %rbx, %rsi callq 0x9706e movq %r14, %rdi movq %rbx, %rsi movq %r15, %rdx callq 0x2b7080 movl $0x0, 0x6c(%rsp) leaq 0x160(%rsp), %r15 leaq 0x70(%rsp), %rbp xorl %ebx, %ebx fldt 0x54(%rsp) fldt 0x48(%rsp) cmpl 0xbc(%r12), %ebx jge 0x2b656f movq 0x18(%r14), %rcx fldt 0xfc(%rsp) fld %st(1) fstpt 0x48(%rsp) fmulp %st, %st(1) fldt 0xf0(%rsp) fstpt 0x30(%rsp) fxch %st(1) fstpt 0x20(%rsp) leaq 0x128(%rsp), %rdi movq %r15, %rsi movq %rbp, %rdx fldt 0xa0(%rsp) fstpt 0x10(%rsp) fstpt (%rsp) callq 0x2b1688 testl %eax, %eax jne 0x2b66e2 vmovsd 0x88(%rsp), %xmm0 movq 0x80(%rsp), %rdi movq %rbp, %rsi callq 0x9706e movl 0xc0(%r12), %eax imull %ebx, %eax notl %eax movslq %eax, %rdx movq %r14, %rdi movq %rbp, %rsi callq 0x2b7080 movq %r14, %rdi callq 0x2b7106 fldt 0x54(%rsp) fldt 0x48(%rsp) testb %al, %al jne 0x2b656f movl %ebx, 0x90(%rsp) pushq $0x1 popq %rax movl %eax, %ebx leal -0x1(%rbx), %r13d cmpl 0xc0(%r12), %r13d jge 0x2b682b movq 0x18(%r14), %rcx fxch %st(1) fstpt 0x20(%rsp) leaq 0x128(%rsp), %rdi movq %r15, %rsi movq %rbp, %rdx fldt 0xa0(%rsp) fstpt 0x10(%rsp) fld %st(0) fstpt 0x48(%rsp) fstpt (%rsp) callq 0x2b1458 testl %eax, %eax je 0x2b67db vcvtsi2sd %r13d, %xmm1, %xmm0 vucomisd 0xa0(%r12), %xmm0 jbe 0x2b67a9 movl %eax, 0xac(%rsp) fildl 0xac(%rsp) movl 0x1d8(%rsp), %eax movl %eax, 0xb0(%rsp) fidivl 0xb0(%rsp) fldt 0xbc(%rsp) fstpt 0x10(%rsp) fstpt (%rsp) callq 0xb110 fldt 0xc8(%rsp) fmulp %st, %st(1) fldt 0x48(%rsp) faddp %st, %st(1) jmp 0x2b67ad fldt 0x48(%rsp) fldt 0xd4(%rsp) fxch %st(1) fucomi %st(1), %st fstp %st(1) ja 0x2b681b movq %r14, %rdi fstpt 0x48(%rsp) callq 0x2b7106 testb %al, %al jne 0x2b6823 fldt 0x54(%rsp) fldt 0x48(%rsp) incl %ebx jmp 0x2b6706 vmovsd 0x88(%rsp), %xmm0 movq 0x80(%rsp), %rdi movq %rbp, %rsi callq 0x9706e movl 0xc0(%r12), %eax imull 0x6c(%rsp), %eax subl %ebx, %eax movslq %eax, %rdx movq %r14, %rdi movq %rbp, %rsi callq 0x2b7080 fldt 0x54(%rsp) fldt 0x48(%rsp) jmp 0x2b682b fldt 0x54(%rsp) fxch %st(1) jmp 0x2b682b fldt 0x54(%rsp) fldt 0x48(%rsp) movl 0x90(%rsp), %ebx decl 0x6c(%rsp) incl %ebx jmp 0x2b6657 jmp 0x2b6878 jmp 0x2b6851 fstp %st(1) fstp %st(0) jmp 0x2b6878 movq %rax, %rbx jmp 0x2b6888 movq %rax, %rbx jmp 0x2b6895 movq %rax, %rbx jmp 0x2b68a2 movq %rax, %rbx jmp 0x2b68af movq %rax, %rbx jmp 0x2b68b9 jmp 0x2b6878 jmp 0x2b6878 jmp 0x2b6878 leaq 0x1e8(%rsp), %rdi movq %rax, %rbx callq 0x357fe jmp 0x2b6888 leaq 0x1e0(%rsp), %rdi movq %rax, %rbx callq 0x435d0 leaq 0x128(%rsp), %rdi callq 0x393ac leaq 0x160(%rsp), %rdi callq 0x2b192c leaq 0x108(%rsp), %rdi callq 0x968b6 leaq 0x78(%rsp), %rdi callq 0x357fe movq 0x60(%rsp), %rdi callq 0xea60 movq %rbx, %rdi callq 0xb3d0 nop
/quesnel[P]baryonyx/lib/src/itm-solver-common.hpp
bool baryonyx::itm::affect<baryonyx::itm::solver_inequalities_101coeff<float, baryonyx::itm::maximize_tag, baryonyx::itm::quadratic_cost_type<float>, false>, baryonyx::bit_array, baryonyx::sparse_matrix<int>::row_value*, float>(baryonyx::itm::solver_inequalities_101coeff<float, baryonyx::itm::maximize_tag, baryonyx::itm::quadratic_cost_type<float>, false>&, baryonyx::bit_array&, baryonyx::sparse_matrix<int>::row_value*, int, int, int, float, float)
bool affect(Solver& slv, Xtype& x, Iterator it, int k, int selected, int r_size, const Float kappa, const Float delta) { constexpr Float one{ 1 }; constexpr Float two{ 2 }; constexpr Float middle{ (two + one) / two }; const auto old_pi = slv.pi[k]; auto d = delta; if (selected < 0) { // slv.pi[k] += slv.R[0].value / two; d += (kappa / (one - kappa)) * (slv.R[0].value / two); for (int i = 0; i != r_size; ++i) { auto var = it + slv.R[i].id; if (slv.R[i].is_negative_factor()) { x.set(var->column); slv.P[var->value] += d; } else { x.unset(var->column); slv.P[var->value] -= d; } } } else if (selected + 1 >= r_size) { // slv.pi[k] += slv.R[selected].value * middle; d += (kappa / (one - kappa)) * (slv.R[selected].value * middle); for (int i = 0; i != r_size; ++i) { auto var = it + slv.R[i].id; if (slv.R[i].is_negative_factor()) { x.unset(var->column); slv.P[var->value] -= d; } else { x.set(var->column); slv.P[var->value] += d; } } } else { slv.pi[k] += ((slv.R[selected].value + slv.R[selected + 1].value) / two); d += (kappa / (one - kappa)) * (slv.R[selected + 1].value - slv.R[selected].value); int i = 0; for (; i <= selected; ++i) { auto var = it + slv.R[i].id; if (slv.R[i].is_negative_factor()) { x.unset(var->column); slv.P[var->value] -= d; } else { x.set(var->column); slv.P[var->value] += d; } } for (; i != r_size; ++i) { auto var = it + slv.R[i].id; if (slv.R[i].is_negative_factor()) { x.set(var->column); slv.P[var->value] += d; } else { x.unset(var->column); slv.P[var->value] -= d; } } } // TODO job: develops is_valid_constraint for all the solvers bx_expects(is_valid_constraint(slv, k, x)); return is_signbit_change(old_pi, slv.pi[k]); }
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x38, %rsp movq 0x78(%rdi), %rax movl %ecx, 0xc(%rsp) movslq %ecx, %rcx movq %rdx, %r15 movq %rsi, %r14 movq %rdi, %rbx movq %rcx, 0x10(%rsp) vmovss (%rax,%rcx,4), %xmm2 vmovaps %xmm2, 0x20(%rsp) testl %r8d, %r8d js 0x2cffbb leal 0x1(%r8), %ecx cmpl %r9d, %ecx jge 0x2d004d movq 0x68(%rbx), %rsi movl %r8d, %edx imulq $0xc, %rdx, %rdx movl %ecx, %ecx imulq $0xc, %rcx, %rbp movq %rcx, 0x18(%rsp) movq 0x10(%rsp), %rcx xorl %r13d, %r13d movl %r9d, 0x8(%rsp) vmovss (%rsi,%rdx), %xmm2 vaddss (%rsi,%rbp), %xmm2, %xmm2 vmulss 0x189ba4(%rip), %xmm2, %xmm2 # 0x459ad0 vaddss 0x20(%rsp), %xmm2, %xmm2 vmovss %xmm2, (%rax,%rcx,4) vmovss 0x189b85(%rip), %xmm2 # 0x459ac4 vsubss %xmm0, %xmm2, %xmm2 vdivss %xmm2, %xmm0, %xmm0 vmovss (%rsi,%rbp), %xmm2 vsubss (%rsi,%rdx), %xmm2, %xmm2 vfmadd213ss %xmm1, %xmm0, %xmm2 # xmm2 = (xmm0 * xmm2) + xmm1 vmovss %xmm2, 0x4(%rsp) cmpq %r13, %rbp je 0x2d00e7 movq 0x68(%rbx), %rax movq %r14, %rdi movslq 0x8(%rax,%r13), %r12 cmpl $0x0, 0x4(%rax,%r13) movl 0x4(%r15,%r12,8), %esi js 0x2cff98 callq 0x369c8 movslq (%r15,%r12,8), %rax movq 0x50(%rbx), %rcx vmovss 0x4(%rsp), %xmm0 vaddss (%rcx,%rax,4), %xmm0, %xmm0 jmp 0x2cffb0 callq 0x36a06 movslq (%r15,%r12,8), %rax movq 0x50(%rbx), %rcx vmovss (%rcx,%rax,4), %xmm0 vsubss 0x4(%rsp), %xmm0, %xmm0 vmovss %xmm0, (%rcx,%rax,4) addq $0xc, %r13 jmp 0x2cff5c vmovss 0x189b01(%rip), %xmm2 # 0x459ac4 movq 0x68(%rbx), %rax xorl %ebp, %ebp vsubss %xmm0, %xmm2, %xmm2 vdivss %xmm2, %xmm0, %xmm0 vmovss (%rax), %xmm2 movl %r9d, %eax imulq $0xc, %rax, %r12 vmulss 0x189aec(%rip), %xmm2, %xmm2 # 0x459ad0 vfmadd213ss %xmm1, %xmm0, %xmm2 # xmm2 = (xmm0 * xmm2) + xmm1 vmovss %xmm2, 0x4(%rsp) cmpq %rbp, %r12 je 0x2d014f movq 0x68(%rbx), %rax movq %r14, %rdi movslq 0x8(%rax,%rbp), %r13 cmpl $0x0, 0x4(%rax,%rbp) movl 0x4(%r15,%r13,8), %esi js 0x2d002a callq 0x36a06 movslq (%r15,%r13,8), %rax movq 0x50(%rbx), %rcx vmovss (%rcx,%rax,4), %xmm0 vsubss 0x4(%rsp), %xmm0, %xmm0 jmp 0x2d0042 callq 0x369c8 movslq (%r15,%r13,8), %rax movq 0x50(%rbx), %rcx vmovss 0x4(%rsp), %xmm0 vaddss (%rcx,%rax,4), %xmm0, %xmm0 vmovss %xmm0, (%rcx,%rax,4) addq $0xc, %rbp jmp 0x2cffef vmovss 0x189a6f(%rip), %xmm2 # 0x459ac4 movq 0x68(%rbx), %rcx movl %r8d, %eax imulq $0xc, %rax, %rax xorl %ebp, %ebp vsubss %xmm0, %xmm2, %xmm2 vdivss %xmm2, %xmm0, %xmm0 vmovss (%rcx,%rax), %xmm2 movl %r9d, %eax imulq $0xc, %rax, %r12 vmulss 0x18bc4a(%rip), %xmm2, %xmm2 # 0x45bcc8 vfmadd213ss %xmm1, %xmm0, %xmm2 # xmm2 = (xmm0 * xmm2) + xmm1 vmovss %xmm2, 0x4(%rsp) cmpq %rbp, %r12 je 0x2d014f movq 0x68(%rbx), %rax movq %r14, %rdi movslq 0x8(%rax,%rbp), %r13 cmpl $0x0, 0x4(%rax,%rbp) movl 0x4(%r15,%r13,8), %esi js 0x2d00c4 callq 0x369c8 movslq (%r15,%r13,8), %rax movq 0x50(%rbx), %rcx vmovss 0x4(%rsp), %xmm0 vaddss (%rcx,%rax,4), %xmm0, %xmm0 jmp 0x2d00dc callq 0x36a06 movslq (%r15,%r13,8), %rax movq 0x50(%rbx), %rcx vmovss (%rcx,%rax,4), %xmm0 vsubss 0x4(%rsp), %xmm0, %xmm0 vmovss %xmm0, (%rcx,%rax,4) addq $0xc, %rbp jmp 0x2d0089 movl 0x8(%rsp), %r12d addq $0x8, %rbp subq 0x18(%rsp), %r12 subq $0x1, %r12 jb 0x2d014f movq 0x68(%rbx), %rax movq %r14, %rdi movslq (%rax,%rbp), %r13 cmpl $0x0, -0x4(%rax,%rbp) movl 0x4(%r15,%r13,8), %esi js 0x2d012c callq 0x36a06 movslq (%r15,%r13,8), %rax movq 0x50(%rbx), %rcx vmovss (%rcx,%rax,4), %xmm0 vsubss 0x4(%rsp), %xmm0, %xmm0 jmp 0x2d0144 callq 0x369c8 movslq (%r15,%r13,8), %rax movq 0x50(%rbx), %rcx vmovss 0x4(%rsp), %xmm0 vaddss (%rcx,%rax,4), %xmm0, %xmm0 vmovss %xmm0, (%rcx,%rax,4) addq $0xc, %rbp jmp 0x2d00f5 movl 0xc(%rsp), %esi movq %rbx, %rdi movq %r14, %rdx callq 0x2cf68e testb %al, %al je 0x2d018d vmovaps 0x20(%rsp), %xmm0 movq 0x78(%rbx), %rcx movq 0x10(%rsp), %rdx vmovmskps %xmm0, %eax shll $0x1f, %eax xorl (%rcx,%rdx,4), %eax shrl $0x1f, %eax addq $0x38, %rsp popq %rbx popq %r12 popq %r13 popq %r14 popq %r15 popq %rbp retq leaq 0x18804a(%rip), %rdi # 0x4581de leaq 0x18c181(%rip), %rsi # 0x45c31c leaq 0x189dea(%rip), %rdx # 0x459f8c leaq 0x18c192(%rip), %rcx # 0x45c33b callq 0x2813f
/quesnel[P]baryonyx/lib/src/itm-common.hpp
void baryonyx::itm::random_shuffle_unique<baryonyx::itm::solver_inequalities_101coeff<float, baryonyx::itm::maximize_tag, baryonyx::itm::quadratic_cost_type<float>, false>::rc_data*>(baryonyx::itm::solver_inequalities_101coeff<float, baryonyx::itm::maximize_tag, baryonyx::itm::quadratic_cost_type<float>, false>::rc_data*, baryonyx::itm::solver_inequalities_101coeff<float, baryonyx::itm::maximize_tag, baryonyx::itm::quadratic_cost_type<float>, false>::rc_data*, std::linear_congruential_engine<unsigned long, 16807ul, 0ul, 2147483647ul>&)
inline void random_shuffle_unique(iteratorT begin, iteratorT end, random_engine& rng) noexcept { auto ret = begin++; for (; begin != end; ++begin) { if (ret->value != begin->value) { std::shuffle(ret, begin, rng); ret = begin; } } std::shuffle(ret, begin, rng); }
pushq %r15 pushq %r14 pushq %rbx movq %rdx, %rbx movq %rsi, %r14 leaq 0xc(%rdi), %rsi cmpq %r14, %rsi je 0x2d01e9 vmovss (%rdi), %xmm0 leaq 0xc(%rsi), %r15 vucomiss (%rsi), %xmm0 movq %r15, %rsi jne 0x2d01d5 jnp 0x2d01bd addq $-0xc, %r15 movq %r15, %rsi movq %rbx, %rdx callq 0x2d077f movq %r15, %rdi jmp 0x2d01b9 movq %rbx, %rdx callq 0x2d077f popq %rbx popq %r14 popq %r15 retq jmp 0x2d01f9 movq %rax, %rdi callq 0xeb9f
/quesnel[P]baryonyx/lib/src/itm-common.hpp
bool baryonyx::itm::solver_inequalities_101coeff<float, baryonyx::itm::maximize_tag, baryonyx::itm::quadratic_cost_type<float>, false>::push_and_compute_update_row<baryonyx::bit_array, __gnu_cxx::__normal_iterator<std::pair<int, int> const*, std::vector<std::pair<int, int>, std::allocator<std::pair<int, int>>>>>(baryonyx::bit_array&, __gnu_cxx::__normal_iterator<std::pair<int, int> const*, std::vector<std::pair<int, int>, std::allocator<std::pair<int, int>>>>, __gnu_cxx::__normal_iterator<std::pair<int, int> const*, std::vector<std::pair<int, int>, std::allocator<std::pair<int, int>>>>, float, float, float, float)
bool push_and_compute_update_row(Xtype& x, Iterator first, Iterator last, Float kappa, Float delta, Float theta, Float obj_amp) { auto at_least_one_pi_changed{ false }; logger::log("push-update-row {} {} {}\n", kappa, delta, theta); for (; first != last; ++first) { auto k = constraint(first); bx_expects(k < m); const auto it = ap.row(k); decrease_preference(std::get<0>(it), std::get<1>(it), theta); const auto sizes = compute_reduced_costs(std::get<0>(it), std::get<1>(it), x); for (int i = 0; i != sizes.r_size; ++i) R[i].value += obj_amp * c((std::get<0>(it) + R[i].id)->column, x); calculator_sort<Mode>(R.get(), R.get() + sizes.r_size, rng); int selected = select_variables(sizes, b[k].min, b[k].max); logger::log("constraints {}: {} <= ", k, b[k].min); for (int i = 0; i < sizes.r_size; ++i) logger::log("{} ({} {}) ", R[i].f, R[i].value, R[i].id); logger::log("<= {} => Selected: {}\n", b[k].max, selected); auto pi_change = affect(*this, x, std::get<0>(it), k, selected, sizes.r_size, kappa, delta); at_least_one_pi_changed = at_least_one_pi_changed || pi_change; } return at_least_one_pi_changed; }
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x48, %rsp leaq 0x10(%rdi), %rax vmovss %xmm3, 0x1c(%rsp) vmovss %xmm2, 0x18(%rsp) vmovss %xmm1, 0x14(%rsp) vmovss %xmm0, 0x10(%rsp) movq %rcx, 0x38(%rsp) movq %rdx, %r14 movq %rsi, %r12 movq %rdi, %r13 movq %rax, 0x30(%rsp) xorl %eax, %eax cmpq 0x38(%rsp), %r14 je 0x2d209c movslq (%r14), %rbp cmpl 0x88(%r13), %ebp jge 0x2d20ad movq 0x30(%rsp), %rsi leaq 0x20(%rsp), %rdi movl %ebp, %edx movq %rax, 0x40(%rsp) callq 0x3de66 vmovss 0x18(%rsp), %xmm0 movq 0x20(%rsp), %rdx movq 0x28(%rsp), %rsi movq %r13, %rdi callq 0x2cfc94 movq 0x20(%rsp), %rdx movq 0x28(%rsp), %rsi movq %r13, %rdi movq %r12, %rcx callq 0x2cfcb2 movq %rax, 0x8(%rsp) xorl %ebx, %ebx xorl %r15d, %r15d movl %eax, %ecx cmpq %rcx, %r15 je 0x2d2031 movq 0x68(%r13), %rax movq 0x28(%rsp), %rcx movq 0x80(%r13), %rdi movq %r12, %rdx movslq 0x8(%rax,%rbx), %rax movl 0x4(%rcx,%rax,8), %esi callq 0x49ffa movq 0x68(%r13), %rax vmovss 0x1c(%rsp), %xmm1 incq %r15 vfmadd213ss (%rax,%rbx), %xmm1, %xmm0 # xmm0 = (xmm1 * xmm0) + mem vmovss %xmm0, (%rax,%rbx) addq $0xc, %rbx movl 0x8(%rsp), %eax jmp 0x2d1fe7 movq 0x68(%r13), %rdi cltq movq 0x8(%r13), %rdx imulq $0xc, %rax, %rsi addq %rdi, %rsi callq 0x2cfe0b movq 0x70(%r13), %rax imulq $0xc, %rbp, %rcx leaq 0x8(%rsp), %rsi movq %r13, %rdi movl (%rax,%rcx), %edx movl 0x4(%rax,%rcx), %ecx callq 0x2cfe3e vmovss 0x10(%rsp), %xmm0 vmovss 0x14(%rsp), %xmm1 movl 0x8(%rsp), %r9d movq 0x28(%rsp), %rdx movq %r13, %rdi movq %r12, %rsi movl %ebp, %ecx movl %eax, %r8d callq 0x2cfeaf movq 0x40(%rsp), %rcx addq $0x8, %r14 orb %al, %cl movq %rcx, %rax jmp 0x2d1f7f andb $0x1, %al addq $0x48, %rsp popq %rbx popq %r12 popq %r13 popq %r14 popq %r15 popq %rbp retq leaq 0x18612a(%rip), %rdi # 0x4581de leaq 0x1883bf(%rip), %rsi # 0x45a47a leaq 0x1915d6(%rip), %rdx # 0x463698 leaq 0x19167e(%rip), %rcx # 0x463747 callq 0x2813f
/quesnel[P]baryonyx/lib/src/itm-solver-inequalities-101.cpp
baryonyx::itm::solver_functor<baryonyx::itm::solver_inequalities_101coeff<float, baryonyx::itm::maximize_tag, baryonyx::itm::quadratic_cost_type<float>, false>, float, baryonyx::itm::maximize_tag, baryonyx::itm::quadratic_cost_type<float>, baryonyx::pnm_observer>::operator()(std::vector<baryonyx::itm::merged_constraint, std::allocator<baryonyx::itm::merged_constraint>> const&, int, baryonyx::itm::quadratic_cost_type<float> const&, double)
result operator()(const std::vector<merged_constraint>& constraints, int variables, const Cost& original_costs, double cost_constant) { result r; bit_array x(variables); int best_remaining = INT_MAX; auto& p = m_ctx.parameters; auto norm_costs = normalize_costs<Float, Cost>( m_ctx, original_costs, m_rng, variables); const auto kappa_step = static_cast<Float>(p.kappa_step); const auto kappa_max = static_cast<Float>(p.kappa_max); const auto alpha = static_cast<Float>(p.alpha); const auto theta = static_cast<Float>(p.theta); const auto delta = p.delta < 0 ? compute_delta<Float, Cost>(m_ctx, norm_costs, theta, variables) : static_cast<Float>(p.delta); const auto pushing_k_factor = static_cast<Float>(p.pushing_k_factor); const auto pushing_objective_amplifier = static_cast<Float>(p.pushing_objective_amplifier); const long int w_limit = static_cast<long int>(p.w); Solver slv( m_rng, length(constraints), variables, norm_costs, constraints); compute_order compute(p.order, variables); { std::bernoulli_distribution choose_mutation( m_ctx.parameters.init_policy_random); bit_array empty_x; switch (p.init_policy) { case solver_parameters::init_policy_type::pessimistic_solve: init_with_pre_solve<Cost, Mode>( x, empty_x, m_rng, original_costs, constraints, 1.0); break; case solver_parameters::init_policy_type::optimistic_solve: init_with_pre_solve<Cost, Mode>( empty_x, x, m_rng, original_costs, constraints, 1.0); break; case solver_parameters::init_policy_type::bastert: init_with_bastert<Cost, Mode>(x, original_costs, variables, 0); break; } for (int i = 0, e = x.size(); i != e; ++i) if (choose_mutation(m_rng)) x.invert(i); } bool start_push = false; auto kappa = static_cast<Float>(p.kappa_min); Observer obs("img", slv.m, slv.n, p.limit); m_begin = std::chrono::steady_clock::now(); m_end = std::chrono::steady_clock::now(); compute.init(slv, x); for (long int i = 0; i != p.limit; ++i) { auto remaining = compute.run(slv, x, m_rng, kappa, delta, theta); obs.make_observation(slv.ap, slv.P.get(), slv.pi.get()); if (remaining == 0) { store_if_better( x, original_costs.results(x, cost_constant), i); best_remaining = remaining; start_push = true; break; } if (remaining < best_remaining) { store_if_better(x, remaining, i); best_remaining = remaining; } if (i > w_limit) kappa += kappa_step * std::pow(static_cast<Float>(remaining) / static_cast<Float>(slv.m), alpha); if (kappa > kappa_max) { r.status = result_status::kappa_max_reached; break; } if (is_timelimit_reached()) { r.status = result_status::time_limit_reached; break; } } if (!start_push) { r.status = result_status::limit_reached; } else { for (int push = 0; push < p.pushes_limit; ++push) { auto remaining = compute.push_and_run(slv, x, m_rng, pushing_k_factor * kappa, delta, theta, pushing_objective_amplifier); if (remaining == 0) store_if_better(x, original_costs.results(x, cost_constant), -push * p.pushing_iteration_limit - 1); if (is_timelimit_reached()) break; for (int iter = 0; iter < p.pushing_iteration_limit; ++iter) { remaining = compute.run(slv, x, m_rng, kappa, delta, theta); if (remaining == 0) { store_if_better( x, original_costs.results(x, cost_constant), -push * p.pushing_iteration_limit - iter - 1); break; } if (iter > p.w) kappa += kappa_step * std::pow(static_cast<Float>(remaining) / static_cast<Float>(slv.m), alpha); if (kappa > kappa_max) break; if (is_timelimit_reached()) break; } } } if (m_best.remaining_constraints == 0) r.status = result_status::success; if (!m_best.x.empty()) { r.solutions.resize(1); convert(m_best, r.solutions[0], variables); } return r; }
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x3c8, %rsp # imm = 0x3C8 vmovsd %xmm0, 0x60(%rsp) vxorps %xmm0, %xmm0, %xmm0 movq %r8, %r13 movl %ecx, %ebx movq %rdx, %r15 movq %rsi, %r14 movq %rdi, 0x20(%rsp) vmovups %zmm0, 0x58(%rdi) vmovups %zmm0, 0x40(%rdi) vmovups %zmm0, (%rdi) movq $0x7fffffff, 0x98(%rdi) # imm = 0x7FFFFFFF movl $0x2, 0xa0(%rdi) leaq 0x30(%rsp), %rdi movl %ecx, %esi vzeroupper callq 0x368ae movq 0x10(%r14), %r12 movq 0x18(%r14), %rcx leaq 0x88(%rsp), %rdi movq %r12, %rsi movq %r13, %rdx movl %ebx, %r8d callq 0x46c07 vmovsd 0x20(%r12), %xmm0 vmovsd 0x10(%r12), %xmm2 vxorpd %xmm1, %xmm1, %xmm1 vmovsd %xmm0, 0x50(%rsp) vmovsd 0x28(%r12), %xmm0 vucomisd %xmm2, %xmm1 vmovsd %xmm0, 0x48(%rsp) vmovsd 0x30(%r12), %xmm0 vmovsd %xmm0, 0x40(%rsp) vcvtsd2ss 0x8(%r12), %xmm3, %xmm0 vmovss %xmm0, 0xc(%rsp) jbe 0x2d3626 movq 0x10(%r14), %rdi leaq 0x88(%rsp), %rsi movl %ebx, %edx callq 0x46ce1 vmovss %xmm0, 0x1c(%rsp) jmp 0x2d3630 vcvtsd2ss %xmm2, %xmm2, %xmm0 vmovss %xmm0, 0x1c(%rsp) vcvttsd2si 0xa0(%r12), %rax vmovsd 0x38(%r12), %xmm0 vmovsd 0x40(%r12), %xmm1 movq 0x18(%r14), %rsi movq %rax, 0x78(%rsp) movq 0x8(%r15), %rax vmovsd %xmm0, 0x70(%rsp) vmovsd %xmm1, 0x68(%rsp) subq (%r15), %rax pushq $0x28 popq %rcx cqto idivq %rcx leaq 0xe0(%rsp), %rdi leaq 0x88(%rsp), %r8 movl %eax, %edx movl %ebx, %ecx movq %r15, %r9 callq 0x2cecc8 movl 0xcc(%r12), %esi leaq 0xa8(%rsp), %rdi movl %ebx, %edx callq 0x38cd8 movq 0x10(%r14), %rax movl %ebx, 0x84(%rsp) vmovsd 0x48(%rax), %xmm0 vmovsd %xmm0, 0x10(%rsp) vxorps %xmm0, %xmm0, %xmm0 vmovaps %xmm0, 0x170(%rsp) movl 0xd4(%r12), %eax testl %eax, %eax je 0x2d3711 cmpl $0x1, %eax je 0x2d36eb cmpl $0x2, %eax jne 0x2d3722 movq 0x18(%r14), %rdx leaq 0x170(%rsp), %rdi leaq 0x30(%rsp), %rsi jmp 0x2d36fc movq 0x18(%r14), %rdx leaq 0x30(%rsp), %rdi leaq 0x170(%rsp), %rsi vmovsd 0x180934(%rip), %xmm0 # 0x454038 movq %r13, %rcx movq %r15, %r8 callq 0x44e5a jmp 0x2d3722 leaq 0x30(%rsp), %rdi movq %r13, %rsi movl %ebx, %edx xorl %ecx, %ecx callq 0x44de5 vcvtsd2ss 0x50(%rsp), %xmm3, %xmm0 leaq 0x30(%rsp), %r15 movq %r13, 0x58(%rsp) xorl %ebx, %ebx movl (%r15), %ebp vmovss %xmm0, 0x2c(%rsp) vcvtsd2ss 0x48(%rsp), %xmm3, %xmm0 vmovss %xmm0, 0x50(%rsp) vcvtsd2ss 0x40(%rsp), %xmm3, %xmm0 vmovss %xmm0, 0x48(%rsp) vcvtsd2ss 0x70(%rsp), %xmm3, %xmm0 vmovss %xmm0, 0x70(%rsp) vcvtsd2ss 0x68(%rsp), %xmm3, %xmm0 vmovss %xmm0, 0x68(%rsp) cmpl %ebx, %ebp je 0x2d3794 movq 0x18(%r14), %rdi callq 0x36ac3 vmovsd 0x10(%rsp), %xmm1 vucomisd %xmm0, %xmm1 jbe 0x2d3790 movq %r15, %rdi movl %ebx, %esi callq 0x3626e incl %ebx jmp 0x2d376d leaq 0x178(%rsp), %rdi callq 0x357fe vmovsd 0x18(%r12), %xmm0 movl 0x168(%rsp), %ecx movl 0x16c(%rsp), %r8d movq 0xb0(%r12), %r9 vmovsd %xmm0, 0x10(%rsp) leaq 0x186dab(%rip), %rdx # 0x45a577 leaq 0x170(%rsp), %rdi pushq $0x3 popq %rsi callq 0x42b5a callq 0xb5a0 movq %rax, (%r14) callq 0xb5a0 movq %rax, 0x8(%r14) leaq 0xa8(%rsp), %rdi leaq 0xe0(%rsp), %rsi leaq 0x30(%rsp), %rdx callq 0x2cef5e vcvtsd2ss 0x10(%rsp), %xmm3, %xmm0 vmovss 0xc(%rsp), %xmm2 movl $0x7fffffff, %ebp # imm = 0x7FFFFFFF leaq 0x30(%rsp), %r13 xorl %r15d, %r15d cmpq 0xb0(%r12), %r15 je 0x2d3905 movq 0x18(%r14), %rcx vmovss 0x1c(%rsp), %xmm1 leaq 0xa8(%rsp), %rdi leaq 0xe0(%rsp), %rsi movq %r13, %rdx vmovss %xmm0, 0x10(%rsp) callq 0x2cefa0 movq 0x130(%rsp), %rdx movq 0x158(%rsp), %rcx movl %eax, %ebx leaq 0x170(%rsp), %rdi leaq 0xf0(%rsp), %rsi callq 0x42bbe testl %ebx, %ebx je 0x2d39ad cmpl %ebp, %ebx jge 0x2d389b movq %r14, %rdi movq %r13, %rsi movl %ebx, %edx movq %r15, %rcx callq 0x2d49c4 movl %ebx, %ebp cmpq 0x78(%rsp), %r15 jle 0x2d38d1 vcvtsi2ss %ebx, %xmm3, %xmm0 vcvtsi2ssl 0x168(%rsp), %xmm3, %xmm1 vdivss %xmm1, %xmm0, %xmm0 vmovss 0x48(%rsp), %xmm1 callq 0xb530 vmovss 0x10(%rsp), %xmm1 vfmadd231ss 0x2c(%rsp), %xmm0, %xmm1 # xmm1 = (xmm0 * mem) + xmm1 vmovaps %xmm1, %xmm0 jmp 0x2d38d7 vmovss 0x10(%rsp), %xmm0 vucomiss 0x50(%rsp), %xmm0 ja 0x2d3905 movq %r14, %rdi vmovss %xmm0, 0x10(%rsp) callq 0x2d49fc vmovss 0x10(%rsp), %xmm0 vmovss 0xc(%rsp), %xmm2 testb %al, %al jne 0x2d3905 incq %r15 jmp 0x2d3820 movq 0x20(%rsp), %rax movl $0x5, 0xa0(%rax) cmpl $0x0, 0x50(%r14) jne 0x2d3927 movq 0x20(%rsp), %rax andl $0x0, 0xa0(%rax) cmpq $0x0, 0x28(%r14) je 0x2d3958 movq 0x20(%rsp), %rax leaq 0x68(%rax), %rbx pushq $0x1 popq %rsi movq %rbx, %rdi callq 0x35292 movq (%rbx), %rsi addq $0x20, %r14 movl 0x84(%rsp), %edx movq %r14, %rdi callq 0x35781 leaq 0x170(%rsp), %rdi callq 0x42cbc leaq 0xa8(%rsp), %rdi callq 0x393ac leaq 0xe0(%rsp), %rdi callq 0x2cf400 leaq 0x88(%rsp), %rdi callq 0x4495c leaq 0x38(%rsp), %rdi callq 0x357fe movq 0x20(%rsp), %rax addq $0x3c8, %rsp # imm = 0x3C8 popq %rbx popq %r12 popq %r13 popq %r14 popq %r15 popq %rbp retq vmovsd 0x60(%rsp), %xmm0 movq 0x58(%rsp), %rdi leaq 0x30(%rsp), %rbx movq %rbx, %rsi callq 0x45118 movq %r14, %rdi movq %rbx, %rsi movq %r15, %rdx callq 0x2d497a vmovss 0xc(%rsp), %xmm2 vmovss 0x10(%rsp), %xmm0 leaq 0xe0(%rsp), %r15 leaq 0x30(%rsp), %rbp xorl %ebx, %ebx movl $0x0, 0x40(%rsp) cmpl 0xbc(%r12), %ebx jge 0x2d3914 vmovss %xmm0, 0x10(%rsp) vmulss 0x70(%rsp), %xmm0, %xmm0 movq 0x18(%r14), %rcx vmovss 0x1c(%rsp), %xmm1 vmovss 0x68(%rsp), %xmm3 leaq 0xa8(%rsp), %rdi movq %r15, %rsi movq %rbp, %rdx callq 0x2cf1ac testl %eax, %eax jne 0x2d3a65 vmovsd 0x60(%rsp), %xmm0 movq 0x58(%rsp), %rdi movq %rbp, %rsi callq 0x45118 movl 0xc0(%r12), %eax imull %ebx, %eax notl %eax movslq %eax, %rdx movq %r14, %rdi movq %rbp, %rsi callq 0x2d497a movq %r14, %rdi callq 0x2d49fc vmovss 0xc(%rsp), %xmm2 vmovss 0x10(%rsp), %xmm0 testb %al, %al jne 0x2d3914 movl %ebx, 0x78(%rsp) pushq $0x1 popq %rax movl %eax, %ebx leal -0x1(%rbx), %r13d cmpl 0xc0(%r12), %r13d jge 0x2d3b8c movq 0x18(%r14), %rcx vmovss 0x1c(%rsp), %xmm1 leaq 0xa8(%rsp), %rdi movq %r15, %rsi movq %rbp, %rdx vmovss %xmm0, 0x10(%rsp) callq 0x2cefa0 testl %eax, %eax je 0x2d3b42 vcvtsi2sd %r13d, %xmm4, %xmm0 vucomisd 0xa0(%r12), %xmm0 jbe 0x2d3b09 vcvtsi2ss %eax, %xmm4, %xmm0 vcvtsi2ssl 0x168(%rsp), %xmm4, %xmm1 vdivss %xmm1, %xmm0, %xmm0 vmovss 0x48(%rsp), %xmm1 callq 0xb530 vmovss 0x10(%rsp), %xmm1 vmovss 0xc(%rsp), %xmm2 vfmadd231ss 0x2c(%rsp), %xmm0, %xmm1 # xmm1 = (xmm0 * mem) + xmm1 vmovaps %xmm1, %xmm0 jmp 0x2d3b15 vmovss 0xc(%rsp), %xmm2 vmovss 0x10(%rsp), %xmm0 vucomiss 0x50(%rsp), %xmm0 ja 0x2d3b8c movq %r14, %rdi vmovss %xmm0, 0x10(%rsp) callq 0x2d49fc testb %al, %al jne 0x2d3b80 vmovss 0xc(%rsp), %xmm2 vmovss 0x10(%rsp), %xmm0 incl %ebx jmp 0x2d3a8a vmovsd 0x60(%rsp), %xmm0 movq 0x58(%rsp), %rdi movq %rbp, %rsi callq 0x45118 movl 0xc0(%r12), %eax imull 0x40(%rsp), %eax subl %ebx, %eax movslq %eax, %rdx movq %r14, %rdi movq %rbp, %rsi callq 0x2d497a vmovss 0xc(%rsp), %xmm2 vmovss 0x10(%rsp), %xmm0 jmp 0x2d3b8c vmovss 0xc(%rsp), %xmm2 vmovss 0x10(%rsp), %xmm0 movl 0x78(%rsp), %ebx decl 0x40(%rsp) incl %ebx jmp 0x2d39f6 jmp 0x2d3bd2 jmp 0x2d3bab jmp 0x2d3bd2 movq %rax, %r14 jmp 0x2d3be2 movq %rax, %r14 jmp 0x2d3bef movq %rax, %r14 jmp 0x2d3bfc movq %rax, %r14 jmp 0x2d3c09 movq %rax, %r14 jmp 0x2d3c13 jmp 0x2d3bd2 jmp 0x2d3bd2 jmp 0x2d3bd2 leaq 0x178(%rsp), %rdi movq %rax, %r14 callq 0x357fe jmp 0x2d3be2 leaq 0x170(%rsp), %rdi movq %rax, %r14 callq 0x42cbc leaq 0xa8(%rsp), %rdi callq 0x393ac leaq 0xe0(%rsp), %rdi callq 0x2cf400 leaq 0x88(%rsp), %rdi callq 0x4495c leaq 0x38(%rsp), %rdi callq 0x357fe movq 0x20(%rsp), %rdi callq 0xea60 movq %r14, %rdi callq 0xb3d0 nop
/quesnel[P]baryonyx/lib/src/itm-solver-common.hpp
baryonyx::itm::solver_functor<baryonyx::itm::solver_inequalities_101coeff<float, baryonyx::itm::maximize_tag, baryonyx::itm::quadratic_cost_type<float>, false>, float, baryonyx::itm::maximize_tag, baryonyx::itm::quadratic_cost_type<float>, baryonyx::pnm_observer>::store_if_better(baryonyx::bit_array const&, double, long)
inline bool is_better_solution(Float lhs, Float rhs) noexcept { if constexpr (std::is_same_v<Mode, minimize_tag>) return lhs < rhs; else return lhs > rhs; }
vucomisd 0x30(%rdi), %xmm0 jbe 0x2d49c2 pushq %r14 pushq %rbx pushq %rax movq %rdi, %r14 addq $0x20, %rdi movq %rdx, %rbx vmovsd %xmm0, (%rsp) callq 0x36a44 movq %r14, %rdi callq 0x2d4a3c vmovsd (%rsp), %xmm1 vmovsd %xmm0, 0x38(%r14) movq %rbx, 0x48(%r14) andl $0x0, 0x50(%r14) vmovsd %xmm1, 0x30(%r14) addq $0x8, %rsp popq %rbx popq %r14 retq nop
/quesnel[P]baryonyx/lib/src/itm-common.hpp
baryonyx::itm::solver_functor<baryonyx::itm::solver_inequalities_101coeff<float, baryonyx::itm::maximize_tag, baryonyx::itm::quadratic_cost_type<float>, false>, float, baryonyx::itm::maximize_tag, baryonyx::itm::quadratic_cost_type<float>, baryonyx::pnm_observer>::is_timelimit_reached()
bool is_timelimit_reached() { m_end = std::chrono::steady_clock::now(); return is_time_limit(m_ctx.parameters.time_limit, m_begin, m_end); }
pushq %rbx movq %rdi, %rbx callq 0xb5a0 movq %rax, 0x8(%rbx) vxorpd %xmm2, %xmm2, %xmm2 movq 0x10(%rbx), %rcx subq (%rbx), %rax vmovsd (%rcx), %xmm0 vcvtsi2sd %rax, %xmm1, %xmm1 vdivsd 0x18505b(%rip), %xmm1, %xmm1 # 0x459a80 vcmpnlepd %xmm2, %xmm0, %k1 vcmpltpd %xmm1, %xmm0, %k0 {%k1} kmovd %k0, %eax andb $0x1, %al popq %rbx retq nop
/quesnel[P]baryonyx/lib/src/itm-solver-common.hpp
baryonyx::itm::solver_functor<baryonyx::itm::solver_inequalities_101coeff<float, baryonyx::itm::maximize_tag, baryonyx::itm::quadratic_cost_type<float>, false>, float, baryonyx::itm::maximize_tag, baryonyx::itm::quadratic_cost_type<float>, baryonyx::file_observer>::store_if_better(baryonyx::bit_array const&, double, long)
inline bool is_better_solution(Float lhs, Float rhs) noexcept { if constexpr (std::is_same_v<Mode, minimize_tag>) return lhs < rhs; else return lhs > rhs; }
vucomisd 0x30(%rdi), %xmm0 jbe 0x2d4aa4 pushq %r14 pushq %rbx pushq %rax movq %rdi, %r14 addq $0x20, %rdi movq %rdx, %rbx vmovsd %xmm0, (%rsp) callq 0x36a44 movq %r14, %rdi callq 0x2d4b1e vmovsd (%rsp), %xmm1 vmovsd %xmm0, 0x38(%r14) movq %rbx, 0x48(%r14) andl $0x0, 0x50(%r14) vmovsd %xmm1, 0x30(%r14) addq $0x8, %rsp popq %rbx popq %r14 retq nop
/quesnel[P]baryonyx/lib/src/itm-common.hpp
baryonyx::itm::solver_functor<baryonyx::itm::solver_inequalities_101coeff<float, baryonyx::itm::maximize_tag, baryonyx::itm::quadratic_cost_type<float>, false>, float, baryonyx::itm::maximize_tag, baryonyx::itm::quadratic_cost_type<float>, baryonyx::file_observer>::is_timelimit_reached()
bool is_timelimit_reached() { m_end = std::chrono::steady_clock::now(); return is_time_limit(m_ctx.parameters.time_limit, m_begin, m_end); }
pushq %rbx movq %rdi, %rbx callq 0xb5a0 movq %rax, 0x8(%rbx) vxorpd %xmm2, %xmm2, %xmm2 movq 0x10(%rbx), %rcx subq (%rbx), %rax vmovsd (%rcx), %xmm0 vcvtsi2sd %rax, %xmm1, %xmm1 vdivsd 0x184f79(%rip), %xmm1, %xmm1 # 0x459a80 vcmpnlepd %xmm2, %xmm0, %k1 vcmpltpd %xmm1, %xmm0, %k0 {%k1} kmovd %k0, %eax andb $0x1, %al popq %rbx retq nop
/quesnel[P]baryonyx/lib/src/itm-solver-common.hpp
baryonyx::result baryonyx::itm::optimize_problem<baryonyx::itm::solver_inequalities_101coeff<float, baryonyx::itm::minimize_tag, baryonyx::itm::default_cost_type<float>, true>, float, baryonyx::itm::minimize_tag, baryonyx::itm::default_cost_type<float>>(baryonyx::context const&, baryonyx::problem const&)
inline result optimize_problem(const context& ctx, const problem& pb) { result r; if (ctx.start) ctx.start(ctx.parameters); auto constraints{ make_merged_constraints(ctx, pb) }; if (constraints.empty() || pb.vars.values.empty()) { r.status = result_status::success; r.solutions.resize(1); r.solutions.back().value = pb.objective.value; r.strings = pb.strings; r.affected_vars = pb.affected_vars; r.variable_name = pb.vars.names; return r; } random_engine rng(init_random_generator_seed(ctx)); auto variables = length(pb.vars.values); auto cost = Cost(pb.objective, variables); auto cost_constant = pb.objective.value; const auto thread = get_thread_number(ctx); std::vector<optimize_functor<Solver, Float, Mode, Cost>> functors; std::vector<std::thread> pool(thread); best_solution_recorder<Cost, Float, Mode> best_recorder( rng, thread, cost, cost_constant, variables, constraints, ctx.parameters.init_population_size); auto seeds = generate_seed(rng, thread); std::atomic_bool stop_task; stop_task.store(false); for (unsigned i = 0u; i != thread; ++i) functors.emplace_back(ctx, i, seeds[i]); for (unsigned i = 0u; i != thread; ++i) pool[i] = std::thread(std::ref(functors[i]), std::ref(stop_task), std::ref(best_recorder), std::cref(constraints), variables, std::cref(cost), cost_constant); const auto start = std::chrono::steady_clock::now(); auto end = start; do { std::this_thread::sleep_for(std::chrono::seconds{ 1L }); if (ctx.update) { auto call_number = 0L; for (auto i = 0u; i != thread; ++i) call_number += functors[i].m_call_number; int constraints_remaining; long int loop; double value; double duration; best_recorder.get_best( constraints_remaining, value, duration, loop); ctx.update( constraints_remaining, value, loop, duration, call_number); } end = std::chrono::steady_clock::now(); } while (!is_time_limit(ctx.parameters.time_limit, start, end)); stop_task.store(true); for (auto& t : pool) t.join(); r.strings = pb.strings; r.affected_vars = pb.affected_vars; r.variable_name = pb.vars.names; r.variables = variables; r.constraints = length(constraints); const auto& first = best_recorder.get_best(0); if (!first.is_solution()) r.status = result_status::time_limit_reached; else r.status = result_status::success; r.duration = first.duration; r.loop = first.loop; r.remaining_constraints = first.remaining_constraints; switch (ctx.parameters.storage) { case solver_parameters::storage_type::one: { r.solutions.resize(1); convert(first, r.solutions[0], variables); } break; case solver_parameters::storage_type::bound: { r.solutions.resize(2); convert(first, r.solutions[0], variables); convert(best_recorder.get_worst(), r.solutions[1], variables); } break; case solver_parameters::storage_type::five: { r.solutions.resize(5); for (int i = 0; i != 5; ++i) convert(best_recorder.get_best(i), r.solutions[i], variables); } break; } best_recorder.show_population(ctx); if (ctx.finish) ctx.finish(r); return r; }
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x188, %rsp # imm = 0x188 vxorpd %xmm0, %xmm0, %xmm0 movq %rdx, %r13 movq %rsi, %r15 movq %rdi, %r12 vmovupd %zmm0, 0x58(%rdi) vmovupd %zmm0, 0x40(%rdi) vmovupd %zmm0, (%rdi) movq $0x7fffffff, 0x98(%rdi) # imm = 0x7FFFFFFF movl $0x2, 0xa0(%rdi) cmpq $0x0, 0x128(%rsi) je 0x2d4c85 leaq 0x118(%r15), %rdi movq %r15, %rsi vzeroupper callq 0x3527e leaq 0x48(%rsp), %rdi movq %r15, %rsi movq %r13, %rdx vzeroupper callq 0x450078 movq 0x48(%rsp), %rax cmpq 0x50(%rsp), %rax je 0x2d4de1 movq 0xa8(%r13), %rax cmpq 0xb0(%r13), %rax je 0x2d4de1 movq %r15, %rdi callq 0x353ba movl $0x7fffffff, %ecx # imm = 0x7FFFFFFF xorl %edx, %edx divq %rcx cmpq $0x1, %rdx adcq $0x0, %rdx movq %rdx, 0x40(%rsp) movq 0xb0(%r13), %rax subq 0xa8(%r13), %rax pushq $0xc popq %rcx cqto leaq 0x10(%r13), %rsi idivq %rcx movl %eax, 0x4(%rsp) leaq 0x60(%rsp), %rdi movl %eax, %edx movq %r12, 0x10(%rsp) callq 0x353d2 vmovsd 0x40(%r13), %xmm0 vmovsd %xmm0, 0x38(%rsp) movq %r15, %rdi callq 0x35447 andq $0x0, 0x80(%rsp) vxorpd %xmm0, %xmm0, %xmm0 movl %eax, %ebp vmovapd %xmm0, 0x70(%rsp) leaq 0xb0(%rsp), %rdi leaq 0xc8(%rsp), %rdx movq %rbp, %rsi callq 0x35462 vmovsd 0x38(%rsp), %xmm0 movl 0xc4(%r15), %eax movl 0x4(%rsp), %r8d subq $0x8, %rsp movl %ebp, %edx leaq 0xd0(%rsp), %rdi leaq 0x48(%rsp), %rsi leaq 0x68(%rsp), %rcx leaq 0x50(%rsp), %r9 pushq %rax callq 0x4f322 popq %rax popq %rcx leaq 0x30(%rsp), %rdi leaq 0x40(%rsp), %rsi movl %ebp, %edx callq 0x3551c xorl %eax, %eax movb %al, 0x3(%rsp) xorl %ecx, %ecx leaq 0x70(%rsp), %rbx leaq 0x8(%rsp), %r14 xchgb %cl, 0x3(%rsp) leaq 0x90(%r13), %rcx movq %rcx, 0x98(%rsp) movl %eax, 0x8(%rsp) cmpl %ebp, %eax je 0x2d4e3c movl %eax, %ecx shlq $0x3, %rcx addq 0x30(%rsp), %rcx movq %rbx, %rdi movq %r15, %rsi movq %r14, %rdx callq 0x2d5f0a movl 0x8(%rsp), %eax incl %eax jmp 0x2d4db8 andl $0x0, 0xa0(%r12) leaq 0x68(%r12), %rdi pushq $0x1 popq %rsi callq 0x35292 vmovsd 0x40(%r13), %xmm0 movq 0x70(%r12), %rax vmovsd %xmm0, -0x8(%rax) movq %r12, %rdi movq %r13, %rsi callq 0x31d28 leaq 0xc0(%r13), %rsi leaq 0x20(%r12), %rdi callq 0x352c2 addq $0x90, %r13 leaq 0x50(%r12), %rdi movq %r13, %rsi callq 0x352ec jmp 0x2d51d1 movq %r13, 0x90(%rsp) movq %r15, %r13 leaq 0x8(%rsp), %r15 leaq 0xa8(%rsp), %r14 leaq 0x4(%rsp), %rbx xorl %r12d, %r12d cmpq %rbp, %r12 je 0x2d4efd imulq $0xc0, %r12, %rax addq 0x70(%rsp), %rax leaq 0x3(%rsp), %r10 leaq 0xc8(%rsp), %r11 movq %r10, 0x20(%rsp) leaq 0x48(%rsp), %r10 movq %r11, 0x18(%rsp) movq %r10, 0xa8(%rsp) movq %rax, 0x28(%rsp) leaq 0x60(%rsp), %rax movq %rax, 0xa0(%rsp) leaq 0x20(%rsp), %rdx leaq 0x38(%rsp), %rax movq %r15, %rdi leaq 0x28(%rsp), %rsi leaq 0x18(%rsp), %rcx movq %r14, %r8 movq %rbx, %r9 pushq %rax leaq 0xa8(%rsp), %rax pushq %rax callq 0x2d5f56 popq %rax popq %rcx leaq (,%r12,8), %rdi addq 0xb0(%rsp), %rdi movq %r15, %rsi callq 0x356d6 movq %r15, %rdi callq 0x356f0 incq %r12 jmp 0x2d4e5c callq 0xb5a0 imulq $0xc0, %rbp, %r15 leaq 0x138(%r13), %r12 leaq 0x8(%rsp), %rbp movq %rax, %rbx movq $0x1, 0x8(%rsp) movq %rbp, %rdi callq 0x356fd cmpq $0x0, 0x148(%r13) je 0x2d4f92 movq 0x70(%rsp), %rax xorl %ecx, %ecx xorl %r14d, %r14d cmpq %rcx, %r15 je 0x2d4f53 addq 0xb0(%rax,%rcx), %r14 addq $0xc0, %rcx jmp 0x2d4f3d leaq 0xd0(%rsp), %rdi leaq 0x18(%rsp), %rsi leaq 0x28(%rsp), %rdx leaq 0x20(%rsp), %rcx movq %rbp, %r8 callq 0x56bc2 vmovsd 0x28(%rsp), %xmm0 vmovsd 0x20(%rsp), %xmm1 movq 0x8(%rsp), %rdx movl 0x18(%rsp), %esi movq %r12, %rdi movq %r14, %rcx callq 0x35736 callq 0xb5a0 vmovsd (%r13), %xmm0 subq %rbx, %rax vxorpd %xmm1, %xmm1, %xmm1 vucomisd %xmm0, %xmm1 vcvtsi2sd %rax, %xmm2, %xmm1 vdivsd 0x184acb(%rip), %xmm1, %xmm1 # 0x459a80 jae 0x2d4f18 vucomisd %xmm0, %xmm1 jbe 0x2d4f18 movb $0x1, %al xchgb %al, 0x3(%rsp) movq 0xb0(%rsp), %rbx movq 0xb8(%rsp), %r14 movq 0x10(%rsp), %r12 cmpq %r14, %rbx je 0x2d4ff3 movq %rbx, %rdi callq 0xb1a0 addq $0x8, %rbx jmp 0x2d4fe0 movq 0x90(%rsp), %rbx movq %r12, %rdi movq %rbx, %rsi callq 0x31d28 addq $0xc0, %rbx leaq 0x20(%r12), %rdi movq %rbx, %rsi callq 0x352c2 leaq 0x50(%r12), %rdi movq 0x98(%rsp), %rsi callq 0x352ec movl 0x4(%rsp), %eax movl %eax, 0x90(%r12) movq 0x50(%rsp), %rax subq 0x48(%rsp), %rax pushq $0x28 popq %rcx cqto xorl %esi, %esi idivq %rcx movq 0x108(%rsp), %rdi movq 0x120(%rsp), %rcx movl %eax, 0x94(%r12) movslq (%rdi), %rax imulq $0x38, %rax, %rax movl 0x30(%rcx,%rax), %edx leaq (%rcx,%rax), %rbx testl %edx, %edx setne %sil leal (%rsi,%rsi,2), %esi movl %esi, 0xa0(%r12) vmovsd 0x18(%rcx,%rax), %xmm0 vmovsd %xmm0, 0x80(%r12) movq 0x28(%rcx,%rax), %rax movq %rax, 0x88(%r12) movl %edx, 0x98(%r12) movl 0xe8(%r13), %eax testl %eax, %eax je 0x2d514e cmpl $0x1, %eax je 0x2d5116 cmpl $0x2, %eax jne 0x2d516d leaq 0x68(%r12), %rbx pushq $0x5 popq %rsi movq %rbx, %rdi callq 0x35292 xorl %r14d, %r14d xorl %r15d, %r15d cmpq $0x5, %r15 je 0x2d516d movq 0x108(%rsp), %rax movq (%rbx), %rsi movl 0x4(%rsp), %edx movslq (%rax,%r15,4), %rax addq %r14, %rsi imulq $0x38, %rax, %rdi addq 0x120(%rsp), %rdi callq 0x4f485 incq %r15 addq $0x20, %r14 jmp 0x2d50dc leaq 0x68(%r12), %r14 pushq $0x2 popq %rsi movq %r14, %rdi callq 0x35292 movl 0x4(%rsp), %edx movq (%r14), %rsi movq %rbx, %rdi callq 0x4f485 leaq 0xd0(%rsp), %rdi callq 0x56c40 movq (%r14), %rsi movq %rax, %rbx addq $0x20, %rsi jmp 0x2d5161 leaq 0x68(%r12), %r14 pushq $0x1 popq %rsi movq %r14, %rdi callq 0x35292 movq (%r14), %rsi movl 0x4(%rsp), %edx movq %rbx, %rdi callq 0x4f485 leaq 0xc8(%rsp), %rdi movq %r13, %rsi callq 0x4f4da cmpq $0x0, 0x168(%r13) je 0x2d5199 addq $0x158, %r13 # imm = 0x158 movq %r13, %rdi movq %r12, %rsi callq 0x357ea leaq 0x30(%rsp), %rdi callq 0x357fe leaq 0xc8(%rsp), %rdi callq 0x4f4ee leaq 0xb0(%rsp), %rdi callq 0x35836 leaq 0x70(%rsp), %rdi callq 0x2db776 leaq 0x68(%rsp), %rdi callq 0x35c1a leaq 0x48(%rsp), %rdi callq 0x35858 movq %r12, %rax addq $0x188, %rsp # imm = 0x188 popq %rbx popq %r12 popq %r13 popq %r14 popq %r15 popq %rbp retq movq %rax, %rbx jmp 0x2d5232 movq %rax, %rbx jmp 0x2d523f movq %rax, %rbx jmp 0x2d524c jmp 0x2d5216 jmp 0x2d5207 jmp 0x2d5225 jmp 0x2d5225 movq %r12, 0x10(%rsp) movq %rax, %rbx jmp 0x2d526a movq %r12, 0x10(%rsp) movq %rax, %rbx jmp 0x2d5260 jmp 0x2d5225 jmp 0x2d5225 jmp 0x2d5225 jmp 0x2d5225 jmp 0x2d5225 movq %rax, %rbx leaq 0x30(%rsp), %rdi callq 0x357fe leaq 0xc8(%rsp), %rdi callq 0x4f4ee leaq 0xb0(%rsp), %rdi callq 0x35836 leaq 0x70(%rsp), %rdi callq 0x2db776 leaq 0x68(%rsp), %rdi callq 0x35c1a leaq 0x48(%rsp), %rdi callq 0x35858 movq 0x10(%rsp), %rdi callq 0xea60 movq %rbx, %rdi callq 0xb3d0
/quesnel[P]baryonyx/lib/src/itm-optimizer-common.hpp
baryonyx::result baryonyx::itm::solve_problem<baryonyx::itm::solver_inequalities_101coeff<float, baryonyx::itm::minimize_tag, baryonyx::itm::default_cost_type<float>, false>, float, baryonyx::itm::minimize_tag, baryonyx::itm::default_cost_type<float>>(baryonyx::context const&, baryonyx::problem const&)
inline result solve_problem(const context& ctx, const problem& pb) { if (ctx.start) ctx.start(ctx.parameters); result ret; auto variables = length(pb.vars.values); auto constraints{ make_merged_constraints(ctx, pb) }; if (!constraints.empty() && !pb.vars.values.empty()) { random_engine rng(init_random_generator_seed(ctx)); auto cost = Cost(pb.objective, variables); auto cost_constant = pb.objective.value; switch (ctx.parameters.observer) { case solver_parameters::observer_type::pnm: { using obs = pnm_observer; solver_functor<Solver, Float, Mode, Cost, obs> slv(ctx, rng); ret = slv(constraints, variables, cost, cost_constant); } break; case solver_parameters::observer_type::file: { using obs = file_observer; solver_functor<Solver, Float, Mode, Cost, obs> slv(ctx, rng); ret = slv(constraints, variables, cost, cost_constant); } break; default: { using obs = none_observer; solver_functor<Solver, Float, Mode, Cost, obs> slv(ctx, rng); ret = slv(constraints, variables, cost, cost_constant); break; } } } else { ret.status = result_status::success; ret.solutions.resize(1); ret.solutions.back().value = pb.objective.value; } ret.strings = pb.strings; ret.variable_name = std::move(pb.vars.names); ret.affected_vars = std::move(pb.affected_vars); ret.variables = variables; ret.constraints = length(constraints); if (ctx.finish) ctx.finish(ret); return ret; }
pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x130, %rsp # imm = 0x130 cmpq $0x0, 0x128(%rsi) movq %rdx, %r15 movq %rsi, %r14 movq %rdi, %rbx je 0x2d5c23 leaq 0x118(%r14), %rdi movq %r14, %rsi callq 0x3527e vxorps %xmm0, %xmm0, %xmm0 vmovups %zmm0, 0x58(%rbx) vmovups %zmm0, 0x40(%rbx) vmovups %zmm0, (%rbx) movq $0x7fffffff, 0x98(%rbx) # imm = 0x7FFFFFFF movl $0x2, 0xa0(%rbx) movq 0xb0(%r15), %rax subq 0xa8(%r15), %rax pushq $0xc popq %rcx cqto idivq %rcx movq %rax, %r12 leaq 0x8(%rsp), %rdi movq %r14, %rsi movq %r15, %rdx vzeroupper callq 0x450078 movq 0x8(%rsp), %rax cmpq 0x10(%rsp), %rax je 0x2d5d4c movq 0xa8(%r15), %rax cmpq 0xb0(%r15), %rax je 0x2d5d4c movq %r14, %rdi callq 0x353ba movl $0x7fffffff, %ecx # imm = 0x7FFFFFFF xorl %edx, %edx leaq 0x10(%r15), %rsi divq %rcx cmpq $0x1, %rdx adcq $0x0, %rdx movq %rdx, (%rsp) leaq 0x20(%rsp), %rdi movl %r12d, %edx callq 0x353d2 vmovsd 0x40(%r15), %xmm0 movl 0xe4(%r14), %eax vxorps %xmm1, %xmm1, %xmm1 leaq 0x30(%rsp), %rsi cmpl $0x1, %eax je 0x2d5d73 cmpl $0x2, %eax jne 0x2d5dc2 movabsq $0x7ff0000000000000, %r9 # imm = 0x7FF0000000000000 vmovaps %xmm1, (%rsi) movq %r14, 0x10(%rsi) movq %rsp, %rax movq %rax, 0x18(%rsi) vmovaps %xmm1, 0x20(%rsi) movq %r9, 0x30(%rsi) andq $0x0, 0x48(%rsi) vmovups %xmm1, 0x38(%rsi) movl $0x7fffffff, 0x50(%rsi) # imm = 0x7FFFFFFF leaq 0x88(%rsp), %rdi leaq 0x8(%rsp), %rdx leaq 0x20(%rsp), %r8 movl %r12d, %ecx callq 0x2e26a8 jmp 0x2d5e0f andl $0x0, 0xa0(%rbx) leaq 0x68(%rbx), %rdi pushq $0x1 popq %rsi callq 0x35292 vmovsd 0x40(%r15), %xmm0 movq 0x70(%rbx), %rax vmovsd %xmm0, -0x8(%rax) jmp 0x2d5e3e movabsq $0x7ff0000000000000, %r9 # imm = 0x7FF0000000000000 vmovaps %xmm1, (%rsi) movq %r14, 0x10(%rsi) movq %rsp, %rax movq %rax, 0x18(%rsi) vmovaps %xmm1, 0x20(%rsi) movq %r9, 0x30(%rsi) andq $0x0, 0x48(%rsi) vmovups %xmm1, 0x38(%rsi) movl $0x7fffffff, 0x50(%rsi) # imm = 0x7FFFFFFF leaq 0x88(%rsp), %rdi leaq 0x8(%rsp), %rdx leaq 0x20(%rsp), %r8 movl %r12d, %ecx callq 0x2e1fc4 jmp 0x2d5e0f movabsq $0x7ff0000000000000, %r9 # imm = 0x7FF0000000000000 vmovaps %xmm1, (%rsi) movq %r14, 0x10(%rsi) movq %rsp, %rax movq %rax, 0x18(%rsi) vmovaps %xmm1, 0x20(%rsi) movq %r9, 0x30(%rsi) andq $0x0, 0x48(%rsi) vmovups %xmm1, 0x38(%rsi) movl $0x7fffffff, 0x50(%rsi) # imm = 0x7FFFFFFF leaq 0x88(%rsp), %rdi leaq 0x8(%rsp), %rdx leaq 0x20(%rsp), %r8 movl %r12d, %ecx callq 0x2e2d8c leaq 0x88(%rsp), %r13 movq %rbx, %rdi movq %r13, %rsi callq 0x41d9e movq %r13, %rdi callq 0xea60 leaq 0x58(%rsp), %rdi callq 0x357fe leaq 0x28(%rsp), %rdi callq 0x35c1a leaq 0x90(%r15), %r13 movq %rbx, %rdi movq %r15, %rsi callq 0x31d28 leaq 0x50(%rbx), %rdi movq %r13, %rsi callq 0x352ec addq $0xc0, %r15 leaq 0x20(%rbx), %rdi movq %r15, %rsi callq 0x352c2 movq 0x10(%rsp), %rax movl %r12d, 0x90(%rbx) subq 0x8(%rsp), %rax pushq $0x28 popq %rcx cqto idivq %rcx movl %eax, 0x94(%rbx) cmpq $0x0, 0x168(%r14) je 0x2d5eaa addq $0x158, %r14 # imm = 0x158 movq %r14, %rdi movq %rbx, %rsi callq 0x357ea leaq 0x8(%rsp), %rdi callq 0x35858 movq %rbx, %rax addq $0x130, %rsp # imm = 0x130 popq %rbx popq %r12 popq %r13 popq %r14 popq %r15 retq jmp 0x2d5ecc jmp 0x2d5ecc leaq 0x58(%rsp), %rdi movq %rax, %r14 callq 0x357fe leaq 0x28(%rsp), %rdi callq 0x35c1a jmp 0x2d5eef jmp 0x2d5eec movq %rax, %r14 jmp 0x2d5ef9 movq %rax, %r14 leaq 0x8(%rsp), %rdi callq 0x35858 movq %rbx, %rdi callq 0xea60 movq %r14, %rdi callq 0xb3d0 nop
/quesnel[P]baryonyx/lib/src/itm-solver-common.hpp
void baryonyx::itm::compute_order::init<baryonyx::itm::solver_inequalities_101coeff<float, baryonyx::itm::minimize_tag, baryonyx::itm::default_cost_type<float>, true>, baryonyx::bit_array>(baryonyx::itm::solver_inequalities_101coeff<float, baryonyx::itm::minimize_tag, baryonyx::itm::default_cost_type<float>, true> const&, baryonyx::bit_array const&)
void init(const Solver& s, const Xtype& x) { switch (order) { case solver_parameters::constraint_order::infeasibility_decr: case solver_parameters::constraint_order::infeasibility_incr: infeasibility_local_compute_violated_constraints(s, x); break; case solver_parameters::constraint_order::pi_sign_change: std::iota(R.begin(), R.end(), 0); break; case solver_parameters::constraint_order::none: case solver_parameters::constraint_order::reversing: case solver_parameters::constraint_order::random_sorting: case solver_parameters::constraint_order::lagrangian_decr: case solver_parameters::constraint_order::lagrangian_incr: default: compute_violated_constraints(s, x, R); break; } }
movl 0x30(%rdi), %ecx movq %rdi, %rax leal -0x3(%rcx), %edi cmpl $0x2, %edi jae 0x2d6b0e movq %rax, %rdi jmp 0x2d70a8 cmpl $0x7, %ecx jne 0x2d6b2b movq (%rax), %rcx movq 0x8(%rax), %rax xorl %edx, %edx cmpq %rax, %rcx je 0x2d6b39 movl %edx, (%rcx) incl %edx addq $0x4, %rcx jmp 0x2d6b1c movq %rsi, %rdi movq %rdx, %rsi movq %rax, %rdx jmp 0x2d71c3 retq
/quesnel[P]baryonyx/lib/src/itm-common.hpp
baryonyx::itm::solver_functor<baryonyx::itm::solver_inequalities_101coeff<float, baryonyx::itm::minimize_tag, baryonyx::itm::default_cost_type<float>, true>, float, baryonyx::itm::minimize_tag, baryonyx::itm::default_cost_type<float>, baryonyx::file_observer>::operator()(std::vector<baryonyx::itm::merged_constraint, std::allocator<baryonyx::itm::merged_constraint>> const&, int, baryonyx::itm::default_cost_type<float> const&, double)
result operator()(const std::vector<merged_constraint>& constraints, int variables, const Cost& original_costs, double cost_constant) { result r; bit_array x(variables); int best_remaining = INT_MAX; auto& p = m_ctx.parameters; auto norm_costs = normalize_costs<Float, Cost>( m_ctx, original_costs, m_rng, variables); const auto kappa_step = static_cast<Float>(p.kappa_step); const auto kappa_max = static_cast<Float>(p.kappa_max); const auto alpha = static_cast<Float>(p.alpha); const auto theta = static_cast<Float>(p.theta); const auto delta = p.delta < 0 ? compute_delta<Float, Cost>(m_ctx, norm_costs, theta, variables) : static_cast<Float>(p.delta); const auto pushing_k_factor = static_cast<Float>(p.pushing_k_factor); const auto pushing_objective_amplifier = static_cast<Float>(p.pushing_objective_amplifier); const long int w_limit = static_cast<long int>(p.w); Solver slv( m_rng, length(constraints), variables, norm_costs, constraints); compute_order compute(p.order, variables); { std::bernoulli_distribution choose_mutation( m_ctx.parameters.init_policy_random); bit_array empty_x; switch (p.init_policy) { case solver_parameters::init_policy_type::pessimistic_solve: init_with_pre_solve<Cost, Mode>( x, empty_x, m_rng, original_costs, constraints, 1.0); break; case solver_parameters::init_policy_type::optimistic_solve: init_with_pre_solve<Cost, Mode>( empty_x, x, m_rng, original_costs, constraints, 1.0); break; case solver_parameters::init_policy_type::bastert: init_with_bastert<Cost, Mode>(x, original_costs, variables, 0); break; } for (int i = 0, e = x.size(); i != e; ++i) if (choose_mutation(m_rng)) x.invert(i); } bool start_push = false; auto kappa = static_cast<Float>(p.kappa_min); Observer obs("img", slv.m, slv.n, p.limit); m_begin = std::chrono::steady_clock::now(); m_end = std::chrono::steady_clock::now(); compute.init(slv, x); for (long int i = 0; i != p.limit; ++i) { auto remaining = compute.run(slv, x, m_rng, kappa, delta, theta); obs.make_observation(slv.ap, slv.P.get(), slv.pi.get()); if (remaining == 0) { store_if_better( x, original_costs.results(x, cost_constant), i); best_remaining = remaining; start_push = true; break; } if (remaining < best_remaining) { store_if_better(x, remaining, i); best_remaining = remaining; } if (i > w_limit) kappa += kappa_step * std::pow(static_cast<Float>(remaining) / static_cast<Float>(slv.m), alpha); if (kappa > kappa_max) { r.status = result_status::kappa_max_reached; break; } if (is_timelimit_reached()) { r.status = result_status::time_limit_reached; break; } } if (!start_push) { r.status = result_status::limit_reached; } else { for (int push = 0; push < p.pushes_limit; ++push) { auto remaining = compute.push_and_run(slv, x, m_rng, pushing_k_factor * kappa, delta, theta, pushing_objective_amplifier); if (remaining == 0) store_if_better(x, original_costs.results(x, cost_constant), -push * p.pushing_iteration_limit - 1); if (is_timelimit_reached()) break; for (int iter = 0; iter < p.pushing_iteration_limit; ++iter) { remaining = compute.run(slv, x, m_rng, kappa, delta, theta); if (remaining == 0) { store_if_better( x, original_costs.results(x, cost_constant), -push * p.pushing_iteration_limit - iter - 1); break; } if (iter > p.w) kappa += kappa_step * std::pow(static_cast<Float>(remaining) / static_cast<Float>(slv.m), alpha); if (kappa > kappa_max) break; if (is_timelimit_reached()) break; } } } if (m_best.remaining_constraints == 0) r.status = result_status::success; if (!m_best.x.empty()) { r.solutions.resize(1); convert(m_best, r.solutions[0], variables); } return r; }
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x3b8, %rsp # imm = 0x3B8 vmovsd %xmm0, 0x60(%rsp) vxorps %xmm0, %xmm0, %xmm0 movq %r8, %r13 movl %ecx, %ebx movq %rdx, %r15 movq %rsi, %r14 movq %rdi, 0x20(%rsp) vmovups %zmm0, 0x58(%rdi) vmovups %zmm0, 0x40(%rdi) vmovups %zmm0, (%rdi) movq $0x7fffffff, 0x98(%rdi) # imm = 0x7FFFFFFF movl $0x2, 0xa0(%rdi) leaq 0x30(%rsp), %rdi movl %ecx, %esi vzeroupper callq 0x368ae movq 0x10(%r14), %r12 movq 0x18(%r14), %rcx leaq 0x88(%rsp), %rdi movq %r12, %rsi movq %r13, %rdx movl %ebx, %r8d callq 0x38959 vmovsd 0x20(%r12), %xmm0 vmovsd 0x10(%r12), %xmm2 vxorpd %xmm1, %xmm1, %xmm1 vmovsd %xmm0, 0x50(%rsp) vmovsd 0x28(%r12), %xmm0 vucomisd %xmm2, %xmm1 vmovsd %xmm0, 0x48(%rsp) vmovsd 0x30(%r12), %xmm0 vmovsd %xmm0, 0x40(%rsp) vcvtsd2ss 0x8(%r12), %xmm3, %xmm0 vmovss %xmm0, 0xc(%rsp) jbe 0x2dbf66 movq 0x10(%r14), %rdi leaq 0x88(%rsp), %rsi movl %ebx, %edx callq 0x38a37 vmovss %xmm0, 0x1c(%rsp) jmp 0x2dbf70 vcvtsd2ss %xmm2, %xmm2, %xmm0 vmovss %xmm0, 0x1c(%rsp) vcvttsd2si 0xa0(%r12), %rax vmovsd 0x38(%r12), %xmm0 vmovsd 0x40(%r12), %xmm1 movq 0x18(%r14), %rsi movq %rax, 0x78(%rsp) movq 0x8(%r15), %rax vmovsd %xmm0, 0x70(%rsp) vmovsd %xmm1, 0x68(%rsp) subq (%r15), %rax pushq $0x28 popq %rcx cqto idivq %rcx leaq 0xd0(%rsp), %rdi leaq 0x88(%rsp), %r8 movl %eax, %edx movl %ebx, %ecx movq %r15, %r9 callq 0x2d6834 movl 0xcc(%r12), %esi leaq 0x98(%rsp), %rdi movl %ebx, %edx callq 0x38cd8 movq 0x10(%r14), %rax movl %ebx, 0x84(%rsp) vmovsd 0x48(%rax), %xmm0 vmovsd %xmm0, 0x10(%rsp) vxorps %xmm0, %xmm0, %xmm0 vmovaps %xmm0, 0x160(%rsp) movl 0xd4(%r12), %eax testl %eax, %eax je 0x2dc051 cmpl $0x1, %eax je 0x2dc02b cmpl $0x2, %eax jne 0x2dc062 movq 0x18(%r14), %rdx leaq 0x160(%rsp), %rdi leaq 0x30(%rsp), %rsi jmp 0x2dc03c movq 0x18(%r14), %rdx leaq 0x30(%rsp), %rdi leaq 0x160(%rsp), %rsi vmovsd 0x177ff4(%rip), %xmm0 # 0x454038 movq %r13, %rcx movq %r15, %r8 callq 0x4f9c9 jmp 0x2dc062 leaq 0x30(%rsp), %rdi movq %r13, %rsi movl %ebx, %edx xorl %ecx, %ecx callq 0x4f95d vcvtsd2ss 0x50(%rsp), %xmm3, %xmm0 leaq 0x30(%rsp), %r15 movq %r13, 0x58(%rsp) xorl %ebx, %ebx movl (%r15), %ebp vmovss %xmm0, 0x2c(%rsp) vcvtsd2ss 0x48(%rsp), %xmm3, %xmm0 vmovss %xmm0, 0x50(%rsp) vcvtsd2ss 0x40(%rsp), %xmm3, %xmm0 vmovss %xmm0, 0x48(%rsp) vcvtsd2ss 0x70(%rsp), %xmm3, %xmm0 vmovss %xmm0, 0x70(%rsp) vcvtsd2ss 0x68(%rsp), %xmm3, %xmm0 vmovss %xmm0, 0x68(%rsp) cmpl %ebx, %ebp je 0x2dc0d4 movq 0x18(%r14), %rdi callq 0x36ac3 vmovsd 0x10(%rsp), %xmm1 vucomisd %xmm0, %xmm1 jbe 0x2dc0d0 movq %r15, %rdi movl %ebx, %esi callq 0x3626e incl %ebx jmp 0x2dc0ad leaq 0x168(%rsp), %rdi callq 0x357fe vmovsd 0x18(%r12), %xmm0 movl 0x158(%rsp), %ecx movl 0x15c(%rsp), %r8d movq 0xb0(%r12), %r9 vmovsd %xmm0, 0x10(%rsp) leaq 0x17e46b(%rip), %rdx # 0x45a577 leaq 0x160(%rsp), %rdi pushq $0x3 popq %rsi callq 0x4346e callq 0xb5a0 movq %rax, (%r14) callq 0xb5a0 movq %rax, 0x8(%r14) leaq 0x98(%rsp), %rdi leaq 0xd0(%rsp), %rsi leaq 0x30(%rsp), %rdx callq 0x2d6af8 vcvtsd2ss 0x10(%rsp), %xmm3, %xmm0 vmovss 0xc(%rsp), %xmm2 movl $0x7fffffff, %ebp # imm = 0x7FFFFFFF leaq 0x30(%rsp), %r13 xorl %r15d, %r15d cmpq 0xb0(%r12), %r15 je 0x2dc245 movq 0x18(%r14), %rcx vmovss 0x1c(%rsp), %xmm1 leaq 0x98(%rsp), %rdi leaq 0xd0(%rsp), %rsi movq %r13, %rdx vmovss %xmm0, 0x10(%rsp) callq 0x2d6b3a movq 0x120(%rsp), %rdx movq 0x148(%rsp), %rcx movl %eax, %ebx leaq 0x160(%rsp), %rdi leaq 0xe0(%rsp), %rsi callq 0x434d2 testl %ebx, %ebx je 0x2dc2ed cmpl %ebp, %ebx jge 0x2dc1db movq %r14, %rdi movq %r13, %rsi movl %ebx, %edx movq %r15, %rcx callq 0x2dcd0a movl %ebx, %ebp cmpq 0x78(%rsp), %r15 jle 0x2dc211 vcvtsi2ss %ebx, %xmm3, %xmm0 vcvtsi2ssl 0x158(%rsp), %xmm3, %xmm1 vdivss %xmm1, %xmm0, %xmm0 vmovss 0x48(%rsp), %xmm1 callq 0xb530 vmovss 0x10(%rsp), %xmm1 vfmadd231ss 0x2c(%rsp), %xmm0, %xmm1 # xmm1 = (xmm0 * mem) + xmm1 vmovaps %xmm1, %xmm0 jmp 0x2dc217 vmovss 0x10(%rsp), %xmm0 vucomiss 0x50(%rsp), %xmm0 ja 0x2dc245 movq %r14, %rdi vmovss %xmm0, 0x10(%rsp) callq 0x2dcd42 vmovss 0x10(%rsp), %xmm0 vmovss 0xc(%rsp), %xmm2 testb %al, %al jne 0x2dc245 incq %r15 jmp 0x2dc160 movq 0x20(%rsp), %rax movl $0x5, 0xa0(%rax) cmpl $0x0, 0x50(%r14) jne 0x2dc267 movq 0x20(%rsp), %rax andl $0x0, 0xa0(%rax) cmpq $0x0, 0x28(%r14) je 0x2dc298 movq 0x20(%rsp), %rax leaq 0x68(%rax), %rbx pushq $0x1 popq %rsi movq %rbx, %rdi callq 0x35292 movq (%rbx), %rsi addq $0x20, %r14 movl 0x84(%rsp), %edx movq %r14, %rdi callq 0x4f485 leaq 0x160(%rsp), %rdi callq 0x435d0 leaq 0x98(%rsp), %rdi callq 0x393ac leaq 0xd0(%rsp), %rdi callq 0x2d6f9a leaq 0x90(%rsp), %rdi callq 0x35c1a leaq 0x38(%rsp), %rdi callq 0x357fe movq 0x20(%rsp), %rax addq $0x3b8, %rsp # imm = 0x3B8 popq %rbx popq %r12 popq %r13 popq %r14 popq %r15 popq %rbp retq vmovsd 0x60(%rsp), %xmm0 movq 0x58(%rsp), %rdi leaq 0x30(%rsp), %rbx movq %rbx, %rsi callq 0x365d4 movq %r14, %rdi movq %rbx, %rsi movq %r15, %rdx callq 0x2dccbc vmovss 0xc(%rsp), %xmm2 vmovss 0x10(%rsp), %xmm0 leaq 0xd0(%rsp), %r15 leaq 0x30(%rsp), %rbp xorl %ebx, %ebx movl $0x0, 0x40(%rsp) cmpl 0xbc(%r12), %ebx jge 0x2dc254 vmovss %xmm0, 0x10(%rsp) vmulss 0x70(%rsp), %xmm0, %xmm0 movq 0x18(%r14), %rcx vmovss 0x1c(%rsp), %xmm1 vmovss 0x68(%rsp), %xmm3 leaq 0x98(%rsp), %rdi movq %r15, %rsi movq %rbp, %rdx callq 0x2d6d46 testl %eax, %eax jne 0x2dc3a5 vmovsd 0x60(%rsp), %xmm0 movq 0x58(%rsp), %rdi movq %rbp, %rsi callq 0x365d4 movl 0xc0(%r12), %eax imull %ebx, %eax notl %eax movslq %eax, %rdx movq %r14, %rdi movq %rbp, %rsi callq 0x2dccbc movq %r14, %rdi callq 0x2dcd42 vmovss 0xc(%rsp), %xmm2 vmovss 0x10(%rsp), %xmm0 testb %al, %al jne 0x2dc254 movl %ebx, 0x78(%rsp) pushq $0x1 popq %rax movl %eax, %ebx leal -0x1(%rbx), %r13d cmpl 0xc0(%r12), %r13d jge 0x2dc4cc movq 0x18(%r14), %rcx vmovss 0x1c(%rsp), %xmm1 leaq 0x98(%rsp), %rdi movq %r15, %rsi movq %rbp, %rdx vmovss %xmm0, 0x10(%rsp) callq 0x2d6b3a testl %eax, %eax je 0x2dc482 vcvtsi2sd %r13d, %xmm4, %xmm0 vucomisd 0xa0(%r12), %xmm0 jbe 0x2dc449 vcvtsi2ss %eax, %xmm4, %xmm0 vcvtsi2ssl 0x158(%rsp), %xmm4, %xmm1 vdivss %xmm1, %xmm0, %xmm0 vmovss 0x48(%rsp), %xmm1 callq 0xb530 vmovss 0x10(%rsp), %xmm1 vmovss 0xc(%rsp), %xmm2 vfmadd231ss 0x2c(%rsp), %xmm0, %xmm1 # xmm1 = (xmm0 * mem) + xmm1 vmovaps %xmm1, %xmm0 jmp 0x2dc455 vmovss 0xc(%rsp), %xmm2 vmovss 0x10(%rsp), %xmm0 vucomiss 0x50(%rsp), %xmm0 ja 0x2dc4cc movq %r14, %rdi vmovss %xmm0, 0x10(%rsp) callq 0x2dcd42 testb %al, %al jne 0x2dc4c0 vmovss 0xc(%rsp), %xmm2 vmovss 0x10(%rsp), %xmm0 incl %ebx jmp 0x2dc3ca vmovsd 0x60(%rsp), %xmm0 movq 0x58(%rsp), %rdi movq %rbp, %rsi callq 0x365d4 movl 0xc0(%r12), %eax imull 0x40(%rsp), %eax subl %ebx, %eax movslq %eax, %rdx movq %r14, %rdi movq %rbp, %rsi callq 0x2dccbc vmovss 0xc(%rsp), %xmm2 vmovss 0x10(%rsp), %xmm0 jmp 0x2dc4cc vmovss 0xc(%rsp), %xmm2 vmovss 0x10(%rsp), %xmm0 movl 0x78(%rsp), %ebx decl 0x40(%rsp) incl %ebx jmp 0x2dc336 jmp 0x2dc512 jmp 0x2dc4eb jmp 0x2dc512 movq %rax, %r14 jmp 0x2dc522 movq %rax, %r14 jmp 0x2dc52f movq %rax, %r14 jmp 0x2dc53c movq %rax, %r14 jmp 0x2dc549 movq %rax, %r14 jmp 0x2dc553 jmp 0x2dc512 jmp 0x2dc512 jmp 0x2dc512 leaq 0x168(%rsp), %rdi movq %rax, %r14 callq 0x357fe jmp 0x2dc522 leaq 0x160(%rsp), %rdi movq %rax, %r14 callq 0x435d0 leaq 0x98(%rsp), %rdi callq 0x393ac leaq 0xd0(%rsp), %rdi callq 0x2d6f9a leaq 0x90(%rsp), %rdi callq 0x35c1a leaq 0x38(%rsp), %rdi callq 0x357fe movq 0x20(%rsp), %rdi callq 0xea60 movq %r14, %rdi callq 0xb3d0 nop
/quesnel[P]baryonyx/lib/src/itm-solver-common.hpp
baryonyx::itm::solver_functor<baryonyx::itm::solver_inequalities_101coeff<float, baryonyx::itm::minimize_tag, baryonyx::itm::default_cost_type<float>, true>, float, baryonyx::itm::minimize_tag, baryonyx::itm::default_cost_type<float>, baryonyx::none_observer>::operator()(std::vector<baryonyx::itm::merged_constraint, std::allocator<baryonyx::itm::merged_constraint>> const&, int, baryonyx::itm::default_cost_type<float> const&, double)
result operator()(const std::vector<merged_constraint>& constraints, int variables, const Cost& original_costs, double cost_constant) { result r; bit_array x(variables); int best_remaining = INT_MAX; auto& p = m_ctx.parameters; auto norm_costs = normalize_costs<Float, Cost>( m_ctx, original_costs, m_rng, variables); const auto kappa_step = static_cast<Float>(p.kappa_step); const auto kappa_max = static_cast<Float>(p.kappa_max); const auto alpha = static_cast<Float>(p.alpha); const auto theta = static_cast<Float>(p.theta); const auto delta = p.delta < 0 ? compute_delta<Float, Cost>(m_ctx, norm_costs, theta, variables) : static_cast<Float>(p.delta); const auto pushing_k_factor = static_cast<Float>(p.pushing_k_factor); const auto pushing_objective_amplifier = static_cast<Float>(p.pushing_objective_amplifier); const long int w_limit = static_cast<long int>(p.w); Solver slv( m_rng, length(constraints), variables, norm_costs, constraints); compute_order compute(p.order, variables); { std::bernoulli_distribution choose_mutation( m_ctx.parameters.init_policy_random); bit_array empty_x; switch (p.init_policy) { case solver_parameters::init_policy_type::pessimistic_solve: init_with_pre_solve<Cost, Mode>( x, empty_x, m_rng, original_costs, constraints, 1.0); break; case solver_parameters::init_policy_type::optimistic_solve: init_with_pre_solve<Cost, Mode>( empty_x, x, m_rng, original_costs, constraints, 1.0); break; case solver_parameters::init_policy_type::bastert: init_with_bastert<Cost, Mode>(x, original_costs, variables, 0); break; } for (int i = 0, e = x.size(); i != e; ++i) if (choose_mutation(m_rng)) x.invert(i); } bool start_push = false; auto kappa = static_cast<Float>(p.kappa_min); Observer obs("img", slv.m, slv.n, p.limit); m_begin = std::chrono::steady_clock::now(); m_end = std::chrono::steady_clock::now(); compute.init(slv, x); for (long int i = 0; i != p.limit; ++i) { auto remaining = compute.run(slv, x, m_rng, kappa, delta, theta); obs.make_observation(slv.ap, slv.P.get(), slv.pi.get()); if (remaining == 0) { store_if_better( x, original_costs.results(x, cost_constant), i); best_remaining = remaining; start_push = true; break; } if (remaining < best_remaining) { store_if_better(x, remaining, i); best_remaining = remaining; } if (i > w_limit) kappa += kappa_step * std::pow(static_cast<Float>(remaining) / static_cast<Float>(slv.m), alpha); if (kappa > kappa_max) { r.status = result_status::kappa_max_reached; break; } if (is_timelimit_reached()) { r.status = result_status::time_limit_reached; break; } } if (!start_push) { r.status = result_status::limit_reached; } else { for (int push = 0; push < p.pushes_limit; ++push) { auto remaining = compute.push_and_run(slv, x, m_rng, pushing_k_factor * kappa, delta, theta, pushing_objective_amplifier); if (remaining == 0) store_if_better(x, original_costs.results(x, cost_constant), -push * p.pushing_iteration_limit - 1); if (is_timelimit_reached()) break; for (int iter = 0; iter < p.pushing_iteration_limit; ++iter) { remaining = compute.run(slv, x, m_rng, kappa, delta, theta); if (remaining == 0) { store_if_better( x, original_costs.results(x, cost_constant), -push * p.pushing_iteration_limit - iter - 1); break; } if (iter > p.w) kappa += kappa_step * std::pow(static_cast<Float>(remaining) / static_cast<Float>(slv.m), alpha); if (kappa > kappa_max) break; if (is_timelimit_reached()) break; } } } if (m_best.remaining_constraints == 0) r.status = result_status::success; if (!m_best.x.empty()) { r.solutions.resize(1); convert(m_best, r.solutions[0], variables); } return r; }
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x168, %rsp # imm = 0x168 vmovsd %xmm0, 0x58(%rsp) vxorps %xmm0, %xmm0, %xmm0 movq %r8, %r13 movl %ecx, %ebx movq %rdx, %r15 movq %rsi, %r14 movq %rdi, 0x18(%rsp) vmovups %zmm0, 0x58(%rdi) vmovups %zmm0, 0x40(%rdi) vmovups %zmm0, (%rdi) movq $0x7fffffff, 0x98(%rdi) # imm = 0x7FFFFFFF movl $0x2, 0xa0(%rdi) leaq 0x28(%rsp), %rdi movl %ecx, %esi vzeroupper callq 0x368ae movq 0x10(%r14), %r12 movq 0x18(%r14), %rcx leaq 0x90(%rsp), %rdi movq %r12, %rsi movq %r13, %rdx movl %ebx, %r8d callq 0x38959 vmovsd 0x20(%r12), %xmm0 vmovsd 0x10(%r12), %xmm2 vxorpd %xmm1, %xmm1, %xmm1 vmovsd %xmm0, 0x48(%rsp) vmovsd 0x28(%r12), %xmm0 vucomisd %xmm2, %xmm1 vmovsd %xmm0, 0x40(%rsp) vmovsd 0x30(%r12), %xmm0 vmovsd %xmm0, 0x38(%rsp) vcvtsd2ss 0x8(%r12), %xmm3, %xmm0 vmovss %xmm0, 0x4(%rsp) jbe 0x2dc64a movq 0x10(%r14), %rdi leaq 0x90(%rsp), %rsi movl %ebx, %edx callq 0x38a37 vmovss %xmm0, 0x14(%rsp) jmp 0x2dc654 vcvtsd2ss %xmm2, %xmm2, %xmm0 vmovss %xmm0, 0x14(%rsp) vcvttsd2si 0xa0(%r12), %rax vmovsd 0x38(%r12), %xmm0 vmovsd 0x40(%r12), %xmm1 movq 0x18(%r14), %rsi movq %rax, 0x70(%rsp) movq 0x8(%r15), %rax vmovsd %xmm0, 0x68(%rsp) vmovsd %xmm1, 0x60(%rsp) subq (%r15), %rax pushq $0x28 popq %rcx cqto idivq %rcx leaq 0xd8(%rsp), %rdi leaq 0x90(%rsp), %r8 movl %eax, %edx movl %ebx, %ecx movq %r15, %r9 callq 0x2d6834 movl 0xcc(%r12), %esi leaq 0xa0(%rsp), %rdi movl %ebx, %edx callq 0x38cd8 movq 0x10(%r14), %rax movl %ebx, 0x7c(%rsp) vmovsd 0x48(%rax), %xmm0 vmovsd %xmm0, 0x8(%rsp) vxorps %xmm0, %xmm0, %xmm0 vmovaps %xmm0, 0x80(%rsp) movl 0xd4(%r12), %eax testl %eax, %eax je 0x2dc732 cmpl $0x1, %eax je 0x2dc70c cmpl $0x2, %eax jne 0x2dc743 movq 0x18(%r14), %rdx leaq 0x80(%rsp), %rdi leaq 0x28(%rsp), %rsi jmp 0x2dc71d movq 0x18(%r14), %rdx leaq 0x28(%rsp), %rdi leaq 0x80(%rsp), %rsi vmovsd 0x177913(%rip), %xmm0 # 0x454038 movq %r13, %rcx movq %r15, %r8 callq 0x4f9c9 jmp 0x2dc743 leaq 0x28(%rsp), %rdi movq %r13, %rsi movl %ebx, %edx xorl %ecx, %ecx callq 0x4f95d vcvtsd2ss 0x48(%rsp), %xmm3, %xmm0 leaq 0x28(%rsp), %r15 movq %r13, 0x50(%rsp) xorl %ebx, %ebx movl (%r15), %ebp vmovss %xmm0, 0x24(%rsp) vcvtsd2ss 0x40(%rsp), %xmm3, %xmm0 vmovss %xmm0, 0x48(%rsp) vcvtsd2ss 0x38(%rsp), %xmm3, %xmm0 vmovss %xmm0, 0x40(%rsp) vcvtsd2ss 0x68(%rsp), %xmm3, %xmm0 vmovss %xmm0, 0x68(%rsp) vcvtsd2ss 0x60(%rsp), %xmm3, %xmm0 vmovss %xmm0, 0x60(%rsp) cmpl %ebx, %ebp je 0x2dc7b5 movq 0x18(%r14), %rdi callq 0x36ac3 vmovsd 0x8(%rsp), %xmm1 vucomisd %xmm0, %xmm1 jbe 0x2dc7b1 movq %r15, %rdi movl %ebx, %esi callq 0x3626e incl %ebx jmp 0x2dc78e leaq 0x88(%rsp), %rdi callq 0x357fe vmovsd 0x18(%r12), %xmm0 vmovsd %xmm0, 0x8(%rsp) callq 0xb5a0 movq %rax, (%r14) callq 0xb5a0 movq %rax, 0x8(%r14) leaq 0xa0(%rsp), %rdi leaq 0xd8(%rsp), %rsi leaq 0x28(%rsp), %rdx callq 0x2d6af8 vcvtsd2ss 0x8(%rsp), %xmm3, %xmm0 vmovss 0x4(%rsp), %xmm2 movl $0x7fffffff, %ebp # imm = 0x7FFFFFFF leaq 0x28(%rsp), %rbx xorl %r15d, %r15d cmpq 0xb0(%r12), %r15 je 0x2dc8d8 movq 0x18(%r14), %rcx vmovss 0x14(%rsp), %xmm1 leaq 0xa0(%rsp), %rdi leaq 0xd8(%rsp), %rsi movq %rbx, %rdx vmovss %xmm0, 0x8(%rsp) callq 0x2d6b3a movl %eax, %r13d testl %eax, %eax je 0x2dc970 cmpl %ebp, %r13d jge 0x2dc86d movq %r14, %rdi movq %rbx, %rsi movl %r13d, %edx movq %r15, %rcx callq 0x2dcdf0 movl %r13d, %ebp cmpq 0x70(%rsp), %r15 jle 0x2dc8a4 vcvtsi2ss %r13d, %xmm3, %xmm0 vcvtsi2ssl 0x160(%rsp), %xmm3, %xmm1 vdivss %xmm1, %xmm0, %xmm0 vmovss 0x40(%rsp), %xmm1 callq 0xb530 vmovss 0x8(%rsp), %xmm1 vfmadd231ss 0x24(%rsp), %xmm0, %xmm1 # xmm1 = (xmm0 * mem) + xmm1 vmovaps %xmm1, %xmm0 jmp 0x2dc8aa vmovss 0x8(%rsp), %xmm0 vucomiss 0x48(%rsp), %xmm0 ja 0x2dc8d8 movq %r14, %rdi vmovss %xmm0, 0x8(%rsp) callq 0x2dce28 vmovss 0x8(%rsp), %xmm0 vmovss 0x4(%rsp), %xmm2 testb %al, %al jne 0x2dc8d8 incq %r15 jmp 0x2dc813 movq 0x18(%rsp), %rax movl $0x5, 0xa0(%rax) cmpl $0x0, 0x50(%r14) jne 0x2dc8fa movq 0x18(%rsp), %rax andl $0x0, 0xa0(%rax) cmpq $0x0, 0x28(%r14) je 0x2dc928 movq 0x18(%rsp), %rax leaq 0x68(%rax), %rbx pushq $0x1 popq %rsi movq %rbx, %rdi callq 0x35292 movq (%rbx), %rsi addq $0x20, %r14 movl 0x7c(%rsp), %edx movq %r14, %rdi callq 0x4f485 leaq 0xa0(%rsp), %rdi callq 0x393ac leaq 0xd8(%rsp), %rdi callq 0x2d6f9a leaq 0x98(%rsp), %rdi callq 0x35c1a leaq 0x30(%rsp), %rdi callq 0x357fe movq 0x18(%rsp), %rax addq $0x168, %rsp # imm = 0x168 popq %rbx popq %r12 popq %r13 popq %r14 popq %r15 popq %rbp retq vmovsd 0x58(%rsp), %xmm0 movq 0x50(%rsp), %rdi leaq 0x28(%rsp), %rbx movq %rbx, %rsi callq 0x365d4 movq %r14, %rdi movq %rbx, %rsi movq %r15, %rdx callq 0x2dcda2 vmovss 0x4(%rsp), %xmm2 vmovss 0x8(%rsp), %xmm0 leaq 0xd8(%rsp), %r15 leaq 0x28(%rsp), %rbp xorl %ebx, %ebx movl $0x0, 0x38(%rsp) cmpl 0xbc(%r12), %ebx jge 0x2dc8e7 vmovss %xmm0, 0x8(%rsp) vmulss 0x68(%rsp), %xmm0, %xmm0 movq 0x18(%r14), %rcx vmovss 0x14(%rsp), %xmm1 vmovss 0x60(%rsp), %xmm3 leaq 0xa0(%rsp), %rdi movq %r15, %rsi movq %rbp, %rdx callq 0x2d6d46 testl %eax, %eax jne 0x2dca28 vmovsd 0x58(%rsp), %xmm0 movq 0x50(%rsp), %rdi movq %rbp, %rsi callq 0x365d4 movl 0xc0(%r12), %eax imull %ebx, %eax notl %eax movslq %eax, %rdx movq %r14, %rdi movq %rbp, %rsi callq 0x2dcda2 movq %r14, %rdi callq 0x2dce28 vmovss 0x4(%rsp), %xmm2 vmovss 0x8(%rsp), %xmm0 testb %al, %al jne 0x2dc8e7 movl %ebx, 0x70(%rsp) pushq $0x1 popq %rax movl %eax, %ebx leal -0x1(%rbx), %r13d cmpl 0xc0(%r12), %r13d jge 0x2dcb4f movq 0x18(%r14), %rcx vmovss 0x14(%rsp), %xmm1 leaq 0xa0(%rsp), %rdi movq %r15, %rsi movq %rbp, %rdx vmovss %xmm0, 0x8(%rsp) callq 0x2d6b3a testl %eax, %eax je 0x2dcb05 vcvtsi2sd %r13d, %xmm4, %xmm0 vucomisd 0xa0(%r12), %xmm0 jbe 0x2dcacc vcvtsi2ss %eax, %xmm4, %xmm0 vcvtsi2ssl 0x160(%rsp), %xmm4, %xmm1 vdivss %xmm1, %xmm0, %xmm0 vmovss 0x40(%rsp), %xmm1 callq 0xb530 vmovss 0x8(%rsp), %xmm1 vmovss 0x4(%rsp), %xmm2 vfmadd231ss 0x24(%rsp), %xmm0, %xmm1 # xmm1 = (xmm0 * mem) + xmm1 vmovaps %xmm1, %xmm0 jmp 0x2dcad8 vmovss 0x4(%rsp), %xmm2 vmovss 0x8(%rsp), %xmm0 vucomiss 0x48(%rsp), %xmm0 ja 0x2dcb4f movq %r14, %rdi vmovss %xmm0, 0x8(%rsp) callq 0x2dce28 testb %al, %al jne 0x2dcb43 vmovss 0x4(%rsp), %xmm2 vmovss 0x8(%rsp), %xmm0 incl %ebx jmp 0x2dca4d vmovsd 0x58(%rsp), %xmm0 movq 0x50(%rsp), %rdi movq %rbp, %rsi callq 0x365d4 movl 0xc0(%r12), %eax imull 0x38(%rsp), %eax subl %ebx, %eax movslq %eax, %rdx movq %r14, %rdi movq %rbp, %rsi callq 0x2dcda2 vmovss 0x4(%rsp), %xmm2 vmovss 0x8(%rsp), %xmm0 jmp 0x2dcb4f vmovss 0x4(%rsp), %xmm2 vmovss 0x8(%rsp), %xmm0 movl 0x70(%rsp), %ebx decl 0x38(%rsp) incl %ebx jmp 0x2dc9b9 jmp 0x2dcb90 jmp 0x2dcb69 jmp 0x2dcb90 movq %rax, %r14 jmp 0x2dcba0 movq %rax, %r14 jmp 0x2dcbad movq %rax, %r14 jmp 0x2dcbba movq %rax, %r14 jmp 0x2dcbc4 jmp 0x2dcb90 jmp 0x2dcb90 jmp 0x2dcb90 leaq 0x88(%rsp), %rdi movq %rax, %r14 callq 0x357fe jmp 0x2dcb93 movq %rax, %r14 leaq 0xa0(%rsp), %rdi callq 0x393ac leaq 0xd8(%rsp), %rdi callq 0x2d6f9a leaq 0x98(%rsp), %rdi callq 0x35c1a leaq 0x30(%rsp), %rdi callq 0x357fe movq 0x18(%rsp), %rdi callq 0xea60 movq %r14, %rdi callq 0xb3d0
/quesnel[P]baryonyx/lib/src/itm-solver-common.hpp
baryonyx::itm::solver_functor<baryonyx::itm::solver_inequalities_101coeff<float, baryonyx::itm::minimize_tag, baryonyx::itm::default_cost_type<float>, true>, float, baryonyx::itm::minimize_tag, baryonyx::itm::default_cost_type<float>, baryonyx::file_observer>::store_if_better(baryonyx::bit_array const&, double, long)
void store_if_better(const bit_array& x, double current, long int i) { if (is_better_solution<Mode>(current, m_best.value)) { m_best.x = x; m_best.duration = duration(); m_best.loop = i; m_best.remaining_constraints = 0; m_best.value = current; } }
vmovsd 0x30(%rdi), %xmm1 vucomisd %xmm0, %xmm1 jbe 0x2dcd08 pushq %r14 pushq %rbx pushq %rax movq %rdi, %r14 addq $0x20, %rdi movq %rdx, %rbx vmovsd %xmm0, (%rsp) callq 0x36a44 movq %r14, %rdi callq 0x2dcd82 vmovsd (%rsp), %xmm1 vmovsd %xmm0, 0x38(%r14) movq %rbx, 0x48(%r14) andl $0x0, 0x50(%r14) vmovsd %xmm1, 0x30(%r14) addq $0x8, %rsp popq %rbx popq %r14 retq nop
/quesnel[P]baryonyx/lib/src/itm-solver-common.hpp
baryonyx::itm::solver_functor<baryonyx::itm::solver_inequalities_101coeff<float, baryonyx::itm::minimize_tag, baryonyx::itm::default_cost_type<float>, true>, float, baryonyx::itm::minimize_tag, baryonyx::itm::default_cost_type<float>, baryonyx::file_observer>::is_timelimit_reached()
bool is_timelimit_reached() { m_end = std::chrono::steady_clock::now(); return is_time_limit(m_ctx.parameters.time_limit, m_begin, m_end); }
pushq %rbx movq %rdi, %rbx callq 0xb5a0 movq %rax, 0x8(%rbx) vxorpd %xmm2, %xmm2, %xmm2 movq 0x10(%rbx), %rcx subq (%rbx), %rax vmovsd (%rcx), %xmm0 vcvtsi2sd %rax, %xmm1, %xmm1 vdivsd 0x17cd15(%rip), %xmm1, %xmm1 # 0x459a80 vcmpnlepd %xmm2, %xmm0, %k1 vcmpltpd %xmm1, %xmm0, %k0 {%k1} kmovd %k0, %eax andb $0x1, %al popq %rbx retq nop
/quesnel[P]baryonyx/lib/src/itm-solver-common.hpp
baryonyx::itm::solver_functor<baryonyx::itm::solver_inequalities_101coeff<float, baryonyx::itm::minimize_tag, baryonyx::itm::default_cost_type<float>, true>, float, baryonyx::itm::minimize_tag, baryonyx::itm::default_cost_type<float>, baryonyx::none_observer>::store_if_better(baryonyx::bit_array const&, double, long)
void store_if_better(const bit_array& x, double current, long int i) { if (is_better_solution<Mode>(current, m_best.value)) { m_best.x = x; m_best.duration = duration(); m_best.loop = i; m_best.remaining_constraints = 0; m_best.value = current; } }
vmovsd 0x30(%rdi), %xmm1 vucomisd %xmm0, %xmm1 jbe 0x2dcdee pushq %r14 pushq %rbx pushq %rax movq %rdi, %r14 addq $0x20, %rdi movq %rdx, %rbx vmovsd %xmm0, (%rsp) callq 0x36a44 movq %r14, %rdi callq 0x2dce68 vmovsd (%rsp), %xmm1 vmovsd %xmm0, 0x38(%r14) movq %rbx, 0x48(%r14) andl $0x0, 0x50(%r14) vmovsd %xmm1, 0x30(%r14) addq $0x8, %rsp popq %rbx popq %r14 retq nop
/quesnel[P]baryonyx/lib/src/itm-solver-common.hpp
baryonyx::itm::solver_functor<baryonyx::itm::solver_inequalities_101coeff<float, baryonyx::itm::minimize_tag, baryonyx::itm::default_cost_type<float>, true>, float, baryonyx::itm::minimize_tag, baryonyx::itm::default_cost_type<float>, baryonyx::none_observer>::is_timelimit_reached()
bool is_timelimit_reached() { m_end = std::chrono::steady_clock::now(); return is_time_limit(m_ctx.parameters.time_limit, m_begin, m_end); }
pushq %rbx movq %rdi, %rbx callq 0xb5a0 movq %rax, 0x8(%rbx) vxorpd %xmm2, %xmm2, %xmm2 movq 0x10(%rbx), %rcx subq (%rbx), %rax vmovsd (%rcx), %xmm0 vcvtsi2sd %rax, %xmm1, %xmm1 vdivsd 0x17cc2f(%rip), %xmm1, %xmm1 # 0x459a80 vcmpnlepd %xmm2, %xmm0, %k1 vcmpltpd %xmm1, %xmm0, %k0 {%k1} kmovd %k0, %eax andb $0x1, %al popq %rbx retq nop
/quesnel[P]baryonyx/lib/src/itm-solver-common.hpp
void baryonyx::itm::compute_order::init<baryonyx::itm::solver_inequalities_101coeff<float, baryonyx::itm::minimize_tag, baryonyx::itm::default_cost_type<float>, false>, baryonyx::bit_array>(baryonyx::itm::solver_inequalities_101coeff<float, baryonyx::itm::minimize_tag, baryonyx::itm::default_cost_type<float>, false> const&, baryonyx::bit_array const&)
void init(const Solver& s, const Xtype& x) { switch (order) { case solver_parameters::constraint_order::infeasibility_decr: case solver_parameters::constraint_order::infeasibility_incr: infeasibility_local_compute_violated_constraints(s, x); break; case solver_parameters::constraint_order::pi_sign_change: std::iota(R.begin(), R.end(), 0); break; case solver_parameters::constraint_order::none: case solver_parameters::constraint_order::reversing: case solver_parameters::constraint_order::random_sorting: case solver_parameters::constraint_order::lagrangian_decr: case solver_parameters::constraint_order::lagrangian_incr: default: compute_violated_constraints(s, x, R); break; } }
movl 0x30(%rdi), %ecx movq %rdi, %rax leal -0x3(%rcx), %edi cmpl $0x2, %edi jae 0x2dda5e movq %rax, %rdi jmp 0x2ddff2 cmpl $0x7, %ecx jne 0x2dda7b movq (%rax), %rcx movq 0x8(%rax), %rax xorl %edx, %edx cmpq %rax, %rcx je 0x2dda89 movl %edx, (%rcx) incl %edx addq $0x4, %rcx jmp 0x2dda6c movq %rsi, %rdi movq %rdx, %rsi movq %rax, %rdx jmp 0x2de10d retq
/quesnel[P]baryonyx/lib/src/itm-common.hpp
int baryonyx::itm::compute_order::run<baryonyx::itm::solver_inequalities_101coeff<float, baryonyx::itm::minimize_tag, baryonyx::itm::default_cost_type<float>, false>, baryonyx::bit_array, float>(baryonyx::itm::solver_inequalities_101coeff<float, baryonyx::itm::minimize_tag, baryonyx::itm::default_cost_type<float>, false>&, baryonyx::bit_array&, std::linear_congruential_engine<unsigned long, 16807ul, 0ul, 2147483647ul>&, float, float, float)
int run(Solver& solver, Xtype& x, random_engine& rng, Float kappa, Float delta, Float theta) { bool pi_changed = false; int remaining = 0; switch (order) { case solver_parameters::constraint_order::reversing: solver.compute_update_row( x, R.crbegin(), R.crend(), kappa, delta, theta); return compute_violated_constraints(solver, x, R); case solver_parameters::constraint_order::random_sorting: std::shuffle(R.begin(), R.end(), rng); solver.compute_update_row( x, R.begin(), R.end(), kappa, delta, theta); return compute_violated_constraints(solver, x, R); case solver_parameters::constraint_order::infeasibility_decr: std::sort(m_order.begin(), m_order.end(), [](const auto& lhs, const auto& rhs) { return rhs.second < lhs.second; }); solver.compute_update_row( x, m_order.cbegin(), m_order.cend(), kappa, delta, theta); return infeasibility_local_compute_violated_constraints(solver, x); case solver_parameters::constraint_order::infeasibility_incr: std::sort(m_order.begin(), m_order.end(), [](const auto& lhs, const auto& rhs) { return lhs.second < rhs.second; }); solver.compute_update_row( x, m_order.cbegin(), m_order.cend(), kappa, delta, theta); return infeasibility_local_compute_violated_constraints(solver, x); case solver_parameters::constraint_order::lagrangian_decr: std::sort( R.begin(), R.end(), [&solver](const auto lhs, const auto rhs) { return solver.pi[rhs] < solver.pi[lhs]; }); solver.compute_update_row( x, R.cbegin(), R.cend(), kappa, delta, theta); return compute_violated_constraints(solver, x, R); case solver_parameters::constraint_order::lagrangian_incr: std::sort( R.begin(), R.end(), [&solver](const auto lhs, const auto rhs) { return solver.pi[lhs] < solver.pi[rhs]; }); solver.compute_update_row( x, R.cbegin(), R.cend(), kappa, delta, theta); return compute_violated_constraints(solver, x, R); case solver_parameters::constraint_order::pi_sign_change: std::shuffle(R.begin(), R.end(), rng); pi_changed = solver.compute_update_row( x, R.begin(), R.end(), kappa, delta, theta); remaining = local_compute_violated_constraints(solver, x); if (!pi_changed && remaining == 0) return 0; return remaining; case solver_parameters::constraint_order::none: default: solver.compute_update_row( x, R.begin(), R.end(), kappa, delta, theta); return compute_violated_constraints(solver, x, R); } }
pushq %r15 pushq %r14 pushq %rbx subq $0x20, %rsp movl 0x30(%rdi), %eax movq %rdx, %rbx movq %rsi, %r15 movq %rdi, %r14 decl %eax cmpl $0x6, %eax ja 0x2ddbc9 leaq 0x185767(%rip), %rdx # 0x463218 movslq (%rdx,%rax,4), %rax addq %rdx, %rax jmpq *%rax movq 0x8(%r14), %rax leaq 0x18(%rsp), %rdx leaq 0x10(%rsp), %rcx movq %r15, %rdi movq %rbx, %rsi movq %rax, (%rdx) movq (%r14), %rax movq %rax, (%rcx) callq 0x2de214 movq %r15, %rdi movq %rbx, %rsi movq %r14, %rdx callq 0x2de10d addq $0x20, %rsp popq %rbx popq %r14 popq %r15 retq movq (%r14), %rdi movq 0x8(%r14), %rsi movq %r15, %rdx vmovss %xmm2, 0xc(%rsp) vmovss %xmm1, 0x8(%rsp) vmovss %xmm0, 0x4(%rsp) callq 0x2dfcfc jmp 0x2ddbfe movq 0x18(%r14), %rdi movq 0x20(%r14), %rsi vmovss %xmm2, 0xc(%rsp) vmovss %xmm1, 0x8(%rsp) vmovss %xmm0, 0x4(%rsp) callq 0x2df426 jmp 0x2ddb5a movq 0x18(%r14), %rdi movq 0x20(%r14), %rsi vmovss %xmm2, 0xc(%rsp) vmovss %xmm1, 0x8(%rsp) vmovss %xmm0, 0x4(%rsp) callq 0x2df891 vmovss 0x4(%rsp), %xmm0 vmovss 0x8(%rsp), %xmm1 vmovss 0xc(%rsp), %xmm2 movq 0x18(%r14), %rdx movq 0x20(%r14), %rcx movq %r15, %rdi movq %rbx, %rsi callq 0x2de420 movq %r14, %rdi movq %r15, %rsi movq %rbx, %rdx addq $0x20, %rsp popq %rbx popq %r14 popq %r15 jmp 0x2de520 movq (%r14), %rdi movq 0x8(%r14), %rsi movq %rcx, %rdx vmovss %xmm2, 0xc(%rsp) vmovss %xmm1, 0x8(%rsp) vmovss %xmm0, 0x4(%rsp) callq 0x3ceed vmovss 0x4(%rsp), %xmm0 vmovss 0x8(%rsp), %xmm1 vmovss 0xc(%rsp), %xmm2 movq (%r14), %rdx movq 0x8(%r14), %rcx movq %r15, %rdi movq %rbx, %rsi callq 0x2de320 jmp 0x2ddc22 movq (%r14), %rdi movq 0x8(%r14), %rsi movq %r15, %rdx vmovss %xmm2, 0xc(%rsp) vmovss %xmm1, 0x8(%rsp) vmovss %xmm0, 0x4(%rsp) callq 0x2e01ee vmovss 0x4(%rsp), %xmm0 vmovss 0x8(%rsp), %xmm1 vmovss 0xc(%rsp), %xmm2 movq (%r14), %rdx movq 0x8(%r14), %rcx movq %r15, %rdi movq %rbx, %rsi callq 0x2de63c movq %r15, %rdi movq %rbx, %rsi movq %r14, %rdx addq $0x20, %rsp popq %rbx popq %r14 popq %r15 jmp 0x2de10d movq (%r14), %rdi movq 0x8(%r14), %rsi movq %rcx, %rdx vmovss %xmm2, 0xc(%rsp) vmovss %xmm1, 0x8(%rsp) vmovss %xmm0, 0x4(%rsp) callq 0x3ceed vmovss 0x4(%rsp), %xmm0 vmovss 0x8(%rsp), %xmm1 vmovss 0xc(%rsp), %xmm2 movq (%r14), %rdx movq 0x8(%r14), %rcx movq %r15, %rdi movq %rbx, %rsi callq 0x2de320 movq %r14, %rdi movq %r15, %rsi movq %rbx, %rdx addq $0x20, %rsp popq %rbx popq %r14 popq %r15 jmp 0x2de73c nop
/quesnel[P]baryonyx/lib/src/itm-common.hpp
int baryonyx::itm::compute_order::push_and_run<baryonyx::itm::solver_inequalities_101coeff<float, baryonyx::itm::minimize_tag, baryonyx::itm::default_cost_type<float>, false>, baryonyx::bit_array, float>(baryonyx::itm::solver_inequalities_101coeff<float, baryonyx::itm::minimize_tag, baryonyx::itm::default_cost_type<float>, false>&, baryonyx::bit_array&, std::linear_congruential_engine<unsigned long, 16807ul, 0ul, 2147483647ul>&, float, float, float, float)
int push_and_run(Solver& solver, Xtype& x, random_engine& rng, Float kappa, Float delta, Float theta, Float objective_amplifier) { bool pi_changed = 0; int remaining = 0; if (use_cycle) order = next_state(order); switch (order) { case solver_parameters::constraint_order::reversing: solver.push_and_compute_update_row(x, R.crbegin(), R.crend(), kappa, delta, theta, objective_amplifier); return compute_violated_constraints(solver, x, R); case solver_parameters::constraint_order::random_sorting: std::shuffle(R.begin(), R.end(), rng); solver.push_and_compute_update_row( x, R.begin(), R.end(), kappa, delta, theta, objective_amplifier); return compute_violated_constraints(solver, x, R); case solver_parameters::constraint_order::infeasibility_decr: std::sort(m_order.begin(), m_order.end(), [](const auto& lhs, const auto& rhs) { return rhs.second < lhs.second; }); solver.push_and_compute_update_row(x, m_order.cbegin(), m_order.cend(), kappa, delta, theta, objective_amplifier); return infeasibility_local_compute_violated_constraints(solver, x); case solver_parameters::constraint_order::infeasibility_incr: std::sort(m_order.begin(), m_order.end(), [](const auto& lhs, const auto& rhs) { return lhs.second < rhs.second; }); solver.push_and_compute_update_row(x, m_order.cbegin(), m_order.cend(), kappa, delta, theta, objective_amplifier); return infeasibility_local_compute_violated_constraints(solver, x); case solver_parameters::constraint_order::lagrangian_decr: std::sort( R.begin(), R.end(), [&solver](const auto lhs, const auto rhs) { return solver.pi[rhs] < solver.pi[lhs]; }); solver.push_and_compute_update_row(x, R.cbegin(), R.cend(), kappa, delta, theta, objective_amplifier); return compute_violated_constraints(solver, x, R); case solver_parameters::constraint_order::lagrangian_incr: std::sort( R.begin(), R.end(), [&solver](const auto lhs, const auto rhs) { return solver.pi[lhs] < solver.pi[rhs]; }); solver.push_and_compute_update_row(x, R.cbegin(), R.cend(), kappa, delta, theta, objective_amplifier); return compute_violated_constraints(solver, x, R); case solver_parameters::constraint_order::pi_sign_change: std::shuffle(R.begin(), R.end(), rng); pi_changed = solver.push_and_compute_update_row( x, R.begin(), R.end(), kappa, delta, theta, objective_amplifier); remaining = local_compute_violated_constraints(solver, x); if (!pi_changed && remaining == 0) return 0; return remaining; case solver_parameters::constraint_order::none: default: solver.push_and_compute_update_row(x, R.cbegin(), R.cend(), kappa, delta, theta, objective_amplifier); return compute_violated_constraints(solver, x, R); } }
pushq %r15 pushq %r14 pushq %rbx subq $0x20, %rsp movl 0x30(%rdi), %eax cmpb $0x1, 0x34(%rdi) movq %rdx, %rbx movq %rsi, %r15 movq %rdi, %r14 jne 0x2ddcc2 xorl %esi, %esi leal 0x1(%rax), %edx cmpl $0x7, %eax cmovll %edx, %esi movl %esi, %eax movl %esi, 0x30(%r14) decl %eax cmpl $0x6, %eax ja 0x2dde5b leaq 0x185560(%rip), %rdx # 0x463234 movslq (%rdx,%rax,4), %rax addq %rdx, %rax jmpq *%rax movq 0x8(%r14), %rax leaq 0x18(%rsp), %rdx leaq 0x10(%rsp), %rcx movq %r15, %rdi movq %rbx, %rsi movq %rax, (%rdx) movq (%r14), %rax movq %rax, (%rcx) callq 0x2e06e4 movq %r15, %rdi movq %rbx, %rsi movq %r14, %rdx callq 0x2de10d addq $0x20, %rsp popq %rbx popq %r14 popq %r15 retq movq (%r14), %rdi movq 0x8(%r14), %rsi movq %r15, %rdx vmovss %xmm3, 0xc(%rsp) vmovss %xmm2, 0x8(%rsp) vmovss %xmm1, 0x4(%rsp) vmovss %xmm0, (%rsp) callq 0x2e15b3 jmp 0x2dde44 movq 0x18(%r14), %rdi movq 0x20(%r14), %rsi vmovss %xmm3, 0xc(%rsp) vmovss %xmm2, 0x8(%rsp) vmovss %xmm1, 0x4(%rsp) vmovss %xmm0, (%rsp) callq 0x2e0cdd jmp 0x2ddd8c movq 0x18(%r14), %rdi movq 0x20(%r14), %rsi vmovss %xmm3, 0xc(%rsp) vmovss %xmm2, 0x8(%rsp) vmovss %xmm1, 0x4(%rsp) vmovss %xmm0, (%rsp) callq 0x2e1148 vmovss (%rsp), %xmm0 vmovss 0x4(%rsp), %xmm1 vmovss 0x8(%rsp), %xmm2 vmovss 0xc(%rsp), %xmm3 movq 0x18(%r14), %rdx movq 0x20(%r14), %rcx movq %r15, %rdi movq %rbx, %rsi callq 0x2e09e6 movq %r14, %rdi movq %r15, %rsi movq %rbx, %rdx addq $0x20, %rsp popq %rbx popq %r14 popq %r15 jmp 0x2de520 movq (%r14), %rdi movq 0x8(%r14), %rsi movq %rcx, %rdx vmovss %xmm3, 0xc(%rsp) vmovss %xmm2, 0x8(%rsp) vmovss %xmm1, 0x4(%rsp) vmovss %xmm0, (%rsp) callq 0x3ceed vmovss (%rsp), %xmm0 vmovss 0x4(%rsp), %xmm1 vmovss 0x8(%rsp), %xmm2 vmovss 0xc(%rsp), %xmm3 movq (%r14), %rdx movq 0x8(%r14), %rcx movq %r15, %rdi movq %rbx, %rsi callq 0x2e086a jmp 0x2dde6d movq (%r14), %rdi movq 0x8(%r14), %rsi movq %r15, %rdx vmovss %xmm3, 0xc(%rsp) vmovss %xmm2, 0x8(%rsp) vmovss %xmm1, 0x4(%rsp) vmovss %xmm0, (%rsp) callq 0x2e1aa5 vmovss (%rsp), %xmm0 vmovss 0x4(%rsp), %xmm1 vmovss 0x8(%rsp), %xmm2 vmovss 0xc(%rsp), %xmm3 movq (%r14), %rdx movq 0x8(%r14), %rcx movq %r15, %rdi movq %rbx, %rsi callq 0x2e0b62 movq %r15, %rdi movq %rbx, %rsi movq %r14, %rdx addq $0x20, %rsp popq %rbx popq %r14 popq %r15 jmp 0x2de10d movq (%r14), %rdi movq 0x8(%r14), %rsi movq %rcx, %rdx vmovss %xmm3, 0xc(%rsp) vmovss %xmm2, 0x8(%rsp) vmovss %xmm1, 0x4(%rsp) vmovss %xmm0, (%rsp) callq 0x3ceed vmovss (%rsp), %xmm0 vmovss 0x4(%rsp), %xmm1 vmovss 0x8(%rsp), %xmm2 vmovss 0xc(%rsp), %xmm3 movq (%r14), %rdx movq 0x8(%r14), %rcx movq %r15, %rdi movq %rbx, %rsi callq 0x2e086a movq %r14, %rdi movq %r15, %rsi movq %rbx, %rdx addq $0x20, %rsp popq %rbx popq %r14 popq %r15 jmp 0x2de73c
/quesnel[P]baryonyx/lib/src/itm-common.hpp
int baryonyx::itm::compute_order::infeasibility_local_compute_violated_constraints<baryonyx::itm::solver_inequalities_101coeff<float, baryonyx::itm::minimize_tag, baryonyx::itm::default_cost_type<float>, false> const, baryonyx::bit_array>(baryonyx::itm::solver_inequalities_101coeff<float, baryonyx::itm::minimize_tag, baryonyx::itm::default_cost_type<float>, false> const&, baryonyx::bit_array const&)
int infeasibility_local_compute_violated_constraints(Solver& solver, const Xtype& x) { m_order.clear(); for (int k = 0, e = solver.m; k != e; ++k) { sparse_matrix<int>::const_row_iterator it, et; std::tie(it, et) = solver.ap.row(k); int v = 0; for (; it != et; ++it) v += solver.factor(it->value) * x[it->column]; if (solver.bound_min(k) > v) m_order.push_back(std::make_pair(k, solver.bound_min(k) - v)); else if (solver.bound_max(k) < v) m_order.push_back(std::make_pair(k, v - solver.bound_max(k))); } return length(m_order); }
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x48, %rsp movq 0x18(%rdi), %rax movq %rdx, %r14 movq %rsi, 0x20(%rsp) cmpq %rax, 0x20(%rdi) je 0x2de016 movq %rax, 0x20(%rdi) movq 0x20(%rsp), %rsi leaq 0x18(%rdi), %rax movq %rdi, 0x28(%rsp) xorl %ebx, %ebx movq %rax, 0x18(%rsp) movl 0x88(%rsi), %ecx leaq 0x10(%rsi), %rdx leaq 0x58(%rsi), %rbp movq %rdx, 0x30(%rsp) movq %rcx, 0x38(%rsp) cmpq 0x38(%rsp), %rbx je 0x2de0ed movq 0x30(%rsp), %rsi leaq 0x8(%rsp), %rdi movl %ebx, %edx movq %rbx, 0x40(%rsp) callq 0x3c868 movq 0x8(%rsp), %r15 movq 0x10(%rsp), %rbx xorl %r13d, %r13d cmpq %r15, %rbx je 0x2de09b movl (%rbx), %esi movq %rbp, %rdi callq 0x2b9c1a movl 0x4(%rbx), %esi movq %r14, %rdi movl %eax, %r12d callq 0x3664a imull %r12d, %eax addq $0x8, %rbx addl %eax, %r13d jmp 0x2de071 movq 0x20(%rsp), %rax movq 0x40(%rsp), %rbx movq 0x70(%rax), %rcx imulq $0xc, %rbx, %rdx movl (%rcx,%rdx), %eax subl %r13d, %eax jle 0x2de0c3 shlq $0x20, %rax orq %rbx, %rax movq %rax, 0x8(%rsp) jmp 0x2de0d6 subl 0x4(%rcx,%rdx), %r13d jle 0x2de0e5 shlq $0x20, %r13 orq %rbx, %r13 movq %r13, 0x8(%rsp) movq 0x18(%rsp), %rdi leaq 0x8(%rsp), %rsi callq 0x3c8c2 incq %rbx jmp 0x2de043 movq 0x28(%rsp), %rcx movq 0x20(%rcx), %rax subq 0x18(%rcx), %rax shrq $0x3, %rax addq $0x48, %rsp popq %rbx popq %r12 popq %r13 popq %r14 popq %r15 popq %rbp retq
/quesnel[P]baryonyx/lib/src/itm-common.hpp
int baryonyx::itm::compute_violated_constraints<baryonyx::itm::solver_inequalities_101coeff<float, baryonyx::itm::minimize_tag, baryonyx::itm::default_cost_type<float>, false>, baryonyx::bit_array>(baryonyx::itm::solver_inequalities_101coeff<float, baryonyx::itm::minimize_tag, baryonyx::itm::default_cost_type<float>, false> const&, baryonyx::bit_array const&, std::vector<int, std::allocator<int>>&)
int compute_violated_constraints(const Solver& slv, const Xtype& x, std::vector<int>& out) { out.clear(); for (int k = 0; k != slv.m; ++k) if (!is_valid_constraint(slv, k, x)) out.emplace_back(k); return length(out); }
pushq %r15 pushq %r14 pushq %r12 pushq %rbx pushq %rax movq (%rdx), %rax movq %rdx, %rbx movq %rsi, %r14 movq %rdi, %r15 cmpq %rax, 0x8(%rdx) je 0x2de12b movq %rax, 0x8(%rbx) leaq 0x4(%rsp), %r12 xorl %esi, %esi movl %esi, 0x4(%rsp) cmpl 0x88(%r15), %esi je 0x2de161 movq %r15, %rdi movq %r14, %rdx callq 0x2de178 testb %al, %al jne 0x2de159 movq %rbx, %rdi movq %r12, %rsi callq 0x31460 movl 0x4(%rsp), %esi incl %esi jmp 0x2de132 movq 0x8(%rbx), %rax subq (%rbx), %rax shrq $0x2, %rax addq $0x8, %rsp popq %rbx popq %r12 popq %r14 popq %r15 retq
/quesnel[P]baryonyx/lib/src/itm-common.hpp
bool baryonyx::itm::is_valid_constraint<baryonyx::itm::solver_inequalities_101coeff<float, baryonyx::itm::minimize_tag, baryonyx::itm::default_cost_type<float>, false>, baryonyx::bit_array>(baryonyx::itm::solver_inequalities_101coeff<float, baryonyx::itm::minimize_tag, baryonyx::itm::default_cost_type<float>, false> const&, int, baryonyx::bit_array const&)
bool is_valid_constraint(const Solver& slv, int k, const Xtype& x) { typename sparse_matrix<int>::const_row_iterator it, et; std::tie(it, et) = slv.ap.row(k); int v = 0; for (; it != et; ++it) v += slv.factor(it->value) * x[it->column]; return slv.bound_min(k) <= v && v <= slv.bound_max(k); }
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x28, %rsp movq %rdx, %r14 movl %esi, %edx leaq 0x10(%rdi), %rsi leaq 0x18(%rsp), %r15 movq %rdi, %r12 movl %edx, 0xc(%rsp) movq %r15, %rdi callq 0x3c868 movq (%r15), %rbp movq 0x8(%r15), %rbx leaq 0x58(%r12), %r15 xorl %r13d, %r13d movq %r12, 0x10(%rsp) cmpq %rbp, %rbx je 0x2de1e1 movl (%rbx), %esi movq %r15, %rdi callq 0x2b9c1a movl 0x4(%rbx), %esi movq %r14, %rdi movl %eax, %r12d callq 0x3664a imull %r12d, %eax addq $0x8, %rbx addl %eax, %r13d jmp 0x2de1b7 movq 0x10(%rsp), %rax movslq 0xc(%rsp), %rcx movq 0x70(%rax), %rax imulq $0xc, %rcx, %rcx cmpl %r13d, (%rax,%rcx) jle 0x2de1fd xorl %eax, %eax jmp 0x2de205 cmpl 0x4(%rax,%rcx), %r13d setle %al addq $0x28, %rsp popq %rbx popq %r12 popq %r13 popq %r14 popq %r15 popq %rbp retq
/quesnel[P]baryonyx/lib/src/itm-common.hpp
bool baryonyx::itm::solver_inequalities_101coeff<float, baryonyx::itm::minimize_tag, baryonyx::itm::default_cost_type<float>, false>::compute_update_row<baryonyx::bit_array, std::reverse_iterator<__gnu_cxx::__normal_iterator<int const*, std::vector<int, std::allocator<int>>>>>(baryonyx::bit_array&, std::reverse_iterator<__gnu_cxx::__normal_iterator<int const*, std::vector<int, std::allocator<int>>>>, std::reverse_iterator<__gnu_cxx::__normal_iterator<int const*, std::vector<int, std::allocator<int>>>>, float, float, float)
bool compute_update_row(Xtype& x, Iterator first, Iterator last, Float kappa, Float delta, Float theta) { auto at_least_one_pi_changed{ false }; logger::log("update-row {} {} {}\n", kappa, delta, theta); for (; first != last; ++first) { auto k = constraint(first); const auto it = ap.row(k); decrease_preference(std::get<0>(it), std::get<1>(it), theta); const auto sizes = compute_reduced_costs(std::get<0>(it), std::get<1>(it), x); calculator_sort<Mode>(R.get(), R.get() + sizes.r_size, rng); int selected = select_variables(sizes, b[k].min, b[k].max); logger::log("constraints {}: {} <= ", k, b[k].min); for (int i = 0; i < sizes.r_size; ++i) logger::log("{} ({} {}) ", R[i].f, R[i].value, R[i].id); logger::log("<= {} => Selected: {}\n", b[k].max, selected); auto pi_change = affect(*this, x, std::get<0>(it), k, selected, sizes.r_size, kappa, delta); at_least_one_pi_changed = at_least_one_pi_changed || pi_change; } return at_least_one_pi_changed; }
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x38, %rsp movq (%rdx), %rax leaq 0x10(%rdi), %r13 vmovss %xmm2, 0x14(%rsp) vmovss %xmm1, 0x10(%rsp) vmovss %xmm0, 0xc(%rsp) movq %rcx, 0x30(%rsp) movq %rdx, %r14 movq %rsi, %r15 movq %rdi, %r12 xorl %ebp, %ebp movq 0x30(%rsp), %rcx cmpq (%rcx), %rax je 0x2de30b movslq -0x4(%rax), %rbx leaq 0x20(%rsp), %rdi movq %r13, %rsi movl %ebx, %edx callq 0x3de66 vmovss 0x14(%rsp), %xmm0 movq 0x20(%rsp), %rdx movq 0x28(%rsp), %rsi movq %r12, %rdi callq 0x2de77e movq 0x20(%rsp), %rdx movq 0x28(%rsp), %rsi movq %r12, %rdi movq %r15, %rcx callq 0x2de79c movq %rax, 0x18(%rsp) cltq movq 0x68(%r12), %rdi movq 0x8(%r12), %rdx imulq $0xc, %rax, %rsi addq %rdi, %rsi callq 0x2de8d9 movq 0x70(%r12), %rax imulq $0xc, %rbx, %rcx leaq 0x18(%rsp), %rsi movq %r12, %rdi movl (%rax,%rcx), %edx movl 0x4(%rax,%rcx), %ecx callq 0x2de90c vmovss 0xc(%rsp), %xmm0 vmovss 0x10(%rsp), %xmm1 movl 0x18(%rsp), %r9d movq 0x28(%rsp), %rdx movq %r12, %rdi movq %r15, %rsi movl %ebx, %ecx movl %eax, %r8d callq 0x2de97d orb %al, %bpl movq (%r14), %rax addq $-0x4, %rax movq %rax, (%r14) jmp 0x2de24b andb $0x1, %bpl movl %ebp, %eax addq $0x38, %rsp popq %rbx popq %r12 popq %r13 popq %r14 popq %r15 popq %rbp retq
/quesnel[P]baryonyx/lib/src/itm-solver-inequalities-101.cpp
bool baryonyx::itm::solver_inequalities_101coeff<float, baryonyx::itm::minimize_tag, baryonyx::itm::default_cost_type<float>, false>::compute_update_row<baryonyx::bit_array, __gnu_cxx::__normal_iterator<int*, std::vector<int, std::allocator<int>>>>(baryonyx::bit_array&, __gnu_cxx::__normal_iterator<int*, std::vector<int, std::allocator<int>>>, __gnu_cxx::__normal_iterator<int*, std::vector<int, std::allocator<int>>>, float, float, float)
bool compute_update_row(Xtype& x, Iterator first, Iterator last, Float kappa, Float delta, Float theta) { auto at_least_one_pi_changed{ false }; logger::log("update-row {} {} {}\n", kappa, delta, theta); for (; first != last; ++first) { auto k = constraint(first); const auto it = ap.row(k); decrease_preference(std::get<0>(it), std::get<1>(it), theta); const auto sizes = compute_reduced_costs(std::get<0>(it), std::get<1>(it), x); calculator_sort<Mode>(R.get(), R.get() + sizes.r_size, rng); int selected = select_variables(sizes, b[k].min, b[k].max); logger::log("constraints {}: {} <= ", k, b[k].min); for (int i = 0; i < sizes.r_size; ++i) logger::log("{} ({} {}) ", R[i].f, R[i].value, R[i].id); logger::log("<= {} => Selected: {}\n", b[k].max, selected); auto pi_change = affect(*this, x, std::get<0>(it), k, selected, sizes.r_size, kappa, delta); at_least_one_pi_changed = at_least_one_pi_changed || pi_change; } return at_least_one_pi_changed; }
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x38, %rsp leaq 0x10(%rdi), %r13 vmovss %xmm2, 0x14(%rsp) vmovss %xmm1, 0x10(%rsp) vmovss %xmm0, 0xc(%rsp) movq %rcx, 0x30(%rsp) movq %rdx, %r14 movq %rsi, %r15 movq %rdi, %r12 xorl %ebp, %ebp cmpq 0x30(%rsp), %r14 je 0x2de40a movslq (%r14), %rbx leaq 0x20(%rsp), %rdi movq %r13, %rsi movl %ebx, %edx callq 0x3de66 vmovss 0x14(%rsp), %xmm0 movq 0x20(%rsp), %rdx movq 0x28(%rsp), %rsi movq %r12, %rdi callq 0x2de77e movq 0x20(%rsp), %rdx movq 0x28(%rsp), %rsi movq %r12, %rdi movq %r15, %rcx callq 0x2de79c movq %rax, 0x18(%rsp) cltq movq 0x68(%r12), %rdi movq 0x8(%r12), %rdx imulq $0xc, %rax, %rsi addq %rdi, %rsi callq 0x2de8d9 movq 0x70(%r12), %rax imulq $0xc, %rbx, %rcx leaq 0x18(%rsp), %rsi movq %r12, %rdi movl (%rax,%rcx), %edx movl 0x4(%rax,%rcx), %ecx callq 0x2de90c vmovss 0xc(%rsp), %xmm0 vmovss 0x10(%rsp), %xmm1 movl 0x18(%rsp), %r9d movq 0x28(%rsp), %rdx movq %r12, %rdi movq %r15, %rsi movl %ebx, %ecx movl %eax, %r8d callq 0x2de97d orb %al, %bpl addq $0x4, %r14 jmp 0x2de354 andb $0x1, %bpl movl %ebp, %eax addq $0x38, %rsp popq %rbx popq %r12 popq %r13 popq %r14 popq %r15 popq %rbp retq nop
/quesnel[P]baryonyx/lib/src/itm-solver-inequalities-101.cpp
bool baryonyx::itm::solver_inequalities_101coeff<float, baryonyx::itm::minimize_tag, baryonyx::itm::default_cost_type<float>, false>::compute_update_row<baryonyx::bit_array, __gnu_cxx::__normal_iterator<std::pair<int, int> const*, std::vector<std::pair<int, int>, std::allocator<std::pair<int, int>>>>>(baryonyx::bit_array&, __gnu_cxx::__normal_iterator<std::pair<int, int> const*, std::vector<std::pair<int, int>, std::allocator<std::pair<int, int>>>>, __gnu_cxx::__normal_iterator<std::pair<int, int> const*, std::vector<std::pair<int, int>, std::allocator<std::pair<int, int>>>>, float, float, float)
bool compute_update_row(Xtype& x, Iterator first, Iterator last, Float kappa, Float delta, Float theta) { auto at_least_one_pi_changed{ false }; logger::log("update-row {} {} {}\n", kappa, delta, theta); for (; first != last; ++first) { auto k = constraint(first); const auto it = ap.row(k); decrease_preference(std::get<0>(it), std::get<1>(it), theta); const auto sizes = compute_reduced_costs(std::get<0>(it), std::get<1>(it), x); calculator_sort<Mode>(R.get(), R.get() + sizes.r_size, rng); int selected = select_variables(sizes, b[k].min, b[k].max); logger::log("constraints {}: {} <= ", k, b[k].min); for (int i = 0; i < sizes.r_size; ++i) logger::log("{} ({} {}) ", R[i].f, R[i].value, R[i].id); logger::log("<= {} => Selected: {}\n", b[k].max, selected); auto pi_change = affect(*this, x, std::get<0>(it), k, selected, sizes.r_size, kappa, delta); at_least_one_pi_changed = at_least_one_pi_changed || pi_change; } return at_least_one_pi_changed; }
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x38, %rsp leaq 0x10(%rdi), %r13 vmovss %xmm2, 0x14(%rsp) vmovss %xmm1, 0x10(%rsp) vmovss %xmm0, 0xc(%rsp) movq %rcx, 0x30(%rsp) movq %rdx, %r14 movq %rsi, %r15 movq %rdi, %r12 xorl %ebp, %ebp cmpq 0x30(%rsp), %r14 je 0x2de50a movslq (%r14), %rbx leaq 0x20(%rsp), %rdi movq %r13, %rsi movl %ebx, %edx callq 0x3de66 vmovss 0x14(%rsp), %xmm0 movq 0x20(%rsp), %rdx movq 0x28(%rsp), %rsi movq %r12, %rdi callq 0x2de77e movq 0x20(%rsp), %rdx movq 0x28(%rsp), %rsi movq %r12, %rdi movq %r15, %rcx callq 0x2de79c movq %rax, 0x18(%rsp) cltq movq 0x68(%r12), %rdi movq 0x8(%r12), %rdx imulq $0xc, %rax, %rsi addq %rdi, %rsi callq 0x2de8d9 movq 0x70(%r12), %rax imulq $0xc, %rbx, %rcx leaq 0x18(%rsp), %rsi movq %r12, %rdi movl (%rax,%rcx), %edx movl 0x4(%rax,%rcx), %ecx callq 0x2de90c vmovss 0xc(%rsp), %xmm0 vmovss 0x10(%rsp), %xmm1 movl 0x18(%rsp), %r9d movq 0x28(%rsp), %rdx movq %r12, %rdi movq %r15, %rsi movl %ebx, %ecx movl %eax, %r8d callq 0x2de97d orb %al, %bpl addq $0x8, %r14 jmp 0x2de454 andb $0x1, %bpl movl %ebp, %eax addq $0x38, %rsp popq %rbx popq %r12 popq %r13 popq %r14 popq %r15 popq %rbp retq nop
/quesnel[P]baryonyx/lib/src/itm-solver-inequalities-101.cpp
int baryonyx::itm::compute_order::infeasibility_local_compute_violated_constraints<baryonyx::itm::solver_inequalities_101coeff<float, baryonyx::itm::minimize_tag, baryonyx::itm::default_cost_type<float>, false>, baryonyx::bit_array>(baryonyx::itm::solver_inequalities_101coeff<float, baryonyx::itm::minimize_tag, baryonyx::itm::default_cost_type<float>, false>&, baryonyx::bit_array const&)
int infeasibility_local_compute_violated_constraints(Solver& solver, const Xtype& x) { m_order.clear(); for (int k = 0, e = solver.m; k != e; ++k) { sparse_matrix<int>::const_row_iterator it, et; std::tie(it, et) = solver.ap.row(k); int v = 0; for (; it != et; ++it) v += solver.factor(it->value) * x[it->column]; if (solver.bound_min(k) > v) m_order.push_back(std::make_pair(k, solver.bound_min(k) - v)); else if (solver.bound_max(k) < v) m_order.push_back(std::make_pair(k, v - solver.bound_max(k))); } return length(m_order); }
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x48, %rsp movq 0x18(%rdi), %rax movq %rdx, %r14 movq %rsi, 0x20(%rsp) cmpq %rax, 0x20(%rdi) je 0x2de544 movq %rax, 0x20(%rdi) movq 0x20(%rsp), %rsi leaq 0x18(%rdi), %rax movq %rdi, 0x28(%rsp) xorl %ebx, %ebx movq %rax, 0x18(%rsp) movl 0x88(%rsi), %ecx leaq 0x10(%rsi), %rdx leaq 0x58(%rsi), %rbp movq %rdx, 0x30(%rsp) movq %rcx, 0x38(%rsp) cmpq 0x38(%rsp), %rbx je 0x2de61b movq 0x30(%rsp), %rsi leaq 0x8(%rsp), %rdi movl %ebx, %edx movq %rbx, 0x40(%rsp) callq 0x3de66 movq 0x8(%rsp), %r15 movq 0x10(%rsp), %rbx xorl %r13d, %r13d cmpq %r15, %rbx je 0x2de5c9 movl (%rbx), %esi movq %rbp, %rdi callq 0x2b9c1a movl 0x4(%rbx), %esi movq %r14, %rdi movl %eax, %r12d callq 0x3664a imull %r12d, %eax addq $0x8, %rbx addl %eax, %r13d jmp 0x2de59f movq 0x20(%rsp), %rax movq 0x40(%rsp), %rbx movq 0x70(%rax), %rcx imulq $0xc, %rbx, %rdx movl (%rcx,%rdx), %eax subl %r13d, %eax jle 0x2de5f1 shlq $0x20, %rax orq %rbx, %rax movq %rax, 0x8(%rsp) jmp 0x2de604 subl 0x4(%rcx,%rdx), %r13d jle 0x2de613 shlq $0x20, %r13 orq %rbx, %r13 movq %r13, 0x8(%rsp) movq 0x18(%rsp), %rdi leaq 0x8(%rsp), %rsi callq 0x3c8c2 incq %rbx jmp 0x2de571 movq 0x28(%rsp), %rcx movq 0x20(%rcx), %rax subq 0x18(%rcx), %rax shrq $0x3, %rax addq $0x48, %rsp popq %rbx popq %r12 popq %r13 popq %r14 popq %r15 popq %rbp retq nop
/quesnel[P]baryonyx/lib/src/itm-common.hpp
bool baryonyx::itm::solver_inequalities_101coeff<float, baryonyx::itm::minimize_tag, baryonyx::itm::default_cost_type<float>, false>::compute_update_row<baryonyx::bit_array, __gnu_cxx::__normal_iterator<int const*, std::vector<int, std::allocator<int>>>>(baryonyx::bit_array&, __gnu_cxx::__normal_iterator<int const*, std::vector<int, std::allocator<int>>>, __gnu_cxx::__normal_iterator<int const*, std::vector<int, std::allocator<int>>>, float, float, float)
bool compute_update_row(Xtype& x, Iterator first, Iterator last, Float kappa, Float delta, Float theta) { auto at_least_one_pi_changed{ false }; logger::log("update-row {} {} {}\n", kappa, delta, theta); for (; first != last; ++first) { auto k = constraint(first); const auto it = ap.row(k); decrease_preference(std::get<0>(it), std::get<1>(it), theta); const auto sizes = compute_reduced_costs(std::get<0>(it), std::get<1>(it), x); calculator_sort<Mode>(R.get(), R.get() + sizes.r_size, rng); int selected = select_variables(sizes, b[k].min, b[k].max); logger::log("constraints {}: {} <= ", k, b[k].min); for (int i = 0; i < sizes.r_size; ++i) logger::log("{} ({} {}) ", R[i].f, R[i].value, R[i].id); logger::log("<= {} => Selected: {}\n", b[k].max, selected); auto pi_change = affect(*this, x, std::get<0>(it), k, selected, sizes.r_size, kappa, delta); at_least_one_pi_changed = at_least_one_pi_changed || pi_change; } return at_least_one_pi_changed; }
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x38, %rsp leaq 0x10(%rdi), %r13 vmovss %xmm2, 0x14(%rsp) vmovss %xmm1, 0x10(%rsp) vmovss %xmm0, 0xc(%rsp) movq %rcx, 0x30(%rsp) movq %rdx, %r14 movq %rsi, %r15 movq %rdi, %r12 xorl %ebp, %ebp cmpq 0x30(%rsp), %r14 je 0x2de726 movslq (%r14), %rbx leaq 0x20(%rsp), %rdi movq %r13, %rsi movl %ebx, %edx callq 0x3de66 vmovss 0x14(%rsp), %xmm0 movq 0x20(%rsp), %rdx movq 0x28(%rsp), %rsi movq %r12, %rdi callq 0x2de77e movq 0x20(%rsp), %rdx movq 0x28(%rsp), %rsi movq %r12, %rdi movq %r15, %rcx callq 0x2de79c movq %rax, 0x18(%rsp) cltq movq 0x68(%r12), %rdi movq 0x8(%r12), %rdx imulq $0xc, %rax, %rsi addq %rdi, %rsi callq 0x2de8d9 movq 0x70(%r12), %rax imulq $0xc, %rbx, %rcx leaq 0x18(%rsp), %rsi movq %r12, %rdi movl (%rax,%rcx), %edx movl 0x4(%rax,%rcx), %ecx callq 0x2de90c vmovss 0xc(%rsp), %xmm0 vmovss 0x10(%rsp), %xmm1 movl 0x18(%rsp), %r9d movq 0x28(%rsp), %rdx movq %r12, %rdi movq %r15, %rsi movl %ebx, %ecx movl %eax, %r8d callq 0x2de97d orb %al, %bpl addq $0x4, %r14 jmp 0x2de670 andb $0x1, %bpl movl %ebp, %eax addq $0x38, %rsp popq %rbx popq %r12 popq %r13 popq %r14 popq %r15 popq %rbp retq nop
/quesnel[P]baryonyx/lib/src/itm-solver-inequalities-101.cpp
int baryonyx::itm::compute_order::local_compute_violated_constraints<baryonyx::itm::solver_inequalities_101coeff<float, baryonyx::itm::minimize_tag, baryonyx::itm::default_cost_type<float>, false>, baryonyx::bit_array>(baryonyx::itm::solver_inequalities_101coeff<float, baryonyx::itm::minimize_tag, baryonyx::itm::default_cost_type<float>, false> const&, baryonyx::bit_array const&)
int local_compute_violated_constraints(const Solver& slv, const Xtype& x) { int remaining = 0; for (int k = 0; k != slv.m; ++k) if (!is_valid_constraint(slv, k, x)) ++remaining; return remaining; }
pushq %rbp pushq %r15 pushq %r14 pushq %rbx pushq %rax movq %rdx, %rbx movq %rsi, %r14 xorl %ebp, %ebp xorl %r15d, %r15d cmpl 0x88(%r14), %r15d je 0x2de771 movq %r14, %rdi movl %r15d, %esi movq %rbx, %rdx callq 0x2de178 xorb $0x1, %al incl %r15d movzbl %al, %eax addl %eax, %ebp jmp 0x2de74e movl %ebp, %eax addq $0x8, %rsp popq %rbx popq %r14 popq %r15 popq %rbp retq
/quesnel[P]baryonyx/lib/src/itm-common.hpp
void baryonyx::itm::random_shuffle_unique<baryonyx::itm::solver_inequalities_101coeff<float, baryonyx::itm::minimize_tag, baryonyx::itm::default_cost_type<float>, false>::rc_data*>(baryonyx::itm::solver_inequalities_101coeff<float, baryonyx::itm::minimize_tag, baryonyx::itm::default_cost_type<float>, false>::rc_data*, baryonyx::itm::solver_inequalities_101coeff<float, baryonyx::itm::minimize_tag, baryonyx::itm::default_cost_type<float>, false>::rc_data*, std::linear_congruential_engine<unsigned long, 16807ul, 0ul, 2147483647ul>&)
inline void random_shuffle_unique(iteratorT begin, iteratorT end, random_engine& rng) noexcept { auto ret = begin++; for (; begin != end; ++begin) { if (ret->value != begin->value) { std::shuffle(ret, begin, rng); ret = begin; } } std::shuffle(ret, begin, rng); }
pushq %r15 pushq %r14 pushq %rbx movq %rdx, %rbx movq %rsi, %r14 leaq 0xc(%rdi), %rsi cmpq %r14, %rsi je 0x2decb7 vmovss (%rdi), %xmm0 leaq 0xc(%rsi), %r15 vucomiss (%rsi), %xmm0 movq %r15, %rsi jne 0x2deca3 jnp 0x2dec8b addq $-0xc, %r15 movq %r15, %rsi movq %rbx, %rdx callq 0x2df24d movq %r15, %rdi jmp 0x2dec87 movq %rbx, %rdx callq 0x2df24d popq %rbx popq %r14 popq %r15 retq jmp 0x2decc7 movq %rax, %rdi callq 0xeb9f
/quesnel[P]baryonyx/lib/src/itm-common.hpp
bool baryonyx::itm::solver_inequalities_101coeff<float, baryonyx::itm::minimize_tag, baryonyx::itm::default_cost_type<float>, false>::push_and_compute_update_row<baryonyx::bit_array, std::reverse_iterator<__gnu_cxx::__normal_iterator<int const*, std::vector<int, std::allocator<int>>>>>(baryonyx::bit_array&, std::reverse_iterator<__gnu_cxx::__normal_iterator<int const*, std::vector<int, std::allocator<int>>>>, std::reverse_iterator<__gnu_cxx::__normal_iterator<int const*, std::vector<int, std::allocator<int>>>>, float, float, float, float)
bool push_and_compute_update_row(Xtype& x, Iterator first, Iterator last, Float kappa, Float delta, Float theta, Float obj_amp) { auto at_least_one_pi_changed{ false }; logger::log("push-update-row {} {} {}\n", kappa, delta, theta); for (; first != last; ++first) { auto k = constraint(first); bx_expects(k < m); const auto it = ap.row(k); decrease_preference(std::get<0>(it), std::get<1>(it), theta); const auto sizes = compute_reduced_costs(std::get<0>(it), std::get<1>(it), x); for (int i = 0; i != sizes.r_size; ++i) R[i].value += obj_amp * c((std::get<0>(it) + R[i].id)->column, x); calculator_sort<Mode>(R.get(), R.get() + sizes.r_size, rng); int selected = select_variables(sizes, b[k].min, b[k].max); logger::log("constraints {}: {} <= ", k, b[k].min); for (int i = 0; i < sizes.r_size; ++i) logger::log("{} ({} {}) ", R[i].f, R[i].value, R[i].id); logger::log("<= {} => Selected: {}\n", b[k].max, selected); auto pi_change = affect(*this, x, std::get<0>(it), k, selected, sizes.r_size, kappa, delta); at_least_one_pi_changed = at_least_one_pi_changed || pi_change; } return at_least_one_pi_changed; }
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x38, %rsp leaq 0x10(%rdi), %rax vmovss %xmm3, 0x14(%rsp) vmovss %xmm2, 0x10(%rsp) vmovss %xmm1, 0xc(%rsp) vmovss %xmm0, 0x8(%rsp) movq %rcx, %rbx movq %rdx, %r14 movq %rsi, %r15 movq %rdi, %r13 xorl %r12d, %r12d movq %rax, 0x30(%rsp) movq (%rdx), %rax cmpq (%rbx), %rax je 0x2e0832 movslq -0x4(%rax), %rbp cmpl 0x88(%r13), %ebp jge 0x2e0848 movq 0x30(%rsp), %rsi leaq 0x18(%rsp), %rdi movl %ebp, %edx callq 0x3de66 vmovss 0x10(%rsp), %xmm0 movq 0x18(%rsp), %rdx movq 0x20(%rsp), %rsi movq %r13, %rdi callq 0x2de77e movq 0x18(%rsp), %rdx movq 0x20(%rsp), %rsi movq %r13, %rdi movq %r15, %rcx callq 0x2de79c vmovss 0x14(%rsp), %xmm1 movq %rax, 0x28(%rsp) movq 0x20(%rsp), %rdx movl %eax, %esi imulq $0xc, %rsi, %rsi xorl %r8d, %r8d movq 0x68(%r13), %rdi movq 0x80(%r13), %rcx cmpq %r8, %rsi je 0x2e07cc movslq 0x8(%rdi,%r8), %r9 movq 0x8(%rcx), %r10 movslq 0x4(%rdx,%r9,8), %r9 vmovss (%r10,%r9,4), %xmm0 vfmadd213ss (%rdi,%r8), %xmm1, %xmm0 # xmm0 = (xmm1 * xmm0) + mem vmovss %xmm0, (%rdi,%r8) addq $0xc, %r8 jmp 0x2e07a1 cltq movq 0x8(%r13), %rdx imulq $0xc, %rax, %rsi addq %rdi, %rsi callq 0x2de8d9 movq 0x70(%r13), %rax imulq $0xc, %rbp, %rcx leaq 0x28(%rsp), %rsi movq %r13, %rdi movl (%rax,%rcx), %edx movl 0x4(%rax,%rcx), %ecx callq 0x2de90c vmovss 0x8(%rsp), %xmm0 vmovss 0xc(%rsp), %xmm1 movl 0x28(%rsp), %r9d movq 0x20(%rsp), %rdx movq %r13, %rdi movq %r15, %rsi movl %ebp, %ecx movl %eax, %r8d callq 0x2de97d orb %al, %r12b movq (%r14), %rax addq $-0x4, %rax movq %rax, (%r14) jmp 0x2e0725 andb $0x1, %r12b movl %r12d, %eax addq $0x38, %rsp popq %rbx popq %r12 popq %r13 popq %r14 popq %r15 popq %rbp retq leaq 0x17798f(%rip), %rdi # 0x4581de leaq 0x179c24(%rip), %rsi # 0x45a47a leaq 0x182e3b(%rip), %rdx # 0x463698 leaq 0x182ee3(%rip), %rcx # 0x463747 callq 0x2813f nop
/quesnel[P]baryonyx/lib/src/itm-solver-inequalities-101.cpp
bool baryonyx::itm::solver_inequalities_101coeff<float, baryonyx::itm::minimize_tag, baryonyx::itm::default_cost_type<float>, false>::push_and_compute_update_row<baryonyx::bit_array, __gnu_cxx::__normal_iterator<int*, std::vector<int, std::allocator<int>>>>(baryonyx::bit_array&, __gnu_cxx::__normal_iterator<int*, std::vector<int, std::allocator<int>>>, __gnu_cxx::__normal_iterator<int*, std::vector<int, std::allocator<int>>>, float, float, float, float)
bool push_and_compute_update_row(Xtype& x, Iterator first, Iterator last, Float kappa, Float delta, Float theta, Float obj_amp) { auto at_least_one_pi_changed{ false }; logger::log("push-update-row {} {} {}\n", kappa, delta, theta); for (; first != last; ++first) { auto k = constraint(first); bx_expects(k < m); const auto it = ap.row(k); decrease_preference(std::get<0>(it), std::get<1>(it), theta); const auto sizes = compute_reduced_costs(std::get<0>(it), std::get<1>(it), x); for (int i = 0; i != sizes.r_size; ++i) R[i].value += obj_amp * c((std::get<0>(it) + R[i].id)->column, x); calculator_sort<Mode>(R.get(), R.get() + sizes.r_size, rng); int selected = select_variables(sizes, b[k].min, b[k].max); logger::log("constraints {}: {} <= ", k, b[k].min); for (int i = 0; i < sizes.r_size; ++i) logger::log("{} ({} {}) ", R[i].f, R[i].value, R[i].id); logger::log("<= {} => Selected: {}\n", b[k].max, selected); auto pi_change = affect(*this, x, std::get<0>(it), k, selected, sizes.r_size, kappa, delta); at_least_one_pi_changed = at_least_one_pi_changed || pi_change; } return at_least_one_pi_changed; }
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x38, %rsp leaq 0x10(%rdi), %rax vmovss %xmm3, 0x14(%rsp) vmovss %xmm2, 0x10(%rsp) vmovss %xmm1, 0xc(%rsp) vmovss %xmm0, 0x8(%rsp) movq %rcx, %rbx movq %rdx, %r14 movq %rsi, %r12 movq %rdi, %r13 xorl %r15d, %r15d movq %rax, 0x30(%rsp) cmpq %rbx, %r14 je 0x2e09ae movslq (%r14), %rbp cmpl 0x88(%r13), %ebp jge 0x2e09c4 movq 0x30(%rsp), %rsi leaq 0x18(%rsp), %rdi movl %ebp, %edx callq 0x3de66 vmovss 0x10(%rsp), %xmm0 movq 0x18(%rsp), %rdx movq 0x20(%rsp), %rsi movq %r13, %rdi callq 0x2de77e movq 0x18(%rsp), %rdx movq 0x20(%rsp), %rsi movq %r13, %rdi movq %r12, %rcx callq 0x2de79c vmovss 0x14(%rsp), %xmm1 movq %rax, 0x28(%rsp) movq 0x20(%rsp), %rdx movl %eax, %esi imulq $0xc, %rsi, %rsi xorl %r8d, %r8d movq 0x68(%r13), %rdi movq 0x80(%r13), %rcx cmpq %r8, %rsi je 0x2e094e movslq 0x8(%rdi,%r8), %r9 movq 0x8(%rcx), %r10 movslq 0x4(%rdx,%r9,8), %r9 vmovss (%r10,%r9,4), %xmm0 vfmadd213ss (%rdi,%r8), %xmm1, %xmm0 # xmm0 = (xmm1 * xmm0) + mem vmovss %xmm0, (%rdi,%r8) addq $0xc, %r8 jmp 0x2e0923 cltq movq 0x8(%r13), %rdx imulq $0xc, %rax, %rsi addq %rdi, %rsi callq 0x2de8d9 movq 0x70(%r13), %rax imulq $0xc, %rbp, %rcx leaq 0x28(%rsp), %rsi movq %r13, %rdi movl (%rax,%rcx), %edx movl 0x4(%rax,%rcx), %ecx callq 0x2de90c vmovss 0x8(%rsp), %xmm0 vmovss 0xc(%rsp), %xmm1 movl 0x28(%rsp), %r9d movq 0x20(%rsp), %rdx movq %r13, %rdi movq %r12, %rsi movl %ebp, %ecx movl %eax, %r8d callq 0x2de97d orb %al, %r15b addq $0x4, %r14 jmp 0x2e08a8 andb $0x1, %r15b movl %r15d, %eax addq $0x38, %rsp popq %rbx popq %r12 popq %r13 popq %r14 popq %r15 popq %rbp retq leaq 0x177813(%rip), %rdi # 0x4581de leaq 0x179aa8(%rip), %rsi # 0x45a47a leaq 0x182cbf(%rip), %rdx # 0x463698 leaq 0x182d67(%rip), %rcx # 0x463747 callq 0x2813f nop
/quesnel[P]baryonyx/lib/src/itm-solver-inequalities-101.cpp
bool baryonyx::itm::solver_inequalities_101coeff<float, baryonyx::itm::minimize_tag, baryonyx::itm::default_cost_type<float>, false>::push_and_compute_update_row<baryonyx::bit_array, __gnu_cxx::__normal_iterator<std::pair<int, int> const*, std::vector<std::pair<int, int>, std::allocator<std::pair<int, int>>>>>(baryonyx::bit_array&, __gnu_cxx::__normal_iterator<std::pair<int, int> const*, std::vector<std::pair<int, int>, std::allocator<std::pair<int, int>>>>, __gnu_cxx::__normal_iterator<std::pair<int, int> const*, std::vector<std::pair<int, int>, std::allocator<std::pair<int, int>>>>, float, float, float, float)
bool push_and_compute_update_row(Xtype& x, Iterator first, Iterator last, Float kappa, Float delta, Float theta, Float obj_amp) { auto at_least_one_pi_changed{ false }; logger::log("push-update-row {} {} {}\n", kappa, delta, theta); for (; first != last; ++first) { auto k = constraint(first); bx_expects(k < m); const auto it = ap.row(k); decrease_preference(std::get<0>(it), std::get<1>(it), theta); const auto sizes = compute_reduced_costs(std::get<0>(it), std::get<1>(it), x); for (int i = 0; i != sizes.r_size; ++i) R[i].value += obj_amp * c((std::get<0>(it) + R[i].id)->column, x); calculator_sort<Mode>(R.get(), R.get() + sizes.r_size, rng); int selected = select_variables(sizes, b[k].min, b[k].max); logger::log("constraints {}: {} <= ", k, b[k].min); for (int i = 0; i < sizes.r_size; ++i) logger::log("{} ({} {}) ", R[i].f, R[i].value, R[i].id); logger::log("<= {} => Selected: {}\n", b[k].max, selected); auto pi_change = affect(*this, x, std::get<0>(it), k, selected, sizes.r_size, kappa, delta); at_least_one_pi_changed = at_least_one_pi_changed || pi_change; } return at_least_one_pi_changed; }
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x38, %rsp leaq 0x10(%rdi), %rax vmovss %xmm3, 0x14(%rsp) vmovss %xmm2, 0x10(%rsp) vmovss %xmm1, 0xc(%rsp) vmovss %xmm0, 0x8(%rsp) movq %rcx, %rbx movq %rdx, %r14 movq %rsi, %r12 movq %rdi, %r13 xorl %r15d, %r15d movq %rax, 0x30(%rsp) cmpq %rbx, %r14 je 0x2e0b2a movslq (%r14), %rbp cmpl 0x88(%r13), %ebp jge 0x2e0b40 movq 0x30(%rsp), %rsi leaq 0x18(%rsp), %rdi movl %ebp, %edx callq 0x3de66 vmovss 0x10(%rsp), %xmm0 movq 0x18(%rsp), %rdx movq 0x20(%rsp), %rsi movq %r13, %rdi callq 0x2de77e movq 0x18(%rsp), %rdx movq 0x20(%rsp), %rsi movq %r13, %rdi movq %r12, %rcx callq 0x2de79c vmovss 0x14(%rsp), %xmm1 movq %rax, 0x28(%rsp) movq 0x20(%rsp), %rdx movl %eax, %esi imulq $0xc, %rsi, %rsi xorl %r8d, %r8d movq 0x68(%r13), %rdi movq 0x80(%r13), %rcx cmpq %r8, %rsi je 0x2e0aca movslq 0x8(%rdi,%r8), %r9 movq 0x8(%rcx), %r10 movslq 0x4(%rdx,%r9,8), %r9 vmovss (%r10,%r9,4), %xmm0 vfmadd213ss (%rdi,%r8), %xmm1, %xmm0 # xmm0 = (xmm1 * xmm0) + mem vmovss %xmm0, (%rdi,%r8) addq $0xc, %r8 jmp 0x2e0a9f cltq movq 0x8(%r13), %rdx imulq $0xc, %rax, %rsi addq %rdi, %rsi callq 0x2de8d9 movq 0x70(%r13), %rax imulq $0xc, %rbp, %rcx leaq 0x28(%rsp), %rsi movq %r13, %rdi movl (%rax,%rcx), %edx movl 0x4(%rax,%rcx), %ecx callq 0x2de90c vmovss 0x8(%rsp), %xmm0 vmovss 0xc(%rsp), %xmm1 movl 0x28(%rsp), %r9d movq 0x20(%rsp), %rdx movq %r13, %rdi movq %r12, %rsi movl %ebp, %ecx movl %eax, %r8d callq 0x2de97d orb %al, %r15b addq $0x8, %r14 jmp 0x2e0a24 andb $0x1, %r15b movl %r15d, %eax addq $0x38, %rsp popq %rbx popq %r12 popq %r13 popq %r14 popq %r15 popq %rbp retq leaq 0x177697(%rip), %rdi # 0x4581de leaq 0x17992c(%rip), %rsi # 0x45a47a leaq 0x182b43(%rip), %rdx # 0x463698 leaq 0x182beb(%rip), %rcx # 0x463747 callq 0x2813f nop
/quesnel[P]baryonyx/lib/src/itm-solver-inequalities-101.cpp
bool baryonyx::itm::solver_inequalities_101coeff<float, baryonyx::itm::minimize_tag, baryonyx::itm::default_cost_type<float>, false>::push_and_compute_update_row<baryonyx::bit_array, __gnu_cxx::__normal_iterator<int const*, std::vector<int, std::allocator<int>>>>(baryonyx::bit_array&, __gnu_cxx::__normal_iterator<int const*, std::vector<int, std::allocator<int>>>, __gnu_cxx::__normal_iterator<int const*, std::vector<int, std::allocator<int>>>, float, float, float, float)
bool push_and_compute_update_row(Xtype& x, Iterator first, Iterator last, Float kappa, Float delta, Float theta, Float obj_amp) { auto at_least_one_pi_changed{ false }; logger::log("push-update-row {} {} {}\n", kappa, delta, theta); for (; first != last; ++first) { auto k = constraint(first); bx_expects(k < m); const auto it = ap.row(k); decrease_preference(std::get<0>(it), std::get<1>(it), theta); const auto sizes = compute_reduced_costs(std::get<0>(it), std::get<1>(it), x); for (int i = 0; i != sizes.r_size; ++i) R[i].value += obj_amp * c((std::get<0>(it) + R[i].id)->column, x); calculator_sort<Mode>(R.get(), R.get() + sizes.r_size, rng); int selected = select_variables(sizes, b[k].min, b[k].max); logger::log("constraints {}: {} <= ", k, b[k].min); for (int i = 0; i < sizes.r_size; ++i) logger::log("{} ({} {}) ", R[i].f, R[i].value, R[i].id); logger::log("<= {} => Selected: {}\n", b[k].max, selected); auto pi_change = affect(*this, x, std::get<0>(it), k, selected, sizes.r_size, kappa, delta); at_least_one_pi_changed = at_least_one_pi_changed || pi_change; } return at_least_one_pi_changed; }
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x38, %rsp leaq 0x10(%rdi), %rax vmovss %xmm3, 0x14(%rsp) vmovss %xmm2, 0x10(%rsp) vmovss %xmm1, 0xc(%rsp) vmovss %xmm0, 0x8(%rsp) movq %rcx, %rbx movq %rdx, %r14 movq %rsi, %r12 movq %rdi, %r13 xorl %r15d, %r15d movq %rax, 0x30(%rsp) cmpq %rbx, %r14 je 0x2e0ca6 movslq (%r14), %rbp cmpl 0x88(%r13), %ebp jge 0x2e0cbc movq 0x30(%rsp), %rsi leaq 0x18(%rsp), %rdi movl %ebp, %edx callq 0x3de66 vmovss 0x10(%rsp), %xmm0 movq 0x18(%rsp), %rdx movq 0x20(%rsp), %rsi movq %r13, %rdi callq 0x2de77e movq 0x18(%rsp), %rdx movq 0x20(%rsp), %rsi movq %r13, %rdi movq %r12, %rcx callq 0x2de79c vmovss 0x14(%rsp), %xmm1 movq %rax, 0x28(%rsp) movq 0x20(%rsp), %rdx movl %eax, %esi imulq $0xc, %rsi, %rsi xorl %r8d, %r8d movq 0x68(%r13), %rdi movq 0x80(%r13), %rcx cmpq %r8, %rsi je 0x2e0c46 movslq 0x8(%rdi,%r8), %r9 movq 0x8(%rcx), %r10 movslq 0x4(%rdx,%r9,8), %r9 vmovss (%r10,%r9,4), %xmm0 vfmadd213ss (%rdi,%r8), %xmm1, %xmm0 # xmm0 = (xmm1 * xmm0) + mem vmovss %xmm0, (%rdi,%r8) addq $0xc, %r8 jmp 0x2e0c1b cltq movq 0x8(%r13), %rdx imulq $0xc, %rax, %rsi addq %rdi, %rsi callq 0x2de8d9 movq 0x70(%r13), %rax imulq $0xc, %rbp, %rcx leaq 0x28(%rsp), %rsi movq %r13, %rdi movl (%rax,%rcx), %edx movl 0x4(%rax,%rcx), %ecx callq 0x2de90c vmovss 0x8(%rsp), %xmm0 vmovss 0xc(%rsp), %xmm1 movl 0x28(%rsp), %r9d movq 0x20(%rsp), %rdx movq %r13, %rdi movq %r12, %rsi movl %ebp, %ecx movl %eax, %r8d callq 0x2de97d orb %al, %r15b addq $0x4, %r14 jmp 0x2e0ba0 andb $0x1, %r15b movl %r15d, %eax addq $0x38, %rsp popq %rbx popq %r12 popq %r13 popq %r14 popq %r15 popq %rbp retq leaq 0x17751b(%rip), %rdi # 0x4581de leaq 0x1797b0(%rip), %rsi # 0x45a47a leaq 0x1829c7(%rip), %rdx # 0x463698 leaq 0x182a6f(%rip), %rcx # 0x463747 callq 0x2813f
/quesnel[P]baryonyx/lib/src/itm-solver-inequalities-101.cpp
baryonyx::itm::solver_functor<baryonyx::itm::solver_inequalities_101coeff<float, baryonyx::itm::minimize_tag, baryonyx::itm::default_cost_type<float>, false>, float, baryonyx::itm::minimize_tag, baryonyx::itm::default_cost_type<float>, baryonyx::pnm_observer>::operator()(std::vector<baryonyx::itm::merged_constraint, std::allocator<baryonyx::itm::merged_constraint>> const&, int, baryonyx::itm::default_cost_type<float> const&, double)
result operator()(const std::vector<merged_constraint>& constraints, int variables, const Cost& original_costs, double cost_constant) { result r; bit_array x(variables); int best_remaining = INT_MAX; auto& p = m_ctx.parameters; auto norm_costs = normalize_costs<Float, Cost>( m_ctx, original_costs, m_rng, variables); const auto kappa_step = static_cast<Float>(p.kappa_step); const auto kappa_max = static_cast<Float>(p.kappa_max); const auto alpha = static_cast<Float>(p.alpha); const auto theta = static_cast<Float>(p.theta); const auto delta = p.delta < 0 ? compute_delta<Float, Cost>(m_ctx, norm_costs, theta, variables) : static_cast<Float>(p.delta); const auto pushing_k_factor = static_cast<Float>(p.pushing_k_factor); const auto pushing_objective_amplifier = static_cast<Float>(p.pushing_objective_amplifier); const long int w_limit = static_cast<long int>(p.w); Solver slv( m_rng, length(constraints), variables, norm_costs, constraints); compute_order compute(p.order, variables); { std::bernoulli_distribution choose_mutation( m_ctx.parameters.init_policy_random); bit_array empty_x; switch (p.init_policy) { case solver_parameters::init_policy_type::pessimistic_solve: init_with_pre_solve<Cost, Mode>( x, empty_x, m_rng, original_costs, constraints, 1.0); break; case solver_parameters::init_policy_type::optimistic_solve: init_with_pre_solve<Cost, Mode>( empty_x, x, m_rng, original_costs, constraints, 1.0); break; case solver_parameters::init_policy_type::bastert: init_with_bastert<Cost, Mode>(x, original_costs, variables, 0); break; } for (int i = 0, e = x.size(); i != e; ++i) if (choose_mutation(m_rng)) x.invert(i); } bool start_push = false; auto kappa = static_cast<Float>(p.kappa_min); Observer obs("img", slv.m, slv.n, p.limit); m_begin = std::chrono::steady_clock::now(); m_end = std::chrono::steady_clock::now(); compute.init(slv, x); for (long int i = 0; i != p.limit; ++i) { auto remaining = compute.run(slv, x, m_rng, kappa, delta, theta); obs.make_observation(slv.ap, slv.P.get(), slv.pi.get()); if (remaining == 0) { store_if_better( x, original_costs.results(x, cost_constant), i); best_remaining = remaining; start_push = true; break; } if (remaining < best_remaining) { store_if_better(x, remaining, i); best_remaining = remaining; } if (i > w_limit) kappa += kappa_step * std::pow(static_cast<Float>(remaining) / static_cast<Float>(slv.m), alpha); if (kappa > kappa_max) { r.status = result_status::kappa_max_reached; break; } if (is_timelimit_reached()) { r.status = result_status::time_limit_reached; break; } } if (!start_push) { r.status = result_status::limit_reached; } else { for (int push = 0; push < p.pushes_limit; ++push) { auto remaining = compute.push_and_run(slv, x, m_rng, pushing_k_factor * kappa, delta, theta, pushing_objective_amplifier); if (remaining == 0) store_if_better(x, original_costs.results(x, cost_constant), -push * p.pushing_iteration_limit - 1); if (is_timelimit_reached()) break; for (int iter = 0; iter < p.pushing_iteration_limit; ++iter) { remaining = compute.run(slv, x, m_rng, kappa, delta, theta); if (remaining == 0) { store_if_better( x, original_costs.results(x, cost_constant), -push * p.pushing_iteration_limit - iter - 1); break; } if (iter > p.w) kappa += kappa_step * std::pow(static_cast<Float>(remaining) / static_cast<Float>(slv.m), alpha); if (kappa > kappa_max) break; if (is_timelimit_reached()) break; } } } if (m_best.remaining_constraints == 0) r.status = result_status::success; if (!m_best.x.empty()) { r.solutions.resize(1); convert(m_best, r.solutions[0], variables); } return r; }
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x3b8, %rsp # imm = 0x3B8 vmovsd %xmm0, 0x60(%rsp) vxorps %xmm0, %xmm0, %xmm0 movq %r8, %r13 movl %ecx, %ebx movq %rdx, %r15 movq %rsi, %r14 movq %rdi, 0x20(%rsp) vmovups %zmm0, 0x58(%rdi) vmovups %zmm0, 0x40(%rdi) vmovups %zmm0, (%rdi) movq $0x7fffffff, 0x98(%rdi) # imm = 0x7FFFFFFF movl $0x2, 0xa0(%rdi) leaq 0x30(%rsp), %rdi movl %ecx, %esi vzeroupper callq 0x368ae movq 0x10(%r14), %r12 movq 0x18(%r14), %rcx leaq 0x88(%rsp), %rdi movq %r12, %rsi movq %r13, %rdx movl %ebx, %r8d callq 0x38959 vmovsd 0x20(%r12), %xmm0 vmovsd 0x10(%r12), %xmm2 vxorpd %xmm1, %xmm1, %xmm1 vmovsd %xmm0, 0x50(%rsp) vmovsd 0x28(%r12), %xmm0 vucomisd %xmm2, %xmm1 vmovsd %xmm0, 0x48(%rsp) vmovsd 0x30(%r12), %xmm0 vmovsd %xmm0, 0x40(%rsp) vcvtsd2ss 0x8(%r12), %xmm3, %xmm0 vmovss %xmm0, 0xc(%rsp) jbe 0x2e20a8 movq 0x10(%r14), %rdi leaq 0x88(%rsp), %rsi movl %ebx, %edx callq 0x38a37 vmovss %xmm0, 0x1c(%rsp) jmp 0x2e20b2 vcvtsd2ss %xmm2, %xmm2, %xmm0 vmovss %xmm0, 0x1c(%rsp) vcvttsd2si 0xa0(%r12), %rax vmovsd 0x38(%r12), %xmm0 vmovsd 0x40(%r12), %xmm1 movq 0x18(%r14), %rsi movq %rax, 0x78(%rsp) movq 0x8(%r15), %rax vmovsd %xmm0, 0x70(%rsp) vmovsd %xmm1, 0x68(%rsp) subq (%r15), %rax pushq $0x28 popq %rcx cqto idivq %rcx leaq 0xd0(%rsp), %rdi leaq 0x88(%rsp), %r8 movl %eax, %edx movl %ebx, %ecx movq %r15, %r9 callq 0x2dd7b2 movl 0xcc(%r12), %esi leaq 0x98(%rsp), %rdi movl %ebx, %edx callq 0x38cd8 movq 0x10(%r14), %rax movl %ebx, 0x84(%rsp) vmovsd 0x48(%rax), %xmm0 vmovsd %xmm0, 0x10(%rsp) vxorps %xmm0, %xmm0, %xmm0 vmovaps %xmm0, 0x160(%rsp) movl 0xd4(%r12), %eax testl %eax, %eax je 0x2e2193 cmpl $0x1, %eax je 0x2e216d cmpl $0x2, %eax jne 0x2e21a4 movq 0x18(%r14), %rdx leaq 0x160(%rsp), %rdi leaq 0x30(%rsp), %rsi jmp 0x2e217e movq 0x18(%r14), %rdx leaq 0x30(%rsp), %rdi leaq 0x160(%rsp), %rsi vmovsd 0x171eb2(%rip), %xmm0 # 0x454038 movq %r13, %rcx movq %r15, %r8 callq 0x4f9c9 jmp 0x2e21a4 leaq 0x30(%rsp), %rdi movq %r13, %rsi movl %ebx, %edx xorl %ecx, %ecx callq 0x4f95d vcvtsd2ss 0x50(%rsp), %xmm3, %xmm0 leaq 0x30(%rsp), %r15 movq %r13, 0x58(%rsp) xorl %ebx, %ebx movl (%r15), %ebp vmovss %xmm0, 0x2c(%rsp) vcvtsd2ss 0x48(%rsp), %xmm3, %xmm0 vmovss %xmm0, 0x50(%rsp) vcvtsd2ss 0x40(%rsp), %xmm3, %xmm0 vmovss %xmm0, 0x48(%rsp) vcvtsd2ss 0x70(%rsp), %xmm3, %xmm0 vmovss %xmm0, 0x70(%rsp) vcvtsd2ss 0x68(%rsp), %xmm3, %xmm0 vmovss %xmm0, 0x68(%rsp) cmpl %ebx, %ebp je 0x2e2216 movq 0x18(%r14), %rdi callq 0x36ac3 vmovsd 0x10(%rsp), %xmm1 vucomisd %xmm0, %xmm1 jbe 0x2e2212 movq %r15, %rdi movl %ebx, %esi callq 0x3626e incl %ebx jmp 0x2e21ef leaq 0x168(%rsp), %rdi callq 0x357fe vmovsd 0x18(%r12), %xmm0 movl 0x158(%rsp), %ecx movl 0x15c(%rsp), %r8d movq 0xb0(%r12), %r9 vmovsd %xmm0, 0x10(%rsp) leaq 0x178329(%rip), %rdx # 0x45a577 leaq 0x160(%rsp), %rdi pushq $0x3 popq %rsi callq 0x42b5a callq 0xb5a0 movq %rax, (%r14) callq 0xb5a0 movq %rax, 0x8(%r14) leaq 0x98(%rsp), %rdi leaq 0xd0(%rsp), %rsi leaq 0x30(%rsp), %rdx callq 0x2dda48 vcvtsd2ss 0x10(%rsp), %xmm3, %xmm0 vmovss 0xc(%rsp), %xmm2 movl $0x7fffffff, %ebp # imm = 0x7FFFFFFF leaq 0x30(%rsp), %r13 xorl %r15d, %r15d cmpq 0xb0(%r12), %r15 je 0x2e2387 movq 0x18(%r14), %rcx vmovss 0x1c(%rsp), %xmm1 leaq 0x98(%rsp), %rdi leaq 0xd0(%rsp), %rsi movq %r13, %rdx vmovss %xmm0, 0x10(%rsp) callq 0x2dda8a movq 0x120(%rsp), %rdx movq 0x148(%rsp), %rcx movl %eax, %ebx leaq 0x160(%rsp), %rdi leaq 0xe0(%rsp), %rsi callq 0x42bbe testl %ebx, %ebx je 0x2e242f cmpl %ebp, %ebx jge 0x2e231d movq %r14, %rdi movq %r13, %rsi movl %ebx, %edx movq %r15, %rcx callq 0x2e344a movl %ebx, %ebp cmpq 0x78(%rsp), %r15 jle 0x2e2353 vcvtsi2ss %ebx, %xmm3, %xmm0 vcvtsi2ssl 0x158(%rsp), %xmm3, %xmm1 vdivss %xmm1, %xmm0, %xmm0 vmovss 0x48(%rsp), %xmm1 callq 0xb530 vmovss 0x10(%rsp), %xmm1 vfmadd231ss 0x2c(%rsp), %xmm0, %xmm1 # xmm1 = (xmm0 * mem) + xmm1 vmovaps %xmm1, %xmm0 jmp 0x2e2359 vmovss 0x10(%rsp), %xmm0 vucomiss 0x50(%rsp), %xmm0 ja 0x2e2387 movq %r14, %rdi vmovss %xmm0, 0x10(%rsp) callq 0x2e3482 vmovss 0x10(%rsp), %xmm0 vmovss 0xc(%rsp), %xmm2 testb %al, %al jne 0x2e2387 incq %r15 jmp 0x2e22a2 movq 0x20(%rsp), %rax movl $0x5, 0xa0(%rax) cmpl $0x0, 0x50(%r14) jne 0x2e23a9 movq 0x20(%rsp), %rax andl $0x0, 0xa0(%rax) cmpq $0x0, 0x28(%r14) je 0x2e23da movq 0x20(%rsp), %rax leaq 0x68(%rax), %rbx pushq $0x1 popq %rsi movq %rbx, %rdi callq 0x35292 movq (%rbx), %rsi addq $0x20, %r14 movl 0x84(%rsp), %edx movq %r14, %rdi callq 0x4f485 leaq 0x160(%rsp), %rdi callq 0x42cbc leaq 0x98(%rsp), %rdi callq 0x393ac leaq 0xd0(%rsp), %rdi callq 0x2ddeea leaq 0x90(%rsp), %rdi callq 0x35c1a leaq 0x38(%rsp), %rdi callq 0x357fe movq 0x20(%rsp), %rax addq $0x3b8, %rsp # imm = 0x3B8 popq %rbx popq %r12 popq %r13 popq %r14 popq %r15 popq %rbp retq vmovsd 0x60(%rsp), %xmm0 movq 0x58(%rsp), %rdi leaq 0x30(%rsp), %rbx movq %rbx, %rsi callq 0x365d4 movq %r14, %rdi movq %rbx, %rsi movq %r15, %rdx callq 0x2e33fc vmovss 0xc(%rsp), %xmm2 vmovss 0x10(%rsp), %xmm0 leaq 0xd0(%rsp), %r15 leaq 0x30(%rsp), %rbp xorl %ebx, %ebx movl $0x0, 0x40(%rsp) cmpl 0xbc(%r12), %ebx jge 0x2e2396 vmovss %xmm0, 0x10(%rsp) vmulss 0x70(%rsp), %xmm0, %xmm0 movq 0x18(%r14), %rcx vmovss 0x1c(%rsp), %xmm1 vmovss 0x68(%rsp), %xmm3 leaq 0x98(%rsp), %rdi movq %r15, %rsi movq %rbp, %rdx callq 0x2ddc96 testl %eax, %eax jne 0x2e24e7 vmovsd 0x60(%rsp), %xmm0 movq 0x58(%rsp), %rdi movq %rbp, %rsi callq 0x365d4 movl 0xc0(%r12), %eax imull %ebx, %eax notl %eax movslq %eax, %rdx movq %r14, %rdi movq %rbp, %rsi callq 0x2e33fc movq %r14, %rdi callq 0x2e3482 vmovss 0xc(%rsp), %xmm2 vmovss 0x10(%rsp), %xmm0 testb %al, %al jne 0x2e2396 movl %ebx, 0x78(%rsp) pushq $0x1 popq %rax movl %eax, %ebx leal -0x1(%rbx), %r13d cmpl 0xc0(%r12), %r13d jge 0x2e260e movq 0x18(%r14), %rcx vmovss 0x1c(%rsp), %xmm1 leaq 0x98(%rsp), %rdi movq %r15, %rsi movq %rbp, %rdx vmovss %xmm0, 0x10(%rsp) callq 0x2dda8a testl %eax, %eax je 0x2e25c4 vcvtsi2sd %r13d, %xmm4, %xmm0 vucomisd 0xa0(%r12), %xmm0 jbe 0x2e258b vcvtsi2ss %eax, %xmm4, %xmm0 vcvtsi2ssl 0x158(%rsp), %xmm4, %xmm1 vdivss %xmm1, %xmm0, %xmm0 vmovss 0x48(%rsp), %xmm1 callq 0xb530 vmovss 0x10(%rsp), %xmm1 vmovss 0xc(%rsp), %xmm2 vfmadd231ss 0x2c(%rsp), %xmm0, %xmm1 # xmm1 = (xmm0 * mem) + xmm1 vmovaps %xmm1, %xmm0 jmp 0x2e2597 vmovss 0xc(%rsp), %xmm2 vmovss 0x10(%rsp), %xmm0 vucomiss 0x50(%rsp), %xmm0 ja 0x2e260e movq %r14, %rdi vmovss %xmm0, 0x10(%rsp) callq 0x2e3482 testb %al, %al jne 0x2e2602 vmovss 0xc(%rsp), %xmm2 vmovss 0x10(%rsp), %xmm0 incl %ebx jmp 0x2e250c vmovsd 0x60(%rsp), %xmm0 movq 0x58(%rsp), %rdi movq %rbp, %rsi callq 0x365d4 movl 0xc0(%r12), %eax imull 0x40(%rsp), %eax subl %ebx, %eax movslq %eax, %rdx movq %r14, %rdi movq %rbp, %rsi callq 0x2e33fc vmovss 0xc(%rsp), %xmm2 vmovss 0x10(%rsp), %xmm0 jmp 0x2e260e vmovss 0xc(%rsp), %xmm2 vmovss 0x10(%rsp), %xmm0 movl 0x78(%rsp), %ebx decl 0x40(%rsp) incl %ebx jmp 0x2e2478 jmp 0x2e2654 jmp 0x2e262d jmp 0x2e2654 movq %rax, %r14 jmp 0x2e2664 movq %rax, %r14 jmp 0x2e2671 movq %rax, %r14 jmp 0x2e267e movq %rax, %r14 jmp 0x2e268b movq %rax, %r14 jmp 0x2e2695 jmp 0x2e2654 jmp 0x2e2654 jmp 0x2e2654 leaq 0x168(%rsp), %rdi movq %rax, %r14 callq 0x357fe jmp 0x2e2664 leaq 0x160(%rsp), %rdi movq %rax, %r14 callq 0x42cbc leaq 0x98(%rsp), %rdi callq 0x393ac leaq 0xd0(%rsp), %rdi callq 0x2ddeea leaq 0x90(%rsp), %rdi callq 0x35c1a leaq 0x38(%rsp), %rdi callq 0x357fe movq 0x20(%rsp), %rdi callq 0xea60 movq %r14, %rdi callq 0xb3d0 nop
/quesnel[P]baryonyx/lib/src/itm-solver-common.hpp
baryonyx::itm::solver_functor<baryonyx::itm::solver_inequalities_101coeff<float, baryonyx::itm::minimize_tag, baryonyx::itm::default_cost_type<float>, false>, float, baryonyx::itm::minimize_tag, baryonyx::itm::default_cost_type<float>, baryonyx::file_observer>::operator()(std::vector<baryonyx::itm::merged_constraint, std::allocator<baryonyx::itm::merged_constraint>> const&, int, baryonyx::itm::default_cost_type<float> const&, double)
result operator()(const std::vector<merged_constraint>& constraints, int variables, const Cost& original_costs, double cost_constant) { result r; bit_array x(variables); int best_remaining = INT_MAX; auto& p = m_ctx.parameters; auto norm_costs = normalize_costs<Float, Cost>( m_ctx, original_costs, m_rng, variables); const auto kappa_step = static_cast<Float>(p.kappa_step); const auto kappa_max = static_cast<Float>(p.kappa_max); const auto alpha = static_cast<Float>(p.alpha); const auto theta = static_cast<Float>(p.theta); const auto delta = p.delta < 0 ? compute_delta<Float, Cost>(m_ctx, norm_costs, theta, variables) : static_cast<Float>(p.delta); const auto pushing_k_factor = static_cast<Float>(p.pushing_k_factor); const auto pushing_objective_amplifier = static_cast<Float>(p.pushing_objective_amplifier); const long int w_limit = static_cast<long int>(p.w); Solver slv( m_rng, length(constraints), variables, norm_costs, constraints); compute_order compute(p.order, variables); { std::bernoulli_distribution choose_mutation( m_ctx.parameters.init_policy_random); bit_array empty_x; switch (p.init_policy) { case solver_parameters::init_policy_type::pessimistic_solve: init_with_pre_solve<Cost, Mode>( x, empty_x, m_rng, original_costs, constraints, 1.0); break; case solver_parameters::init_policy_type::optimistic_solve: init_with_pre_solve<Cost, Mode>( empty_x, x, m_rng, original_costs, constraints, 1.0); break; case solver_parameters::init_policy_type::bastert: init_with_bastert<Cost, Mode>(x, original_costs, variables, 0); break; } for (int i = 0, e = x.size(); i != e; ++i) if (choose_mutation(m_rng)) x.invert(i); } bool start_push = false; auto kappa = static_cast<Float>(p.kappa_min); Observer obs("img", slv.m, slv.n, p.limit); m_begin = std::chrono::steady_clock::now(); m_end = std::chrono::steady_clock::now(); compute.init(slv, x); for (long int i = 0; i != p.limit; ++i) { auto remaining = compute.run(slv, x, m_rng, kappa, delta, theta); obs.make_observation(slv.ap, slv.P.get(), slv.pi.get()); if (remaining == 0) { store_if_better( x, original_costs.results(x, cost_constant), i); best_remaining = remaining; start_push = true; break; } if (remaining < best_remaining) { store_if_better(x, remaining, i); best_remaining = remaining; } if (i > w_limit) kappa += kappa_step * std::pow(static_cast<Float>(remaining) / static_cast<Float>(slv.m), alpha); if (kappa > kappa_max) { r.status = result_status::kappa_max_reached; break; } if (is_timelimit_reached()) { r.status = result_status::time_limit_reached; break; } } if (!start_push) { r.status = result_status::limit_reached; } else { for (int push = 0; push < p.pushes_limit; ++push) { auto remaining = compute.push_and_run(slv, x, m_rng, pushing_k_factor * kappa, delta, theta, pushing_objective_amplifier); if (remaining == 0) store_if_better(x, original_costs.results(x, cost_constant), -push * p.pushing_iteration_limit - 1); if (is_timelimit_reached()) break; for (int iter = 0; iter < p.pushing_iteration_limit; ++iter) { remaining = compute.run(slv, x, m_rng, kappa, delta, theta); if (remaining == 0) { store_if_better( x, original_costs.results(x, cost_constant), -push * p.pushing_iteration_limit - iter - 1); break; } if (iter > p.w) kappa += kappa_step * std::pow(static_cast<Float>(remaining) / static_cast<Float>(slv.m), alpha); if (kappa > kappa_max) break; if (is_timelimit_reached()) break; } } } if (m_best.remaining_constraints == 0) r.status = result_status::success; if (!m_best.x.empty()) { r.solutions.resize(1); convert(m_best, r.solutions[0], variables); } return r; }
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x3b8, %rsp # imm = 0x3B8 vmovsd %xmm0, 0x60(%rsp) vxorps %xmm0, %xmm0, %xmm0 movq %r8, %r13 movl %ecx, %ebx movq %rdx, %r15 movq %rsi, %r14 movq %rdi, 0x20(%rsp) vmovups %zmm0, 0x58(%rdi) vmovups %zmm0, 0x40(%rdi) vmovups %zmm0, (%rdi) movq $0x7fffffff, 0x98(%rdi) # imm = 0x7FFFFFFF movl $0x2, 0xa0(%rdi) leaq 0x30(%rsp), %rdi movl %ecx, %esi vzeroupper callq 0x368ae movq 0x10(%r14), %r12 movq 0x18(%r14), %rcx leaq 0x88(%rsp), %rdi movq %r12, %rsi movq %r13, %rdx movl %ebx, %r8d callq 0x38959 vmovsd 0x20(%r12), %xmm0 vmovsd 0x10(%r12), %xmm2 vxorpd %xmm1, %xmm1, %xmm1 vmovsd %xmm0, 0x50(%rsp) vmovsd 0x28(%r12), %xmm0 vucomisd %xmm2, %xmm1 vmovsd %xmm0, 0x48(%rsp) vmovsd 0x30(%r12), %xmm0 vmovsd %xmm0, 0x40(%rsp) vcvtsd2ss 0x8(%r12), %xmm3, %xmm0 vmovss %xmm0, 0xc(%rsp) jbe 0x2e278c movq 0x10(%r14), %rdi leaq 0x88(%rsp), %rsi movl %ebx, %edx callq 0x38a37 vmovss %xmm0, 0x1c(%rsp) jmp 0x2e2796 vcvtsd2ss %xmm2, %xmm2, %xmm0 vmovss %xmm0, 0x1c(%rsp) vcvttsd2si 0xa0(%r12), %rax vmovsd 0x38(%r12), %xmm0 vmovsd 0x40(%r12), %xmm1 movq 0x18(%r14), %rsi movq %rax, 0x78(%rsp) movq 0x8(%r15), %rax vmovsd %xmm0, 0x70(%rsp) vmovsd %xmm1, 0x68(%rsp) subq (%r15), %rax pushq $0x28 popq %rcx cqto idivq %rcx leaq 0xd0(%rsp), %rdi leaq 0x88(%rsp), %r8 movl %eax, %edx movl %ebx, %ecx movq %r15, %r9 callq 0x2dd7b2 movl 0xcc(%r12), %esi leaq 0x98(%rsp), %rdi movl %ebx, %edx callq 0x38cd8 movq 0x10(%r14), %rax movl %ebx, 0x84(%rsp) vmovsd 0x48(%rax), %xmm0 vmovsd %xmm0, 0x10(%rsp) vxorps %xmm0, %xmm0, %xmm0 vmovaps %xmm0, 0x160(%rsp) movl 0xd4(%r12), %eax testl %eax, %eax je 0x2e2877 cmpl $0x1, %eax je 0x2e2851 cmpl $0x2, %eax jne 0x2e2888 movq 0x18(%r14), %rdx leaq 0x160(%rsp), %rdi leaq 0x30(%rsp), %rsi jmp 0x2e2862 movq 0x18(%r14), %rdx leaq 0x30(%rsp), %rdi leaq 0x160(%rsp), %rsi vmovsd 0x1717ce(%rip), %xmm0 # 0x454038 movq %r13, %rcx movq %r15, %r8 callq 0x4f9c9 jmp 0x2e2888 leaq 0x30(%rsp), %rdi movq %r13, %rsi movl %ebx, %edx xorl %ecx, %ecx callq 0x4f95d vcvtsd2ss 0x50(%rsp), %xmm3, %xmm0 leaq 0x30(%rsp), %r15 movq %r13, 0x58(%rsp) xorl %ebx, %ebx movl (%r15), %ebp vmovss %xmm0, 0x2c(%rsp) vcvtsd2ss 0x48(%rsp), %xmm3, %xmm0 vmovss %xmm0, 0x50(%rsp) vcvtsd2ss 0x40(%rsp), %xmm3, %xmm0 vmovss %xmm0, 0x48(%rsp) vcvtsd2ss 0x70(%rsp), %xmm3, %xmm0 vmovss %xmm0, 0x70(%rsp) vcvtsd2ss 0x68(%rsp), %xmm3, %xmm0 vmovss %xmm0, 0x68(%rsp) cmpl %ebx, %ebp je 0x2e28fa movq 0x18(%r14), %rdi callq 0x36ac3 vmovsd 0x10(%rsp), %xmm1 vucomisd %xmm0, %xmm1 jbe 0x2e28f6 movq %r15, %rdi movl %ebx, %esi callq 0x3626e incl %ebx jmp 0x2e28d3 leaq 0x168(%rsp), %rdi callq 0x357fe vmovsd 0x18(%r12), %xmm0 movl 0x158(%rsp), %ecx movl 0x15c(%rsp), %r8d movq 0xb0(%r12), %r9 vmovsd %xmm0, 0x10(%rsp) leaq 0x177c45(%rip), %rdx # 0x45a577 leaq 0x160(%rsp), %rdi pushq $0x3 popq %rsi callq 0x4346e callq 0xb5a0 movq %rax, (%r14) callq 0xb5a0 movq %rax, 0x8(%r14) leaq 0x98(%rsp), %rdi leaq 0xd0(%rsp), %rsi leaq 0x30(%rsp), %rdx callq 0x2dda48 vcvtsd2ss 0x10(%rsp), %xmm3, %xmm0 vmovss 0xc(%rsp), %xmm2 movl $0x7fffffff, %ebp # imm = 0x7FFFFFFF leaq 0x30(%rsp), %r13 xorl %r15d, %r15d cmpq 0xb0(%r12), %r15 je 0x2e2a6b movq 0x18(%r14), %rcx vmovss 0x1c(%rsp), %xmm1 leaq 0x98(%rsp), %rdi leaq 0xd0(%rsp), %rsi movq %r13, %rdx vmovss %xmm0, 0x10(%rsp) callq 0x2dda8a movq 0x120(%rsp), %rdx movq 0x148(%rsp), %rcx movl %eax, %ebx leaq 0x160(%rsp), %rdi leaq 0xe0(%rsp), %rsi callq 0x434d2 testl %ebx, %ebx je 0x2e2b13 cmpl %ebp, %ebx jge 0x2e2a01 movq %r14, %rdi movq %r13, %rsi movl %ebx, %edx movq %r15, %rcx callq 0x2e3530 movl %ebx, %ebp cmpq 0x78(%rsp), %r15 jle 0x2e2a37 vcvtsi2ss %ebx, %xmm3, %xmm0 vcvtsi2ssl 0x158(%rsp), %xmm3, %xmm1 vdivss %xmm1, %xmm0, %xmm0 vmovss 0x48(%rsp), %xmm1 callq 0xb530 vmovss 0x10(%rsp), %xmm1 vfmadd231ss 0x2c(%rsp), %xmm0, %xmm1 # xmm1 = (xmm0 * mem) + xmm1 vmovaps %xmm1, %xmm0 jmp 0x2e2a3d vmovss 0x10(%rsp), %xmm0 vucomiss 0x50(%rsp), %xmm0 ja 0x2e2a6b movq %r14, %rdi vmovss %xmm0, 0x10(%rsp) callq 0x2e3568 vmovss 0x10(%rsp), %xmm0 vmovss 0xc(%rsp), %xmm2 testb %al, %al jne 0x2e2a6b incq %r15 jmp 0x2e2986 movq 0x20(%rsp), %rax movl $0x5, 0xa0(%rax) cmpl $0x0, 0x50(%r14) jne 0x2e2a8d movq 0x20(%rsp), %rax andl $0x0, 0xa0(%rax) cmpq $0x0, 0x28(%r14) je 0x2e2abe movq 0x20(%rsp), %rax leaq 0x68(%rax), %rbx pushq $0x1 popq %rsi movq %rbx, %rdi callq 0x35292 movq (%rbx), %rsi addq $0x20, %r14 movl 0x84(%rsp), %edx movq %r14, %rdi callq 0x4f485 leaq 0x160(%rsp), %rdi callq 0x435d0 leaq 0x98(%rsp), %rdi callq 0x393ac leaq 0xd0(%rsp), %rdi callq 0x2ddeea leaq 0x90(%rsp), %rdi callq 0x35c1a leaq 0x38(%rsp), %rdi callq 0x357fe movq 0x20(%rsp), %rax addq $0x3b8, %rsp # imm = 0x3B8 popq %rbx popq %r12 popq %r13 popq %r14 popq %r15 popq %rbp retq vmovsd 0x60(%rsp), %xmm0 movq 0x58(%rsp), %rdi leaq 0x30(%rsp), %rbx movq %rbx, %rsi callq 0x365d4 movq %r14, %rdi movq %rbx, %rsi movq %r15, %rdx callq 0x2e34e2 vmovss 0xc(%rsp), %xmm2 vmovss 0x10(%rsp), %xmm0 leaq 0xd0(%rsp), %r15 leaq 0x30(%rsp), %rbp xorl %ebx, %ebx movl $0x0, 0x40(%rsp) cmpl 0xbc(%r12), %ebx jge 0x2e2a7a vmovss %xmm0, 0x10(%rsp) vmulss 0x70(%rsp), %xmm0, %xmm0 movq 0x18(%r14), %rcx vmovss 0x1c(%rsp), %xmm1 vmovss 0x68(%rsp), %xmm3 leaq 0x98(%rsp), %rdi movq %r15, %rsi movq %rbp, %rdx callq 0x2ddc96 testl %eax, %eax jne 0x2e2bcb vmovsd 0x60(%rsp), %xmm0 movq 0x58(%rsp), %rdi movq %rbp, %rsi callq 0x365d4 movl 0xc0(%r12), %eax imull %ebx, %eax notl %eax movslq %eax, %rdx movq %r14, %rdi movq %rbp, %rsi callq 0x2e34e2 movq %r14, %rdi callq 0x2e3568 vmovss 0xc(%rsp), %xmm2 vmovss 0x10(%rsp), %xmm0 testb %al, %al jne 0x2e2a7a movl %ebx, 0x78(%rsp) pushq $0x1 popq %rax movl %eax, %ebx leal -0x1(%rbx), %r13d cmpl 0xc0(%r12), %r13d jge 0x2e2cf2 movq 0x18(%r14), %rcx vmovss 0x1c(%rsp), %xmm1 leaq 0x98(%rsp), %rdi movq %r15, %rsi movq %rbp, %rdx vmovss %xmm0, 0x10(%rsp) callq 0x2dda8a testl %eax, %eax je 0x2e2ca8 vcvtsi2sd %r13d, %xmm4, %xmm0 vucomisd 0xa0(%r12), %xmm0 jbe 0x2e2c6f vcvtsi2ss %eax, %xmm4, %xmm0 vcvtsi2ssl 0x158(%rsp), %xmm4, %xmm1 vdivss %xmm1, %xmm0, %xmm0 vmovss 0x48(%rsp), %xmm1 callq 0xb530 vmovss 0x10(%rsp), %xmm1 vmovss 0xc(%rsp), %xmm2 vfmadd231ss 0x2c(%rsp), %xmm0, %xmm1 # xmm1 = (xmm0 * mem) + xmm1 vmovaps %xmm1, %xmm0 jmp 0x2e2c7b vmovss 0xc(%rsp), %xmm2 vmovss 0x10(%rsp), %xmm0 vucomiss 0x50(%rsp), %xmm0 ja 0x2e2cf2 movq %r14, %rdi vmovss %xmm0, 0x10(%rsp) callq 0x2e3568 testb %al, %al jne 0x2e2ce6 vmovss 0xc(%rsp), %xmm2 vmovss 0x10(%rsp), %xmm0 incl %ebx jmp 0x2e2bf0 vmovsd 0x60(%rsp), %xmm0 movq 0x58(%rsp), %rdi movq %rbp, %rsi callq 0x365d4 movl 0xc0(%r12), %eax imull 0x40(%rsp), %eax subl %ebx, %eax movslq %eax, %rdx movq %r14, %rdi movq %rbp, %rsi callq 0x2e34e2 vmovss 0xc(%rsp), %xmm2 vmovss 0x10(%rsp), %xmm0 jmp 0x2e2cf2 vmovss 0xc(%rsp), %xmm2 vmovss 0x10(%rsp), %xmm0 movl 0x78(%rsp), %ebx decl 0x40(%rsp) incl %ebx jmp 0x2e2b5c jmp 0x2e2d38 jmp 0x2e2d11 jmp 0x2e2d38 movq %rax, %r14 jmp 0x2e2d48 movq %rax, %r14 jmp 0x2e2d55 movq %rax, %r14 jmp 0x2e2d62 movq %rax, %r14 jmp 0x2e2d6f movq %rax, %r14 jmp 0x2e2d79 jmp 0x2e2d38 jmp 0x2e2d38 jmp 0x2e2d38 leaq 0x168(%rsp), %rdi movq %rax, %r14 callq 0x357fe jmp 0x2e2d48 leaq 0x160(%rsp), %rdi movq %rax, %r14 callq 0x435d0 leaq 0x98(%rsp), %rdi callq 0x393ac leaq 0xd0(%rsp), %rdi callq 0x2ddeea leaq 0x90(%rsp), %rdi callq 0x35c1a leaq 0x38(%rsp), %rdi callq 0x357fe movq 0x20(%rsp), %rdi callq 0xea60 movq %r14, %rdi callq 0xb3d0 nop
/quesnel[P]baryonyx/lib/src/itm-solver-common.hpp
baryonyx::itm::solver_functor<baryonyx::itm::solver_inequalities_101coeff<float, baryonyx::itm::minimize_tag, baryonyx::itm::default_cost_type<float>, false>, float, baryonyx::itm::minimize_tag, baryonyx::itm::default_cost_type<float>, baryonyx::none_observer>::operator()(std::vector<baryonyx::itm::merged_constraint, std::allocator<baryonyx::itm::merged_constraint>> const&, int, baryonyx::itm::default_cost_type<float> const&, double)
result operator()(const std::vector<merged_constraint>& constraints, int variables, const Cost& original_costs, double cost_constant) { result r; bit_array x(variables); int best_remaining = INT_MAX; auto& p = m_ctx.parameters; auto norm_costs = normalize_costs<Float, Cost>( m_ctx, original_costs, m_rng, variables); const auto kappa_step = static_cast<Float>(p.kappa_step); const auto kappa_max = static_cast<Float>(p.kappa_max); const auto alpha = static_cast<Float>(p.alpha); const auto theta = static_cast<Float>(p.theta); const auto delta = p.delta < 0 ? compute_delta<Float, Cost>(m_ctx, norm_costs, theta, variables) : static_cast<Float>(p.delta); const auto pushing_k_factor = static_cast<Float>(p.pushing_k_factor); const auto pushing_objective_amplifier = static_cast<Float>(p.pushing_objective_amplifier); const long int w_limit = static_cast<long int>(p.w); Solver slv( m_rng, length(constraints), variables, norm_costs, constraints); compute_order compute(p.order, variables); { std::bernoulli_distribution choose_mutation( m_ctx.parameters.init_policy_random); bit_array empty_x; switch (p.init_policy) { case solver_parameters::init_policy_type::pessimistic_solve: init_with_pre_solve<Cost, Mode>( x, empty_x, m_rng, original_costs, constraints, 1.0); break; case solver_parameters::init_policy_type::optimistic_solve: init_with_pre_solve<Cost, Mode>( empty_x, x, m_rng, original_costs, constraints, 1.0); break; case solver_parameters::init_policy_type::bastert: init_with_bastert<Cost, Mode>(x, original_costs, variables, 0); break; } for (int i = 0, e = x.size(); i != e; ++i) if (choose_mutation(m_rng)) x.invert(i); } bool start_push = false; auto kappa = static_cast<Float>(p.kappa_min); Observer obs("img", slv.m, slv.n, p.limit); m_begin = std::chrono::steady_clock::now(); m_end = std::chrono::steady_clock::now(); compute.init(slv, x); for (long int i = 0; i != p.limit; ++i) { auto remaining = compute.run(slv, x, m_rng, kappa, delta, theta); obs.make_observation(slv.ap, slv.P.get(), slv.pi.get()); if (remaining == 0) { store_if_better( x, original_costs.results(x, cost_constant), i); best_remaining = remaining; start_push = true; break; } if (remaining < best_remaining) { store_if_better(x, remaining, i); best_remaining = remaining; } if (i > w_limit) kappa += kappa_step * std::pow(static_cast<Float>(remaining) / static_cast<Float>(slv.m), alpha); if (kappa > kappa_max) { r.status = result_status::kappa_max_reached; break; } if (is_timelimit_reached()) { r.status = result_status::time_limit_reached; break; } } if (!start_push) { r.status = result_status::limit_reached; } else { for (int push = 0; push < p.pushes_limit; ++push) { auto remaining = compute.push_and_run(slv, x, m_rng, pushing_k_factor * kappa, delta, theta, pushing_objective_amplifier); if (remaining == 0) store_if_better(x, original_costs.results(x, cost_constant), -push * p.pushing_iteration_limit - 1); if (is_timelimit_reached()) break; for (int iter = 0; iter < p.pushing_iteration_limit; ++iter) { remaining = compute.run(slv, x, m_rng, kappa, delta, theta); if (remaining == 0) { store_if_better( x, original_costs.results(x, cost_constant), -push * p.pushing_iteration_limit - iter - 1); break; } if (iter > p.w) kappa += kappa_step * std::pow(static_cast<Float>(remaining) / static_cast<Float>(slv.m), alpha); if (kappa > kappa_max) break; if (is_timelimit_reached()) break; } } } if (m_best.remaining_constraints == 0) r.status = result_status::success; if (!m_best.x.empty()) { r.solutions.resize(1); convert(m_best, r.solutions[0], variables); } return r; }
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x168, %rsp # imm = 0x168 vmovsd %xmm0, 0x58(%rsp) vxorps %xmm0, %xmm0, %xmm0 movq %r8, %r13 movl %ecx, %ebx movq %rdx, %r15 movq %rsi, %r14 movq %rdi, 0x18(%rsp) vmovups %zmm0, 0x58(%rdi) vmovups %zmm0, 0x40(%rdi) vmovups %zmm0, (%rdi) movq $0x7fffffff, 0x98(%rdi) # imm = 0x7FFFFFFF movl $0x2, 0xa0(%rdi) leaq 0x28(%rsp), %rdi movl %ecx, %esi vzeroupper callq 0x368ae movq 0x10(%r14), %r12 movq 0x18(%r14), %rcx leaq 0x90(%rsp), %rdi movq %r12, %rsi movq %r13, %rdx movl %ebx, %r8d callq 0x38959 vmovsd 0x20(%r12), %xmm0 vmovsd 0x10(%r12), %xmm2 vxorpd %xmm1, %xmm1, %xmm1 vmovsd %xmm0, 0x48(%rsp) vmovsd 0x28(%r12), %xmm0 vucomisd %xmm2, %xmm1 vmovsd %xmm0, 0x40(%rsp) vmovsd 0x30(%r12), %xmm0 vmovsd %xmm0, 0x38(%rsp) vcvtsd2ss 0x8(%r12), %xmm3, %xmm0 vmovss %xmm0, 0x4(%rsp) jbe 0x2e2e70 movq 0x10(%r14), %rdi leaq 0x90(%rsp), %rsi movl %ebx, %edx callq 0x38a37 vmovss %xmm0, 0x14(%rsp) jmp 0x2e2e7a vcvtsd2ss %xmm2, %xmm2, %xmm0 vmovss %xmm0, 0x14(%rsp) vcvttsd2si 0xa0(%r12), %rax vmovsd 0x38(%r12), %xmm0 vmovsd 0x40(%r12), %xmm1 movq 0x18(%r14), %rsi movq %rax, 0x70(%rsp) movq 0x8(%r15), %rax vmovsd %xmm0, 0x68(%rsp) vmovsd %xmm1, 0x60(%rsp) subq (%r15), %rax pushq $0x28 popq %rcx cqto idivq %rcx leaq 0xd8(%rsp), %rdi leaq 0x90(%rsp), %r8 movl %eax, %edx movl %ebx, %ecx movq %r15, %r9 callq 0x2dd7b2 movl 0xcc(%r12), %esi leaq 0xa0(%rsp), %rdi movl %ebx, %edx callq 0x38cd8 movq 0x10(%r14), %rax movl %ebx, 0x7c(%rsp) vmovsd 0x48(%rax), %xmm0 vmovsd %xmm0, 0x8(%rsp) vxorps %xmm0, %xmm0, %xmm0 vmovaps %xmm0, 0x80(%rsp) movl 0xd4(%r12), %eax testl %eax, %eax je 0x2e2f58 cmpl $0x1, %eax je 0x2e2f32 cmpl $0x2, %eax jne 0x2e2f69 movq 0x18(%r14), %rdx leaq 0x80(%rsp), %rdi leaq 0x28(%rsp), %rsi jmp 0x2e2f43 movq 0x18(%r14), %rdx leaq 0x28(%rsp), %rdi leaq 0x80(%rsp), %rsi vmovsd 0x1710ed(%rip), %xmm0 # 0x454038 movq %r13, %rcx movq %r15, %r8 callq 0x4f9c9 jmp 0x2e2f69 leaq 0x28(%rsp), %rdi movq %r13, %rsi movl %ebx, %edx xorl %ecx, %ecx callq 0x4f95d vcvtsd2ss 0x48(%rsp), %xmm3, %xmm0 leaq 0x28(%rsp), %r15 movq %r13, 0x50(%rsp) xorl %ebx, %ebx movl (%r15), %ebp vmovss %xmm0, 0x24(%rsp) vcvtsd2ss 0x40(%rsp), %xmm3, %xmm0 vmovss %xmm0, 0x48(%rsp) vcvtsd2ss 0x38(%rsp), %xmm3, %xmm0 vmovss %xmm0, 0x40(%rsp) vcvtsd2ss 0x68(%rsp), %xmm3, %xmm0 vmovss %xmm0, 0x68(%rsp) vcvtsd2ss 0x60(%rsp), %xmm3, %xmm0 vmovss %xmm0, 0x60(%rsp) cmpl %ebx, %ebp je 0x2e2fdb movq 0x18(%r14), %rdi callq 0x36ac3 vmovsd 0x8(%rsp), %xmm1 vucomisd %xmm0, %xmm1 jbe 0x2e2fd7 movq %r15, %rdi movl %ebx, %esi callq 0x3626e incl %ebx jmp 0x2e2fb4 leaq 0x88(%rsp), %rdi callq 0x357fe vmovsd 0x18(%r12), %xmm0 vmovsd %xmm0, 0x8(%rsp) callq 0xb5a0 movq %rax, (%r14) callq 0xb5a0 movq %rax, 0x8(%r14) leaq 0xa0(%rsp), %rdi leaq 0xd8(%rsp), %rsi leaq 0x28(%rsp), %rdx callq 0x2dda48 vcvtsd2ss 0x8(%rsp), %xmm3, %xmm0 vmovss 0x4(%rsp), %xmm2 movl $0x7fffffff, %ebp # imm = 0x7FFFFFFF leaq 0x28(%rsp), %rbx xorl %r15d, %r15d cmpq 0xb0(%r12), %r15 je 0x2e30fe movq 0x18(%r14), %rcx vmovss 0x14(%rsp), %xmm1 leaq 0xa0(%rsp), %rdi leaq 0xd8(%rsp), %rsi movq %rbx, %rdx vmovss %xmm0, 0x8(%rsp) callq 0x2dda8a movl %eax, %r13d testl %eax, %eax je 0x2e3196 cmpl %ebp, %r13d jge 0x2e3093 movq %r14, %rdi movq %rbx, %rsi movl %r13d, %edx movq %r15, %rcx callq 0x2e3616 movl %r13d, %ebp cmpq 0x70(%rsp), %r15 jle 0x2e30ca vcvtsi2ss %r13d, %xmm3, %xmm0 vcvtsi2ssl 0x160(%rsp), %xmm3, %xmm1 vdivss %xmm1, %xmm0, %xmm0 vmovss 0x40(%rsp), %xmm1 callq 0xb530 vmovss 0x8(%rsp), %xmm1 vfmadd231ss 0x24(%rsp), %xmm0, %xmm1 # xmm1 = (xmm0 * mem) + xmm1 vmovaps %xmm1, %xmm0 jmp 0x2e30d0 vmovss 0x8(%rsp), %xmm0 vucomiss 0x48(%rsp), %xmm0 ja 0x2e30fe movq %r14, %rdi vmovss %xmm0, 0x8(%rsp) callq 0x2e364e vmovss 0x8(%rsp), %xmm0 vmovss 0x4(%rsp), %xmm2 testb %al, %al jne 0x2e30fe incq %r15 jmp 0x2e3039 movq 0x18(%rsp), %rax movl $0x5, 0xa0(%rax) cmpl $0x0, 0x50(%r14) jne 0x2e3120 movq 0x18(%rsp), %rax andl $0x0, 0xa0(%rax) cmpq $0x0, 0x28(%r14) je 0x2e314e movq 0x18(%rsp), %rax leaq 0x68(%rax), %rbx pushq $0x1 popq %rsi movq %rbx, %rdi callq 0x35292 movq (%rbx), %rsi addq $0x20, %r14 movl 0x7c(%rsp), %edx movq %r14, %rdi callq 0x4f485 leaq 0xa0(%rsp), %rdi callq 0x393ac leaq 0xd8(%rsp), %rdi callq 0x2ddeea leaq 0x98(%rsp), %rdi callq 0x35c1a leaq 0x30(%rsp), %rdi callq 0x357fe movq 0x18(%rsp), %rax addq $0x168, %rsp # imm = 0x168 popq %rbx popq %r12 popq %r13 popq %r14 popq %r15 popq %rbp retq vmovsd 0x58(%rsp), %xmm0 movq 0x50(%rsp), %rdi leaq 0x28(%rsp), %rbx movq %rbx, %rsi callq 0x365d4 movq %r14, %rdi movq %rbx, %rsi movq %r15, %rdx callq 0x2e35c8 vmovss 0x4(%rsp), %xmm2 vmovss 0x8(%rsp), %xmm0 leaq 0xd8(%rsp), %r15 leaq 0x28(%rsp), %rbp xorl %ebx, %ebx movl $0x0, 0x38(%rsp) cmpl 0xbc(%r12), %ebx jge 0x2e310d vmovss %xmm0, 0x8(%rsp) vmulss 0x68(%rsp), %xmm0, %xmm0 movq 0x18(%r14), %rcx vmovss 0x14(%rsp), %xmm1 vmovss 0x60(%rsp), %xmm3 leaq 0xa0(%rsp), %rdi movq %r15, %rsi movq %rbp, %rdx callq 0x2ddc96 testl %eax, %eax jne 0x2e324e vmovsd 0x58(%rsp), %xmm0 movq 0x50(%rsp), %rdi movq %rbp, %rsi callq 0x365d4 movl 0xc0(%r12), %eax imull %ebx, %eax notl %eax movslq %eax, %rdx movq %r14, %rdi movq %rbp, %rsi callq 0x2e35c8 movq %r14, %rdi callq 0x2e364e vmovss 0x4(%rsp), %xmm2 vmovss 0x8(%rsp), %xmm0 testb %al, %al jne 0x2e310d movl %ebx, 0x70(%rsp) pushq $0x1 popq %rax movl %eax, %ebx leal -0x1(%rbx), %r13d cmpl 0xc0(%r12), %r13d jge 0x2e3375 movq 0x18(%r14), %rcx vmovss 0x14(%rsp), %xmm1 leaq 0xa0(%rsp), %rdi movq %r15, %rsi movq %rbp, %rdx vmovss %xmm0, 0x8(%rsp) callq 0x2dda8a testl %eax, %eax je 0x2e332b vcvtsi2sd %r13d, %xmm4, %xmm0 vucomisd 0xa0(%r12), %xmm0 jbe 0x2e32f2 vcvtsi2ss %eax, %xmm4, %xmm0 vcvtsi2ssl 0x160(%rsp), %xmm4, %xmm1 vdivss %xmm1, %xmm0, %xmm0 vmovss 0x40(%rsp), %xmm1 callq 0xb530 vmovss 0x8(%rsp), %xmm1 vmovss 0x4(%rsp), %xmm2 vfmadd231ss 0x24(%rsp), %xmm0, %xmm1 # xmm1 = (xmm0 * mem) + xmm1 vmovaps %xmm1, %xmm0 jmp 0x2e32fe vmovss 0x4(%rsp), %xmm2 vmovss 0x8(%rsp), %xmm0 vucomiss 0x48(%rsp), %xmm0 ja 0x2e3375 movq %r14, %rdi vmovss %xmm0, 0x8(%rsp) callq 0x2e364e testb %al, %al jne 0x2e3369 vmovss 0x4(%rsp), %xmm2 vmovss 0x8(%rsp), %xmm0 incl %ebx jmp 0x2e3273 vmovsd 0x58(%rsp), %xmm0 movq 0x50(%rsp), %rdi movq %rbp, %rsi callq 0x365d4 movl 0xc0(%r12), %eax imull 0x38(%rsp), %eax subl %ebx, %eax movslq %eax, %rdx movq %r14, %rdi movq %rbp, %rsi callq 0x2e35c8 vmovss 0x4(%rsp), %xmm2 vmovss 0x8(%rsp), %xmm0 jmp 0x2e3375 vmovss 0x4(%rsp), %xmm2 vmovss 0x8(%rsp), %xmm0 movl 0x70(%rsp), %ebx decl 0x38(%rsp) incl %ebx jmp 0x2e31df jmp 0x2e33b6 jmp 0x2e338f jmp 0x2e33b6 movq %rax, %r14 jmp 0x2e33c6 movq %rax, %r14 jmp 0x2e33d3 movq %rax, %r14 jmp 0x2e33e0 movq %rax, %r14 jmp 0x2e33ea jmp 0x2e33b6 jmp 0x2e33b6 jmp 0x2e33b6 leaq 0x88(%rsp), %rdi movq %rax, %r14 callq 0x357fe jmp 0x2e33b9 movq %rax, %r14 leaq 0xa0(%rsp), %rdi callq 0x393ac leaq 0xd8(%rsp), %rdi callq 0x2ddeea leaq 0x98(%rsp), %rdi callq 0x35c1a leaq 0x30(%rsp), %rdi callq 0x357fe movq 0x18(%rsp), %rdi callq 0xea60 movq %r14, %rdi callq 0xb3d0
/quesnel[P]baryonyx/lib/src/itm-solver-common.hpp
baryonyx::itm::solver_functor<baryonyx::itm::solver_inequalities_101coeff<float, baryonyx::itm::minimize_tag, baryonyx::itm::default_cost_type<float>, false>, float, baryonyx::itm::minimize_tag, baryonyx::itm::default_cost_type<float>, baryonyx::pnm_observer>::store_if_better(baryonyx::bit_array const&, double, long)
void store_if_better(const bit_array& x, double current, long int i) { if (is_better_solution<Mode>(current, m_best.value)) { m_best.x = x; m_best.duration = duration(); m_best.loop = i; m_best.remaining_constraints = 0; m_best.value = current; } }
vmovsd 0x30(%rdi), %xmm1 vucomisd %xmm0, %xmm1 jbe 0x2e3448 pushq %r14 pushq %rbx pushq %rax movq %rdi, %r14 addq $0x20, %rdi movq %rdx, %rbx vmovsd %xmm0, (%rsp) callq 0x36a44 movq %r14, %rdi callq 0x2e34c2 vmovsd (%rsp), %xmm1 vmovsd %xmm0, 0x38(%r14) movq %rbx, 0x48(%r14) andl $0x0, 0x50(%r14) vmovsd %xmm1, 0x30(%r14) addq $0x8, %rsp popq %rbx popq %r14 retq nop
/quesnel[P]baryonyx/lib/src/itm-solver-common.hpp
baryonyx::itm::solver_functor<baryonyx::itm::solver_inequalities_101coeff<float, baryonyx::itm::minimize_tag, baryonyx::itm::default_cost_type<float>, false>, float, baryonyx::itm::minimize_tag, baryonyx::itm::default_cost_type<float>, baryonyx::file_observer>::store_if_better(baryonyx::bit_array const&, double, long)
void store_if_better(const bit_array& x, double current, long int i) { if (is_better_solution<Mode>(current, m_best.value)) { m_best.x = x; m_best.duration = duration(); m_best.loop = i; m_best.remaining_constraints = 0; m_best.value = current; } }
vmovsd 0x30(%rdi), %xmm1 vucomisd %xmm0, %xmm1 jbe 0x2e352e pushq %r14 pushq %rbx pushq %rax movq %rdi, %r14 addq $0x20, %rdi movq %rdx, %rbx vmovsd %xmm0, (%rsp) callq 0x36a44 movq %r14, %rdi callq 0x2e35a8 vmovsd (%rsp), %xmm1 vmovsd %xmm0, 0x38(%r14) movq %rbx, 0x48(%r14) andl $0x0, 0x50(%r14) vmovsd %xmm1, 0x30(%r14) addq $0x8, %rsp popq %rbx popq %r14 retq nop
/quesnel[P]baryonyx/lib/src/itm-solver-common.hpp
baryonyx::itm::solver_functor<baryonyx::itm::solver_inequalities_101coeff<float, baryonyx::itm::minimize_tag, baryonyx::itm::default_cost_type<float>, false>, float, baryonyx::itm::minimize_tag, baryonyx::itm::default_cost_type<float>, baryonyx::file_observer>::is_timelimit_reached()
bool is_timelimit_reached() { m_end = std::chrono::steady_clock::now(); return is_time_limit(m_ctx.parameters.time_limit, m_begin, m_end); }
pushq %rbx movq %rdi, %rbx callq 0xb5a0 movq %rax, 0x8(%rbx) vxorpd %xmm2, %xmm2, %xmm2 movq 0x10(%rbx), %rcx subq (%rbx), %rax vmovsd (%rcx), %xmm0 vcvtsi2sd %rax, %xmm1, %xmm1 vdivsd 0x1764ef(%rip), %xmm1, %xmm1 # 0x459a80 vcmpnlepd %xmm2, %xmm0, %k1 vcmpltpd %xmm1, %xmm0, %k0 {%k1} kmovd %k0, %eax andb $0x1, %al popq %rbx retq nop
/quesnel[P]baryonyx/lib/src/itm-solver-common.hpp
baryonyx::itm::solver_functor<baryonyx::itm::solver_inequalities_101coeff<float, baryonyx::itm::minimize_tag, baryonyx::itm::default_cost_type<float>, false>, float, baryonyx::itm::minimize_tag, baryonyx::itm::default_cost_type<float>, baryonyx::none_observer>::store_if_better(baryonyx::bit_array const&, double, long)
void store_if_better(const bit_array& x, double current, long int i) { if (is_better_solution<Mode>(current, m_best.value)) { m_best.x = x; m_best.duration = duration(); m_best.loop = i; m_best.remaining_constraints = 0; m_best.value = current; } }
vmovsd 0x30(%rdi), %xmm1 vucomisd %xmm0, %xmm1 jbe 0x2e3614 pushq %r14 pushq %rbx pushq %rax movq %rdi, %r14 addq $0x20, %rdi movq %rdx, %rbx vmovsd %xmm0, (%rsp) callq 0x36a44 movq %r14, %rdi callq 0x2e368e vmovsd (%rsp), %xmm1 vmovsd %xmm0, 0x38(%r14) movq %rbx, 0x48(%r14) andl $0x0, 0x50(%r14) vmovsd %xmm1, 0x30(%r14) addq $0x8, %rsp popq %rbx popq %r14 retq nop
/quesnel[P]baryonyx/lib/src/itm-solver-common.hpp
baryonyx::itm::solver_functor<baryonyx::itm::solver_inequalities_101coeff<float, baryonyx::itm::minimize_tag, baryonyx::itm::default_cost_type<float>, false>, float, baryonyx::itm::minimize_tag, baryonyx::itm::default_cost_type<float>, baryonyx::none_observer>::is_timelimit_reached()
bool is_timelimit_reached() { m_end = std::chrono::steady_clock::now(); return is_time_limit(m_ctx.parameters.time_limit, m_begin, m_end); }
pushq %rbx movq %rdi, %rbx callq 0xb5a0 movq %rax, 0x8(%rbx) vxorpd %xmm2, %xmm2, %xmm2 movq 0x10(%rbx), %rcx subq (%rbx), %rax vmovsd (%rcx), %xmm0 vcvtsi2sd %rax, %xmm1, %xmm1 vdivsd 0x176409(%rip), %xmm1, %xmm1 # 0x459a80 vcmpnlepd %xmm2, %xmm0, %k1 vcmpltpd %xmm1, %xmm0, %k0 {%k1} kmovd %k0, %eax andb $0x1, %al popq %rbx retq nop
/quesnel[P]baryonyx/lib/src/itm-solver-common.hpp
baryonyx::result baryonyx::itm::optimize_problem<baryonyx::itm::solver_inequalities_101coeff<float, baryonyx::itm::minimize_tag, baryonyx::itm::quadratic_cost_type<float>, true>, float, baryonyx::itm::minimize_tag, baryonyx::itm::quadratic_cost_type<float>>(baryonyx::context const&, baryonyx::problem const&)
inline result optimize_problem(const context& ctx, const problem& pb) { result r; if (ctx.start) ctx.start(ctx.parameters); auto constraints{ make_merged_constraints(ctx, pb) }; if (constraints.empty() || pb.vars.values.empty()) { r.status = result_status::success; r.solutions.resize(1); r.solutions.back().value = pb.objective.value; r.strings = pb.strings; r.affected_vars = pb.affected_vars; r.variable_name = pb.vars.names; return r; } random_engine rng(init_random_generator_seed(ctx)); auto variables = length(pb.vars.values); auto cost = Cost(pb.objective, variables); auto cost_constant = pb.objective.value; const auto thread = get_thread_number(ctx); std::vector<optimize_functor<Solver, Float, Mode, Cost>> functors; std::vector<std::thread> pool(thread); best_solution_recorder<Cost, Float, Mode> best_recorder( rng, thread, cost, cost_constant, variables, constraints, ctx.parameters.init_population_size); auto seeds = generate_seed(rng, thread); std::atomic_bool stop_task; stop_task.store(false); for (unsigned i = 0u; i != thread; ++i) functors.emplace_back(ctx, i, seeds[i]); for (unsigned i = 0u; i != thread; ++i) pool[i] = std::thread(std::ref(functors[i]), std::ref(stop_task), std::ref(best_recorder), std::cref(constraints), variables, std::cref(cost), cost_constant); const auto start = std::chrono::steady_clock::now(); auto end = start; do { std::this_thread::sleep_for(std::chrono::seconds{ 1L }); if (ctx.update) { auto call_number = 0L; for (auto i = 0u; i != thread; ++i) call_number += functors[i].m_call_number; int constraints_remaining; long int loop; double value; double duration; best_recorder.get_best( constraints_remaining, value, duration, loop); ctx.update( constraints_remaining, value, loop, duration, call_number); } end = std::chrono::steady_clock::now(); } while (!is_time_limit(ctx.parameters.time_limit, start, end)); stop_task.store(true); for (auto& t : pool) t.join(); r.strings = pb.strings; r.affected_vars = pb.affected_vars; r.variable_name = pb.vars.names; r.variables = variables; r.constraints = length(constraints); const auto& first = best_recorder.get_best(0); if (!first.is_solution()) r.status = result_status::time_limit_reached; else r.status = result_status::success; r.duration = first.duration; r.loop = first.loop; r.remaining_constraints = first.remaining_constraints; switch (ctx.parameters.storage) { case solver_parameters::storage_type::one: { r.solutions.resize(1); convert(first, r.solutions[0], variables); } break; case solver_parameters::storage_type::bound: { r.solutions.resize(2); convert(first, r.solutions[0], variables); convert(best_recorder.get_worst(), r.solutions[1], variables); } break; case solver_parameters::storage_type::five: { r.solutions.resize(5); for (int i = 0; i != 5; ++i) convert(best_recorder.get_best(i), r.solutions[i], variables); } break; } best_recorder.show_population(ctx); if (ctx.finish) ctx.finish(r); return r; }
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x198, %rsp # imm = 0x198 vxorpd %xmm0, %xmm0, %xmm0 movq %rdx, %r13 movq %rsi, %r15 movq %rdi, %r12 vmovupd %zmm0, 0x58(%rdi) vmovupd %zmm0, 0x40(%rdi) vmovupd %zmm0, (%rdi) movq $0x7fffffff, 0x98(%rdi) # imm = 0x7FFFFFFF movl $0x2, 0xa0(%rdi) cmpq $0x0, 0x128(%rsi) je 0x2e3713 leaq 0x118(%r15), %rdi movq %r15, %rsi vzeroupper callq 0x3527e leaq 0x48(%rsp), %rdi movq %r15, %rsi movq %r13, %rdx vzeroupper callq 0x450078 movq 0x48(%rsp), %rax cmpq 0x50(%rsp), %rax je 0x2e3872 movq 0xa8(%r13), %rax cmpq 0xb0(%r13), %rax je 0x2e3872 movq %r15, %rdi callq 0x353ba movl $0x7fffffff, %ecx # imm = 0x7FFFFFFF xorl %edx, %edx divq %rcx cmpq $0x1, %rdx adcq $0x0, %rdx movq %rdx, 0x40(%rsp) movq 0xb0(%r13), %rax subq 0xa8(%r13), %rax pushq $0xc popq %rcx cqto leaq 0x10(%r13), %rsi idivq %rcx movl %eax, 0x4(%rsp) leaq 0xb8(%rsp), %rdi movl %eax, %edx movq %r12, 0x10(%rsp) callq 0x445b4 vmovsd 0x40(%r13), %xmm0 vmovsd %xmm0, 0x38(%rsp) movq %r15, %rdi callq 0x35447 andq $0x0, 0x70(%rsp) vxorpd %xmm0, %xmm0, %xmm0 movl %eax, %ebp vmovapd %xmm0, 0x60(%rsp) leaq 0xa0(%rsp), %rdi leaq 0xd8(%rsp), %rdx movq %rbp, %rsi callq 0x35462 vmovsd 0x38(%rsp), %xmm0 movl 0xc4(%r15), %eax movl 0x4(%rsp), %r8d subq $0x8, %rsp movl %ebp, %edx leaq 0xe0(%rsp), %rdi leaq 0x48(%rsp), %rsi leaq 0xc0(%rsp), %rcx leaq 0x50(%rsp), %r9 pushq %rax callq 0x58da6 popq %rax popq %rcx leaq 0x30(%rsp), %rdi leaq 0x40(%rsp), %rsi movl %ebp, %edx callq 0x3551c xorl %eax, %eax movb %al, 0x3(%rsp) xorl %ecx, %ecx leaq 0x60(%rsp), %rbx leaq 0x8(%rsp), %r14 xchgb %cl, 0x3(%rsp) leaq 0x90(%r13), %rcx movq %rcx, 0x88(%rsp) movl %eax, 0x8(%rsp) cmpl %ebp, %eax je 0x2e38cd movl %eax, %ecx shlq $0x3, %rcx addq 0x30(%rsp), %rcx movq %rbx, %rdi movq %r15, %rsi movq %r14, %rdx callq 0x2e49b0 movl 0x8(%rsp), %eax incl %eax jmp 0x2e3849 andl $0x0, 0xa0(%r12) leaq 0x68(%r12), %rdi pushq $0x1 popq %rsi callq 0x35292 vmovsd 0x40(%r13), %xmm0 movq 0x70(%r12), %rax vmovsd %xmm0, -0x8(%rax) movq %r12, %rdi movq %r13, %rsi callq 0x31d28 leaq 0xc0(%r13), %rsi leaq 0x20(%r12), %rdi callq 0x352c2 addq $0x90, %r13 leaq 0x50(%r12), %rdi movq %r13, %rsi callq 0x352ec jmp 0x2e3c68 movq %r13, 0x80(%rsp) movq %r15, %r13 leaq 0x8(%rsp), %r15 leaq 0x98(%rsp), %r14 leaq 0x4(%rsp), %rbx xorl %r12d, %r12d cmpq %rbp, %r12 je 0x2e3991 imulq $0xc0, %r12, %rax addq 0x60(%rsp), %rax leaq 0x3(%rsp), %r10 leaq 0xd8(%rsp), %r11 movq %r10, 0x20(%rsp) leaq 0x48(%rsp), %r10 movq %r11, 0x18(%rsp) movq %r10, 0x98(%rsp) movq %rax, 0x28(%rsp) leaq 0xb8(%rsp), %rax movq %rax, 0x90(%rsp) leaq 0x20(%rsp), %rdx leaq 0x38(%rsp), %rax movq %r15, %rdi leaq 0x28(%rsp), %rsi leaq 0x18(%rsp), %rcx movq %r14, %r8 movq %rbx, %r9 pushq %rax leaq 0x98(%rsp), %rax pushq %rax callq 0x2e49fc popq %rax popq %rcx leaq (,%r12,8), %rdi addq 0xa0(%rsp), %rdi movq %r15, %rsi callq 0x356d6 movq %r15, %rdi callq 0x356f0 incq %r12 jmp 0x2e38ed callq 0xb5a0 imulq $0xc0, %rbp, %r15 leaq 0x138(%r13), %r12 leaq 0x8(%rsp), %rbp movq %rax, %rbx movq $0x1, 0x8(%rsp) movq %rbp, %rdi callq 0x356fd cmpq $0x0, 0x148(%r13) je 0x2e3a26 movq 0x60(%rsp), %rax xorl %ecx, %ecx xorl %r14d, %r14d cmpq %rcx, %r15 je 0x2e39e7 addq 0xb0(%rax,%rcx), %r14 addq $0xc0, %rcx jmp 0x2e39d1 leaq 0xe0(%rsp), %rdi leaq 0x18(%rsp), %rsi leaq 0x28(%rsp), %rdx leaq 0x20(%rsp), %rcx movq %rbp, %r8 callq 0x60418 vmovsd 0x28(%rsp), %xmm0 vmovsd 0x20(%rsp), %xmm1 movq 0x8(%rsp), %rdx movl 0x18(%rsp), %esi movq %r12, %rdi movq %r14, %rcx callq 0x35736 callq 0xb5a0 vmovsd (%r13), %xmm0 subq %rbx, %rax vxorpd %xmm1, %xmm1, %xmm1 vucomisd %xmm0, %xmm1 vcvtsi2sd %rax, %xmm2, %xmm1 vdivsd 0x176037(%rip), %xmm1, %xmm1 # 0x459a80 jae 0x2e39ac vucomisd %xmm0, %xmm1 jbe 0x2e39ac movb $0x1, %al xchgb %al, 0x3(%rsp) movq 0xa0(%rsp), %rbx movq 0xa8(%rsp), %r14 movq 0x10(%rsp), %r12 cmpq %r14, %rbx je 0x2e3a87 movq %rbx, %rdi callq 0xb1a0 addq $0x8, %rbx jmp 0x2e3a74 movq 0x80(%rsp), %rbx movq %r12, %rdi movq %rbx, %rsi callq 0x31d28 addq $0xc0, %rbx leaq 0x20(%r12), %rdi movq %rbx, %rsi callq 0x352c2 leaq 0x50(%r12), %rdi movq 0x88(%rsp), %rsi callq 0x352ec movl 0x4(%rsp), %eax movl %eax, 0x90(%r12) movq 0x50(%rsp), %rax subq 0x48(%rsp), %rax pushq $0x28 popq %rcx cqto xorl %esi, %esi idivq %rcx movq 0x118(%rsp), %rdi movq 0x130(%rsp), %rcx movl %eax, 0x94(%r12) movslq (%rdi), %rax imulq $0x38, %rax, %rax movl 0x30(%rcx,%rax), %edx leaq (%rcx,%rax), %rbx testl %edx, %edx setne %sil leal (%rsi,%rsi,2), %esi movl %esi, 0xa0(%r12) vmovsd 0x18(%rcx,%rax), %xmm0 vmovsd %xmm0, 0x80(%r12) movq 0x28(%rcx,%rax), %rax movq %rax, 0x88(%r12) movl %edx, 0x98(%r12) movl 0xe8(%r13), %eax testl %eax, %eax je 0x2e3be2 cmpl $0x1, %eax je 0x2e3baa cmpl $0x2, %eax jne 0x2e3c01 leaq 0x68(%r12), %rbx pushq $0x5 popq %rsi movq %rbx, %rdi callq 0x35292 xorl %r14d, %r14d xorl %r15d, %r15d cmpq $0x5, %r15 je 0x2e3c01 movq 0x118(%rsp), %rax movq (%rbx), %rsi movl 0x4(%rsp), %edx movslq (%rax,%r15,4), %rax addq %r14, %rsi imulq $0x38, %rax, %rdi addq 0x130(%rsp), %rdi callq 0x4f485 incq %r15 addq $0x20, %r14 jmp 0x2e3b70 leaq 0x68(%r12), %r14 pushq $0x2 popq %rsi movq %r14, %rdi callq 0x35292 movl 0x4(%rsp), %edx movq (%r14), %rsi movq %rbx, %rdi callq 0x4f485 leaq 0xe0(%rsp), %rdi callq 0x60496 movq (%r14), %rsi movq %rax, %rbx addq $0x20, %rsi jmp 0x2e3bf5 leaq 0x68(%r12), %r14 pushq $0x1 popq %rsi movq %r14, %rdi callq 0x35292 movq (%r14), %rsi movl 0x4(%rsp), %edx movq %rbx, %rdi callq 0x4f485 leaq 0xd8(%rsp), %rdi movq %r13, %rsi callq 0x58f0a cmpq $0x0, 0x168(%r13) je 0x2e3c2d addq $0x158, %r13 # imm = 0x158 movq %r13, %rdi movq %r12, %rsi callq 0x357ea leaq 0x30(%rsp), %rdi callq 0x357fe leaq 0xd8(%rsp), %rdi callq 0x58f1e leaq 0xa0(%rsp), %rdi callq 0x35836 leaq 0x60(%rsp), %rdi callq 0x2ea23c leaq 0xb8(%rsp), %rdi callq 0x4495c leaq 0x48(%rsp), %rdi callq 0x35858 movq %r12, %rax addq $0x198, %rsp # imm = 0x198 popq %rbx popq %r12 popq %r13 popq %r14 popq %r15 popq %rbp retq movq %rax, %rbx jmp 0x2e3cc9 movq %rax, %rbx jmp 0x2e3cd6 movq %rax, %rbx jmp 0x2e3ce3 jmp 0x2e3cad jmp 0x2e3c9e jmp 0x2e3cbc jmp 0x2e3cbc movq %r12, 0x10(%rsp) movq %rax, %rbx jmp 0x2e3d04 movq %r12, 0x10(%rsp) movq %rax, %rbx jmp 0x2e3cfa jmp 0x2e3cbc jmp 0x2e3cbc jmp 0x2e3cbc jmp 0x2e3cbc jmp 0x2e3cbc movq %rax, %rbx leaq 0x30(%rsp), %rdi callq 0x357fe leaq 0xd8(%rsp), %rdi callq 0x58f1e leaq 0xa0(%rsp), %rdi callq 0x35836 leaq 0x60(%rsp), %rdi callq 0x2ea23c leaq 0xb8(%rsp), %rdi callq 0x4495c leaq 0x48(%rsp), %rdi callq 0x35858 movq 0x10(%rsp), %rdi callq 0xea60 movq %rbx, %rdi callq 0xb3d0
/quesnel[P]baryonyx/lib/src/itm-optimizer-common.hpp
baryonyx::result baryonyx::itm::solve_problem<baryonyx::itm::solver_inequalities_101coeff<float, baryonyx::itm::minimize_tag, baryonyx::itm::quadratic_cost_type<float>, true>, float, baryonyx::itm::minimize_tag, baryonyx::itm::quadratic_cost_type<float>>(baryonyx::context const&, baryonyx::problem const&)
inline result solve_problem(const context& ctx, const problem& pb) { if (ctx.start) ctx.start(ctx.parameters); result ret; auto variables = length(pb.vars.values); auto constraints{ make_merged_constraints(ctx, pb) }; if (!constraints.empty() && !pb.vars.values.empty()) { random_engine rng(init_random_generator_seed(ctx)); auto cost = Cost(pb.objective, variables); auto cost_constant = pb.objective.value; switch (ctx.parameters.observer) { case solver_parameters::observer_type::pnm: { using obs = pnm_observer; solver_functor<Solver, Float, Mode, Cost, obs> slv(ctx, rng); ret = slv(constraints, variables, cost, cost_constant); } break; case solver_parameters::observer_type::file: { using obs = file_observer; solver_functor<Solver, Float, Mode, Cost, obs> slv(ctx, rng); ret = slv(constraints, variables, cost, cost_constant); } break; default: { using obs = none_observer; solver_functor<Solver, Float, Mode, Cost, obs> slv(ctx, rng); ret = slv(constraints, variables, cost, cost_constant); break; } } } else { ret.status = result_status::success; ret.solutions.resize(1); ret.solutions.back().value = pb.objective.value; } ret.strings = pb.strings; ret.variable_name = std::move(pb.vars.names); ret.affected_vars = std::move(pb.affected_vars); ret.variables = variables; ret.constraints = length(constraints); if (ctx.finish) ctx.finish(ret); return ret; }
pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x140, %rsp # imm = 0x140 cmpq $0x0, 0x128(%rsi) movq %rdx, %r15 movq %rsi, %r14 movq %rdi, %rbx je 0x2e3d48 leaq 0x118(%r14), %rdi movq %r14, %rsi callq 0x3527e vxorps %xmm0, %xmm0, %xmm0 vmovups %zmm0, 0x58(%rbx) vmovups %zmm0, 0x40(%rbx) vmovups %zmm0, (%rbx) movq $0x7fffffff, 0x98(%rbx) # imm = 0x7FFFFFFF movl $0x2, 0xa0(%rbx) movq 0xb0(%r15), %rax subq 0xa8(%r15), %rax pushq $0xc popq %rcx cqto idivq %rcx movq %rax, %r12 leaq 0x8(%rsp), %rdi movq %r14, %rsi movq %r15, %rdx vzeroupper callq 0x450078 movq 0x8(%rsp), %rax cmpq 0x10(%rsp), %rax je 0x2e3e71 movq 0xa8(%r15), %rax cmpq 0xb0(%r15), %rax je 0x2e3e71 movq %r14, %rdi callq 0x353ba movl $0x7fffffff, %ecx # imm = 0x7FFFFFFF xorl %edx, %edx leaq 0x10(%r15), %rsi divq %rcx cmpq $0x1, %rdx adcq $0x0, %rdx movq %rdx, (%rsp) leaq 0x20(%rsp), %rdi movl %r12d, %edx callq 0x445b4 vmovsd 0x40(%r15), %xmm0 movl 0xe4(%r14), %eax vxorps %xmm1, %xmm1, %xmm1 leaq 0x40(%rsp), %rsi cmpl $0x1, %eax je 0x2e3e98 cmpl $0x2, %eax jne 0x2e3ee7 movabsq $0x7ff0000000000000, %r9 # imm = 0x7FF0000000000000 vmovaps %xmm1, (%rsi) movq %r14, 0x10(%rsi) movq %rsp, %rax movq %rax, 0x18(%rsi) vmovaps %xmm1, 0x20(%rsi) movq %r9, 0x30(%rsi) andq $0x0, 0x48(%rsi) vmovups %xmm1, 0x38(%rsi) movl $0x7fffffff, 0x50(%rsi) # imm = 0x7FFFFFFF leaq 0x98(%rsp), %rdi leaq 0x8(%rsp), %rdx leaq 0x20(%rsp), %r8 movl %r12d, %ecx callq 0x2ea948 jmp 0x2e3f34 andl $0x0, 0xa0(%rbx) leaq 0x68(%rbx), %rdi pushq $0x1 popq %rsi callq 0x35292 vmovsd 0x40(%r15), %xmm0 movq 0x70(%rbx), %rax vmovsd %xmm0, -0x8(%rax) jmp 0x2e3f63 movabsq $0x7ff0000000000000, %r9 # imm = 0x7FF0000000000000 vmovaps %xmm1, (%rsi) movq %r14, 0x10(%rsi) movq %rsp, %rax movq %rax, 0x18(%rsi) vmovaps %xmm1, 0x20(%rsi) movq %r9, 0x30(%rsi) andq $0x0, 0x48(%rsi) vmovups %xmm1, 0x38(%rsi) movl $0x7fffffff, 0x50(%rsi) # imm = 0x7FFFFFFF leaq 0x98(%rsp), %rdi leaq 0x8(%rsp), %rdx leaq 0x20(%rsp), %r8 movl %r12d, %ecx callq 0x2ea264 jmp 0x2e3f34 movabsq $0x7ff0000000000000, %r9 # imm = 0x7FF0000000000000 vmovaps %xmm1, (%rsi) movq %r14, 0x10(%rsi) movq %rsp, %rax movq %rax, 0x18(%rsi) vmovaps %xmm1, 0x20(%rsi) movq %r9, 0x30(%rsi) andq $0x0, 0x48(%rsi) vmovups %xmm1, 0x38(%rsi) movl $0x7fffffff, 0x50(%rsi) # imm = 0x7FFFFFFF leaq 0x98(%rsp), %rdi leaq 0x8(%rsp), %rdx leaq 0x20(%rsp), %r8 movl %r12d, %ecx callq 0x2eb02c leaq 0x98(%rsp), %r13 movq %rbx, %rdi movq %r13, %rsi callq 0x41d9e movq %r13, %rdi callq 0xea60 leaq 0x68(%rsp), %rdi callq 0x357fe leaq 0x20(%rsp), %rdi callq 0x4495c leaq 0x90(%r15), %r13 movq %rbx, %rdi movq %r15, %rsi callq 0x31d28 leaq 0x50(%rbx), %rdi movq %r13, %rsi callq 0x352ec addq $0xc0, %r15 leaq 0x20(%rbx), %rdi movq %r15, %rsi callq 0x352c2 movq 0x10(%rsp), %rax movl %r12d, 0x90(%rbx) subq 0x8(%rsp), %rax pushq $0x28 popq %rcx cqto idivq %rcx movl %eax, 0x94(%rbx) cmpq $0x0, 0x168(%r14) je 0x2e3fcf addq $0x158, %r14 # imm = 0x158 movq %r14, %rdi movq %rbx, %rsi callq 0x357ea leaq 0x8(%rsp), %rdi callq 0x35858 movq %rbx, %rax addq $0x140, %rsp # imm = 0x140 popq %rbx popq %r12 popq %r13 popq %r14 popq %r15 retq jmp 0x2e3ff1 jmp 0x2e3ff1 leaq 0x68(%rsp), %rdi movq %rax, %r14 callq 0x357fe leaq 0x20(%rsp), %rdi callq 0x4495c jmp 0x2e4014 jmp 0x2e4011 movq %rax, %r14 jmp 0x2e401e movq %rax, %r14 leaq 0x8(%rsp), %rdi callq 0x35858 movq %rbx, %rdi callq 0xea60 movq %r14, %rdi callq 0xb3d0
/quesnel[P]baryonyx/lib/src/itm-solver-common.hpp
baryonyx::result baryonyx::itm::optimize_problem<baryonyx::itm::solver_inequalities_101coeff<float, baryonyx::itm::minimize_tag, baryonyx::itm::quadratic_cost_type<float>, false>, float, baryonyx::itm::minimize_tag, baryonyx::itm::quadratic_cost_type<float>>(baryonyx::context const&, baryonyx::problem const&)
inline result optimize_problem(const context& ctx, const problem& pb) { result r; if (ctx.start) ctx.start(ctx.parameters); auto constraints{ make_merged_constraints(ctx, pb) }; if (constraints.empty() || pb.vars.values.empty()) { r.status = result_status::success; r.solutions.resize(1); r.solutions.back().value = pb.objective.value; r.strings = pb.strings; r.affected_vars = pb.affected_vars; r.variable_name = pb.vars.names; return r; } random_engine rng(init_random_generator_seed(ctx)); auto variables = length(pb.vars.values); auto cost = Cost(pb.objective, variables); auto cost_constant = pb.objective.value; const auto thread = get_thread_number(ctx); std::vector<optimize_functor<Solver, Float, Mode, Cost>> functors; std::vector<std::thread> pool(thread); best_solution_recorder<Cost, Float, Mode> best_recorder( rng, thread, cost, cost_constant, variables, constraints, ctx.parameters.init_population_size); auto seeds = generate_seed(rng, thread); std::atomic_bool stop_task; stop_task.store(false); for (unsigned i = 0u; i != thread; ++i) functors.emplace_back(ctx, i, seeds[i]); for (unsigned i = 0u; i != thread; ++i) pool[i] = std::thread(std::ref(functors[i]), std::ref(stop_task), std::ref(best_recorder), std::cref(constraints), variables, std::cref(cost), cost_constant); const auto start = std::chrono::steady_clock::now(); auto end = start; do { std::this_thread::sleep_for(std::chrono::seconds{ 1L }); if (ctx.update) { auto call_number = 0L; for (auto i = 0u; i != thread; ++i) call_number += functors[i].m_call_number; int constraints_remaining; long int loop; double value; double duration; best_recorder.get_best( constraints_remaining, value, duration, loop); ctx.update( constraints_remaining, value, loop, duration, call_number); } end = std::chrono::steady_clock::now(); } while (!is_time_limit(ctx.parameters.time_limit, start, end)); stop_task.store(true); for (auto& t : pool) t.join(); r.strings = pb.strings; r.affected_vars = pb.affected_vars; r.variable_name = pb.vars.names; r.variables = variables; r.constraints = length(constraints); const auto& first = best_recorder.get_best(0); if (!first.is_solution()) r.status = result_status::time_limit_reached; else r.status = result_status::success; r.duration = first.duration; r.loop = first.loop; r.remaining_constraints = first.remaining_constraints; switch (ctx.parameters.storage) { case solver_parameters::storage_type::one: { r.solutions.resize(1); convert(first, r.solutions[0], variables); } break; case solver_parameters::storage_type::bound: { r.solutions.resize(2); convert(first, r.solutions[0], variables); convert(best_recorder.get_worst(), r.solutions[1], variables); } break; case solver_parameters::storage_type::five: { r.solutions.resize(5); for (int i = 0; i != 5; ++i) convert(best_recorder.get_best(i), r.solutions[i], variables); } break; } best_recorder.show_population(ctx); if (ctx.finish) ctx.finish(r); return r; }
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x198, %rsp # imm = 0x198 vxorpd %xmm0, %xmm0, %xmm0 movq %rdx, %r13 movq %rsi, %r15 movq %rdi, %r12 vmovupd %zmm0, 0x58(%rdi) vmovupd %zmm0, 0x40(%rdi) vmovupd %zmm0, (%rdi) movq $0x7fffffff, 0x98(%rdi) # imm = 0x7FFFFFFF movl $0x2, 0xa0(%rdi) cmpq $0x0, 0x128(%rsi) je 0x2e4094 leaq 0x118(%r15), %rdi movq %r15, %rsi vzeroupper callq 0x3527e leaq 0x48(%rsp), %rdi movq %r15, %rsi movq %r13, %rdx vzeroupper callq 0x450078 movq 0x48(%rsp), %rax cmpq 0x50(%rsp), %rax je 0x2e41f3 movq 0xa8(%r13), %rax cmpq 0xb0(%r13), %rax je 0x2e41f3 movq %r15, %rdi callq 0x353ba movl $0x7fffffff, %ecx # imm = 0x7FFFFFFF xorl %edx, %edx divq %rcx cmpq $0x1, %rdx adcq $0x0, %rdx movq %rdx, 0x40(%rsp) movq 0xb0(%r13), %rax subq 0xa8(%r13), %rax pushq $0xc popq %rcx cqto leaq 0x10(%r13), %rsi idivq %rcx movl %eax, 0x4(%rsp) leaq 0xb8(%rsp), %rdi movl %eax, %edx movq %r12, 0x10(%rsp) callq 0x445b4 vmovsd 0x40(%r13), %xmm0 vmovsd %xmm0, 0x38(%rsp) movq %r15, %rdi callq 0x35447 andq $0x0, 0x70(%rsp) vxorpd %xmm0, %xmm0, %xmm0 movl %eax, %ebp vmovapd %xmm0, 0x60(%rsp) leaq 0xa0(%rsp), %rdi leaq 0xd8(%rsp), %rdx movq %rbp, %rsi callq 0x35462 vmovsd 0x38(%rsp), %xmm0 movl 0xc4(%r15), %eax movl 0x4(%rsp), %r8d subq $0x8, %rsp movl %ebp, %edx leaq 0xe0(%rsp), %rdi leaq 0x48(%rsp), %rsi leaq 0xc0(%rsp), %rcx leaq 0x50(%rsp), %r9 pushq %rax callq 0x58da6 popq %rax popq %rcx leaq 0x30(%rsp), %rdi leaq 0x40(%rsp), %rsi movl %ebp, %edx callq 0x3551c xorl %eax, %eax movb %al, 0x3(%rsp) xorl %ecx, %ecx leaq 0x60(%rsp), %rbx leaq 0x8(%rsp), %r14 xchgb %cl, 0x3(%rsp) leaq 0x90(%r13), %rcx movq %rcx, 0x88(%rsp) movl %eax, 0x8(%rsp) cmpl %ebp, %eax je 0x2e424e movl %eax, %ecx shlq $0x3, %rcx addq 0x30(%rsp), %rcx movq %rbx, %rdi movq %r15, %rsi movq %r14, %rdx callq 0x2eb94e movl 0x8(%rsp), %eax incl %eax jmp 0x2e41ca andl $0x0, 0xa0(%r12) leaq 0x68(%r12), %rdi pushq $0x1 popq %rsi callq 0x35292 vmovsd 0x40(%r13), %xmm0 movq 0x70(%r12), %rax vmovsd %xmm0, -0x8(%rax) movq %r12, %rdi movq %r13, %rsi callq 0x31d28 leaq 0xc0(%r13), %rsi leaq 0x20(%r12), %rdi callq 0x352c2 addq $0x90, %r13 leaq 0x50(%r12), %rdi movq %r13, %rsi callq 0x352ec jmp 0x2e45e9 movq %r13, 0x80(%rsp) movq %r15, %r13 leaq 0x8(%rsp), %r15 leaq 0x98(%rsp), %r14 leaq 0x4(%rsp), %rbx xorl %r12d, %r12d cmpq %rbp, %r12 je 0x2e4312 imulq $0xc0, %r12, %rax addq 0x60(%rsp), %rax leaq 0x3(%rsp), %r10 leaq 0xd8(%rsp), %r11 movq %r10, 0x20(%rsp) leaq 0x48(%rsp), %r10 movq %r11, 0x18(%rsp) movq %r10, 0x98(%rsp) movq %rax, 0x28(%rsp) leaq 0xb8(%rsp), %rax movq %rax, 0x90(%rsp) leaq 0x20(%rsp), %rdx leaq 0x38(%rsp), %rax movq %r15, %rdi leaq 0x28(%rsp), %rsi leaq 0x18(%rsp), %rcx movq %r14, %r8 movq %rbx, %r9 pushq %rax leaq 0x98(%rsp), %rax pushq %rax callq 0x2eb99a popq %rax popq %rcx leaq (,%r12,8), %rdi addq 0xa0(%rsp), %rdi movq %r15, %rsi callq 0x356d6 movq %r15, %rdi callq 0x356f0 incq %r12 jmp 0x2e426e callq 0xb5a0 imulq $0xc0, %rbp, %r15 leaq 0x138(%r13), %r12 leaq 0x8(%rsp), %rbp movq %rax, %rbx movq $0x1, 0x8(%rsp) movq %rbp, %rdi callq 0x356fd cmpq $0x0, 0x148(%r13) je 0x2e43a7 movq 0x60(%rsp), %rax xorl %ecx, %ecx xorl %r14d, %r14d cmpq %rcx, %r15 je 0x2e4368 addq 0xb0(%rax,%rcx), %r14 addq $0xc0, %rcx jmp 0x2e4352 leaq 0xe0(%rsp), %rdi leaq 0x18(%rsp), %rsi leaq 0x28(%rsp), %rdx leaq 0x20(%rsp), %rcx movq %rbp, %r8 callq 0x60418 vmovsd 0x28(%rsp), %xmm0 vmovsd 0x20(%rsp), %xmm1 movq 0x8(%rsp), %rdx movl 0x18(%rsp), %esi movq %r12, %rdi movq %r14, %rcx callq 0x35736 callq 0xb5a0 vmovsd (%r13), %xmm0 subq %rbx, %rax vxorpd %xmm1, %xmm1, %xmm1 vucomisd %xmm0, %xmm1 vcvtsi2sd %rax, %xmm2, %xmm1 vdivsd 0x1756b6(%rip), %xmm1, %xmm1 # 0x459a80 jae 0x2e432d vucomisd %xmm0, %xmm1 jbe 0x2e432d movb $0x1, %al xchgb %al, 0x3(%rsp) movq 0xa0(%rsp), %rbx movq 0xa8(%rsp), %r14 movq 0x10(%rsp), %r12 cmpq %r14, %rbx je 0x2e4408 movq %rbx, %rdi callq 0xb1a0 addq $0x8, %rbx jmp 0x2e43f5 movq 0x80(%rsp), %rbx movq %r12, %rdi movq %rbx, %rsi callq 0x31d28 addq $0xc0, %rbx leaq 0x20(%r12), %rdi movq %rbx, %rsi callq 0x352c2 leaq 0x50(%r12), %rdi movq 0x88(%rsp), %rsi callq 0x352ec movl 0x4(%rsp), %eax movl %eax, 0x90(%r12) movq 0x50(%rsp), %rax subq 0x48(%rsp), %rax pushq $0x28 popq %rcx cqto xorl %esi, %esi idivq %rcx movq 0x118(%rsp), %rdi movq 0x130(%rsp), %rcx movl %eax, 0x94(%r12) movslq (%rdi), %rax imulq $0x38, %rax, %rax movl 0x30(%rcx,%rax), %edx leaq (%rcx,%rax), %rbx testl %edx, %edx setne %sil leal (%rsi,%rsi,2), %esi movl %esi, 0xa0(%r12) vmovsd 0x18(%rcx,%rax), %xmm0 vmovsd %xmm0, 0x80(%r12) movq 0x28(%rcx,%rax), %rax movq %rax, 0x88(%r12) movl %edx, 0x98(%r12) movl 0xe8(%r13), %eax testl %eax, %eax je 0x2e4563 cmpl $0x1, %eax je 0x2e452b cmpl $0x2, %eax jne 0x2e4582 leaq 0x68(%r12), %rbx pushq $0x5 popq %rsi movq %rbx, %rdi callq 0x35292 xorl %r14d, %r14d xorl %r15d, %r15d cmpq $0x5, %r15 je 0x2e4582 movq 0x118(%rsp), %rax movq (%rbx), %rsi movl 0x4(%rsp), %edx movslq (%rax,%r15,4), %rax addq %r14, %rsi imulq $0x38, %rax, %rdi addq 0x130(%rsp), %rdi callq 0x4f485 incq %r15 addq $0x20, %r14 jmp 0x2e44f1 leaq 0x68(%r12), %r14 pushq $0x2 popq %rsi movq %r14, %rdi callq 0x35292 movl 0x4(%rsp), %edx movq (%r14), %rsi movq %rbx, %rdi callq 0x4f485 leaq 0xe0(%rsp), %rdi callq 0x60496 movq (%r14), %rsi movq %rax, %rbx addq $0x20, %rsi jmp 0x2e4576 leaq 0x68(%r12), %r14 pushq $0x1 popq %rsi movq %r14, %rdi callq 0x35292 movq (%r14), %rsi movl 0x4(%rsp), %edx movq %rbx, %rdi callq 0x4f485 leaq 0xd8(%rsp), %rdi movq %r13, %rsi callq 0x58f0a cmpq $0x0, 0x168(%r13) je 0x2e45ae addq $0x158, %r13 # imm = 0x158 movq %r13, %rdi movq %r12, %rsi callq 0x357ea leaq 0x30(%rsp), %rdi callq 0x357fe leaq 0xd8(%rsp), %rdi callq 0x58f1e leaq 0xa0(%rsp), %rdi callq 0x35836 leaq 0x60(%rsp), %rdi callq 0x2f0ace leaq 0xb8(%rsp), %rdi callq 0x4495c leaq 0x48(%rsp), %rdi callq 0x35858 movq %r12, %rax addq $0x198, %rsp # imm = 0x198 popq %rbx popq %r12 popq %r13 popq %r14 popq %r15 popq %rbp retq movq %rax, %rbx jmp 0x2e464a movq %rax, %rbx jmp 0x2e4657 movq %rax, %rbx jmp 0x2e4664 jmp 0x2e462e jmp 0x2e461f jmp 0x2e463d jmp 0x2e463d movq %r12, 0x10(%rsp) movq %rax, %rbx jmp 0x2e4685 movq %r12, 0x10(%rsp) movq %rax, %rbx jmp 0x2e467b jmp 0x2e463d jmp 0x2e463d jmp 0x2e463d jmp 0x2e463d jmp 0x2e463d movq %rax, %rbx leaq 0x30(%rsp), %rdi callq 0x357fe leaq 0xd8(%rsp), %rdi callq 0x58f1e leaq 0xa0(%rsp), %rdi callq 0x35836 leaq 0x60(%rsp), %rdi callq 0x2f0ace leaq 0xb8(%rsp), %rdi callq 0x4495c leaq 0x48(%rsp), %rdi callq 0x35858 movq 0x10(%rsp), %rdi callq 0xea60 movq %rbx, %rdi callq 0xb3d0
/quesnel[P]baryonyx/lib/src/itm-optimizer-common.hpp
baryonyx::result baryonyx::itm::solve_problem<baryonyx::itm::solver_inequalities_101coeff<float, baryonyx::itm::minimize_tag, baryonyx::itm::quadratic_cost_type<float>, false>, float, baryonyx::itm::minimize_tag, baryonyx::itm::quadratic_cost_type<float>>(baryonyx::context const&, baryonyx::problem const&)
inline result solve_problem(const context& ctx, const problem& pb) { if (ctx.start) ctx.start(ctx.parameters); result ret; auto variables = length(pb.vars.values); auto constraints{ make_merged_constraints(ctx, pb) }; if (!constraints.empty() && !pb.vars.values.empty()) { random_engine rng(init_random_generator_seed(ctx)); auto cost = Cost(pb.objective, variables); auto cost_constant = pb.objective.value; switch (ctx.parameters.observer) { case solver_parameters::observer_type::pnm: { using obs = pnm_observer; solver_functor<Solver, Float, Mode, Cost, obs> slv(ctx, rng); ret = slv(constraints, variables, cost, cost_constant); } break; case solver_parameters::observer_type::file: { using obs = file_observer; solver_functor<Solver, Float, Mode, Cost, obs> slv(ctx, rng); ret = slv(constraints, variables, cost, cost_constant); } break; default: { using obs = none_observer; solver_functor<Solver, Float, Mode, Cost, obs> slv(ctx, rng); ret = slv(constraints, variables, cost, cost_constant); break; } } } else { ret.status = result_status::success; ret.solutions.resize(1); ret.solutions.back().value = pb.objective.value; } ret.strings = pb.strings; ret.variable_name = std::move(pb.vars.names); ret.affected_vars = std::move(pb.affected_vars); ret.variables = variables; ret.constraints = length(constraints); if (ctx.finish) ctx.finish(ret); return ret; }
pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x140, %rsp # imm = 0x140 cmpq $0x0, 0x128(%rsi) movq %rdx, %r15 movq %rsi, %r14 movq %rdi, %rbx je 0x2e46c9 leaq 0x118(%r14), %rdi movq %r14, %rsi callq 0x3527e vxorps %xmm0, %xmm0, %xmm0 vmovups %zmm0, 0x58(%rbx) vmovups %zmm0, 0x40(%rbx) vmovups %zmm0, (%rbx) movq $0x7fffffff, 0x98(%rbx) # imm = 0x7FFFFFFF movl $0x2, 0xa0(%rbx) movq 0xb0(%r15), %rax subq 0xa8(%r15), %rax pushq $0xc popq %rcx cqto idivq %rcx movq %rax, %r12 leaq 0x8(%rsp), %rdi movq %r14, %rsi movq %r15, %rdx vzeroupper callq 0x450078 movq 0x8(%rsp), %rax cmpq 0x10(%rsp), %rax je 0x2e47f2 movq 0xa8(%r15), %rax cmpq 0xb0(%r15), %rax je 0x2e47f2 movq %r14, %rdi callq 0x353ba movl $0x7fffffff, %ecx # imm = 0x7FFFFFFF xorl %edx, %edx leaq 0x10(%r15), %rsi divq %rcx cmpq $0x1, %rdx adcq $0x0, %rdx movq %rdx, (%rsp) leaq 0x20(%rsp), %rdi movl %r12d, %edx callq 0x445b4 vmovsd 0x40(%r15), %xmm0 movl 0xe4(%r14), %eax vxorps %xmm1, %xmm1, %xmm1 leaq 0x40(%rsp), %rsi cmpl $0x1, %eax je 0x2e4819 cmpl $0x2, %eax jne 0x2e4868 movabsq $0x7ff0000000000000, %r9 # imm = 0x7FF0000000000000 vmovaps %xmm1, (%rsi) movq %r14, 0x10(%rsi) movq %rsp, %rax movq %rax, 0x18(%rsi) vmovaps %xmm1, 0x20(%rsi) movq %r9, 0x30(%rsi) andq $0x0, 0x48(%rsi) vmovups %xmm1, 0x38(%rsi) movl $0x7fffffff, 0x50(%rsi) # imm = 0x7FFFFFFF leaq 0x98(%rsp), %rdi leaq 0x8(%rsp), %rdx leaq 0x20(%rsp), %r8 movl %r12d, %ecx callq 0x2f11da jmp 0x2e48b5 andl $0x0, 0xa0(%rbx) leaq 0x68(%rbx), %rdi pushq $0x1 popq %rsi callq 0x35292 vmovsd 0x40(%r15), %xmm0 movq 0x70(%rbx), %rax vmovsd %xmm0, -0x8(%rax) jmp 0x2e48e4 movabsq $0x7ff0000000000000, %r9 # imm = 0x7FF0000000000000 vmovaps %xmm1, (%rsi) movq %r14, 0x10(%rsi) movq %rsp, %rax movq %rax, 0x18(%rsi) vmovaps %xmm1, 0x20(%rsi) movq %r9, 0x30(%rsi) andq $0x0, 0x48(%rsi) vmovups %xmm1, 0x38(%rsi) movl $0x7fffffff, 0x50(%rsi) # imm = 0x7FFFFFFF leaq 0x98(%rsp), %rdi leaq 0x8(%rsp), %rdx leaq 0x20(%rsp), %r8 movl %r12d, %ecx callq 0x2f0af6 jmp 0x2e48b5 movabsq $0x7ff0000000000000, %r9 # imm = 0x7FF0000000000000 vmovaps %xmm1, (%rsi) movq %r14, 0x10(%rsi) movq %rsp, %rax movq %rax, 0x18(%rsi) vmovaps %xmm1, 0x20(%rsi) movq %r9, 0x30(%rsi) andq $0x0, 0x48(%rsi) vmovups %xmm1, 0x38(%rsi) movl $0x7fffffff, 0x50(%rsi) # imm = 0x7FFFFFFF leaq 0x98(%rsp), %rdi leaq 0x8(%rsp), %rdx leaq 0x20(%rsp), %r8 movl %r12d, %ecx callq 0x2f18be leaq 0x98(%rsp), %r13 movq %rbx, %rdi movq %r13, %rsi callq 0x41d9e movq %r13, %rdi callq 0xea60 leaq 0x68(%rsp), %rdi callq 0x357fe leaq 0x20(%rsp), %rdi callq 0x4495c leaq 0x90(%r15), %r13 movq %rbx, %rdi movq %r15, %rsi callq 0x31d28 leaq 0x50(%rbx), %rdi movq %r13, %rsi callq 0x352ec addq $0xc0, %r15 leaq 0x20(%rbx), %rdi movq %r15, %rsi callq 0x352c2 movq 0x10(%rsp), %rax movl %r12d, 0x90(%rbx) subq 0x8(%rsp), %rax pushq $0x28 popq %rcx cqto idivq %rcx movl %eax, 0x94(%rbx) cmpq $0x0, 0x168(%r14) je 0x2e4950 addq $0x158, %r14 # imm = 0x158 movq %r14, %rdi movq %rbx, %rsi callq 0x357ea leaq 0x8(%rsp), %rdi callq 0x35858 movq %rbx, %rax addq $0x140, %rsp # imm = 0x140 popq %rbx popq %r12 popq %r13 popq %r14 popq %r15 retq jmp 0x2e4972 jmp 0x2e4972 leaq 0x68(%rsp), %rdi movq %rax, %r14 callq 0x357fe leaq 0x20(%rsp), %rdi callq 0x4495c jmp 0x2e4995 jmp 0x2e4992 movq %rax, %r14 jmp 0x2e499f movq %rax, %r14 leaq 0x8(%rsp), %rdi callq 0x35858 movq %rbx, %rdi callq 0xea60 movq %r14, %rdi callq 0xb3d0 nop
/quesnel[P]baryonyx/lib/src/itm-solver-common.hpp
baryonyx::itm::optimize_functor<baryonyx::itm::solver_inequalities_101coeff<float, baryonyx::itm::minimize_tag, baryonyx::itm::quadratic_cost_type<float>, true>, float, baryonyx::itm::minimize_tag, baryonyx::itm::quadratic_cost_type<float>>::operator()(std::atomic<bool> const&, baryonyx::itm::best_solution_recorder<baryonyx::itm::quadratic_cost_type<float>, float, baryonyx::itm::minimize_tag>&, std::vector<baryonyx::itm::merged_constraint, std::allocator<baryonyx::itm::merged_constraint>> const&, int, baryonyx::itm::quadratic_cost_type<float> const&, double)
void operator()(const std::atomic_bool& stop_task, best_solution_recorder<Cost, Float, Mode>& best_recorder, const std::vector<merged_constraint>& constraints, int variables, const Cost& original_costs, double cost_constant) { bit_array x(variables); auto& p = m_ctx.parameters; auto norm_costs = normalize_costs<Float, Cost>( m_ctx, original_costs, m_local_ctx.rng, variables); const auto kappa_step = static_cast<Float>(p.kappa_step); const auto kappa_max = static_cast<Float>(p.kappa_max); const auto alpha = static_cast<Float>(p.alpha); const auto theta = static_cast<Float>(p.theta); const auto delta = p.delta < 0 ? compute_delta<Float, Cost>(m_ctx, norm_costs, theta, variables) : static_cast<Float>(p.delta); const auto pushing_k_factor = static_cast<Float>(p.pushing_k_factor); const auto pushing_objective_amplifier = static_cast<Float>(p.pushing_objective_amplifier); const auto w_limit = static_cast<long int>(p.w); Solver slv(m_local_ctx.rng, length(constraints), variables, norm_costs, constraints); compute_order compute(p.order, variables); bool is_a_solution = false; while (!stop_task.load()) { ++m_call_number; const auto kappa_start = static_cast<Float>(best_recorder.reinit( m_local_ctx, is_a_solution, p.kappa_min, p.kappa_max, x)); auto kappa = kappa_start; compute.init(slv, x); auto best_remaining = INT_MAX; is_a_solution = false; for (long int i = 0; !stop_task.load() && i != p.limit; ++i) { auto remaining = compute.run(slv, x, m_local_ctx.rng, kappa, delta, theta); if (remaining == 0) { best_recorder.try_update( m_local_ctx, x, original_costs.results(x, cost_constant), i); best_remaining = 0; is_a_solution = true; break; } else { best_remaining = std::min(remaining, best_remaining); } if (i > w_limit) kappa += kappa_step * std::pow(static_cast<Float>(remaining) / static_cast<Float>(slv.m), alpha); if (kappa > kappa_max) break; } if (best_remaining > 0) { best_recorder.try_advance( m_local_ctx, x, best_remaining, p.limit); continue; } for (int push = 0; !stop_task.load() && push < p.pushes_limit; ++push) { auto remaining = compute.push_and_run(slv, x, m_local_ctx.rng, pushing_k_factor, delta, theta, pushing_objective_amplifier); if (remaining == 0) { best_recorder.try_update( m_local_ctx, x, original_costs.results(x, cost_constant), -push * p.pushing_iteration_limit - 1); is_a_solution = true; } kappa = kappa_start; for (int iter = 0; !stop_task.load() && iter < p.pushing_iteration_limit; ++iter) { remaining = compute.run( slv, x, m_local_ctx.rng, kappa, delta, theta); if (remaining == 0) { best_recorder.try_update( m_local_ctx, x, original_costs.results(x, cost_constant), -push * p.pushing_iteration_limit - iter - 1); is_a_solution = true; break; } if (iter > p.w) kappa += kappa_step * std::pow(static_cast<Float>(remaining) / static_cast<Float>(slv.m), alpha); if (kappa > kappa_max) break; } } } }
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x168, %rsp # imm = 0x168 movq %rdi, %rbp leaq 0x18(%rsp), %rdi movq %rsi, %r15 movl %r8d, %esi vmovsd %xmm0, 0x58(%rsp) movq %r9, %r12 movl %r8d, %ebx movq %rcx, %r14 movq %rdx, 0x28(%rsp) callq 0x368ae movq (%rbp), %r13 movq %rbp, 0x70(%rsp) addq $0x8, %rbp leaq 0x80(%rsp), %rdi movq %r13, %rsi movq %r12, 0x50(%rsp) movq %r12, %rdx movq %rbp, %rcx movl %ebx, %r8d callq 0x46c07 vmovsd 0x20(%r13), %xmm0 vmovsd 0x10(%r13), %xmm2 vxorpd %xmm1, %xmm1, %xmm1 vmovsd %xmm0, (%rsp) vmovsd 0x28(%r13), %xmm0 vucomisd %xmm2, %xmm1 vmovsd %xmm0, 0x48(%rsp) vmovsd 0x30(%r13), %xmm0 vmovsd %xmm0, 0x40(%rsp) vcvtsd2ss 0x8(%r13), %xmm3, %xmm0 vmovss %xmm0, 0xc(%rsp) jbe 0x2e4e85 movq 0x70(%rsp), %rax movq (%rax), %rdi leaq 0x80(%rsp), %rsi movl %ebx, %edx callq 0x46ce1 vmovss %xmm0, 0x8(%rsp) jmp 0x2e4e8f vcvtsd2ss %xmm2, %xmm2, %xmm0 vmovss %xmm0, 0x8(%rsp) vcvttsd2si 0xa0(%r13), %rax vmovsd 0x38(%r13), %xmm0 vmovsd 0x40(%r13), %xmm1 movq %rax, 0x78(%rsp) movq 0x8(%r14), %rax vmovsd %xmm0, 0x38(%rsp) vmovsd %xmm1, 0x60(%rsp) subq (%r14), %rax pushq $0x28 popq %rcx cqto idivq %rcx leaq 0xd8(%rsp), %rdi leaq 0x80(%rsp), %r8 movq %rbp, %rsi movl %eax, %edx movl %ebx, %ecx movq %r14, %r9 callq 0x2e52de movl 0xcc(%r13), %esi leaq 0xa0(%rsp), %rdi movl %ebx, %edx callq 0x38cd8 vcvtsd2ss (%rsp), %xmm3, %xmm0 vmovss %xmm0, 0x34(%rsp) vcvtsd2ss 0x48(%rsp), %xmm3, %xmm0 vmovss %xmm0, 0x48(%rsp) vcvtsd2ss 0x40(%rsp), %xmm3, %xmm0 vmovss %xmm0, 0x40(%rsp) vcvtsd2ss 0x38(%rsp), %xmm3, %xmm0 vmovss %xmm0, 0x6c(%rsp) vcvtsd2ss 0x60(%rsp), %xmm3, %xmm0 vmovss %xmm0, 0x68(%rsp) movq $0x0, 0x10(%rsp) movb (%r15), %al testb $0x1, %al jne 0x2e5245 movq 0x70(%rsp), %rax incq 0xb0(%rax) vmovsd 0x18(%r13), %xmm0 vmovsd 0x28(%r13), %xmm1 movzbl 0x10(%rsp), %edx movq 0x28(%rsp), %rdi leaq 0x18(%rsp), %rbx movq %rbp, %rsi movq %rbx, %rcx andl $0x1, %edx callq 0x5b276 vmovsd %xmm0, (%rsp) leaq 0xa0(%rsp), %rdi leaq 0xd8(%rsp), %rsi movq %rbx, %rdx callq 0x2e55a2 vcvtsd2ss (%rsp), %xmm3, %xmm0 movl $0x7fffffff, %ebx # imm = 0x7FFFFFFF xorl %r14d, %r14d vmovss %xmm0, 0x60(%rsp) movb (%r15), %al testb $0x1, %al jne 0x2e5046 cmpq 0xb0(%r13), %r14 je 0x2e5046 vmovss 0x8(%rsp), %xmm1 vmovss 0xc(%rsp), %xmm2 leaq 0xa0(%rsp), %rdi leaq 0xd8(%rsp), %rsi leaq 0x18(%rsp), %rdx movq %rbp, %rcx vmovss %xmm0, (%rsp) callq 0x2e55e4 testl %eax, %eax je 0x2e505e cmpl %eax, %ebx cmovgel %eax, %ebx cmpq 0x78(%rsp), %r14 jle 0x2e5031 vcvtsi2ss %eax, %xmm3, %xmm0 vcvtsi2ssl 0x160(%rsp), %xmm3, %xmm1 vdivss %xmm1, %xmm0, %xmm0 vmovss 0x40(%rsp), %xmm1 callq 0xb530 vmovss (%rsp), %xmm1 vfmadd231ss 0x34(%rsp), %xmm0, %xmm1 # xmm1 = (xmm0 * mem) + xmm1 vmovaps %xmm1, %xmm0 jmp 0x2e5036 vmovss (%rsp), %xmm0 vucomiss 0x48(%rsp), %xmm0 ja 0x2e5046 incq %r14 jmp 0x2e4fad testl %ebx, %ebx jg 0x2e5225 leaq 0x18(%rsp), %rbx movq $0x0, 0x10(%rsp) jmp 0x2e5090 vmovsd 0x58(%rsp), %xmm0 movq 0x50(%rsp), %rdi leaq 0x18(%rsp), %rbx movq %rbx, %rsi callq 0x45118 movq 0x28(%rsp), %rdi movq %rbp, %rsi movq %rbx, %rdx movq %r14, %rcx callq 0x5b5c4 movb $0x1, %al movq %rax, 0x10(%rsp) xorl %r14d, %r14d movl $0x0, 0x38(%rsp) movb (%r15), %al testb $0x1, %al jne 0x2e4f3d cmpl 0xbc(%r13), %r14d jge 0x2e4f3d vmovss 0x6c(%rsp), %xmm0 vmovss 0x8(%rsp), %xmm1 vmovss 0xc(%rsp), %xmm2 vmovss 0x68(%rsp), %xmm3 leaq 0xa0(%rsp), %rdi leaq 0xd8(%rsp), %rsi movq %rbx, %rdx movq %rbp, %rcx callq 0x2e57f0 testl %eax, %eax jne 0x2e5124 vmovsd 0x58(%rsp), %xmm0 movq 0x50(%rsp), %rdi movq %rbx, %rsi callq 0x45118 movl 0xc0(%r13), %eax movq 0x28(%rsp), %rdi movq %rbp, %rsi movq %rbx, %rdx imull %r14d, %eax notl %eax movslq %eax, %rcx callq 0x5b5c4 movb $0x1, %al movq %rax, 0x10(%rsp) pushq $0x1 popq %rax vmovss 0x60(%rsp), %xmm0 movl %eax, %ebx movb (%r15), %al testb $0x1, %al jne 0x2e51cf leal -0x1(%rbx), %r12d cmpl 0xc0(%r13), %r12d jge 0x2e51cf vmovss 0x8(%rsp), %xmm1 vmovss 0xc(%rsp), %xmm2 leaq 0xa0(%rsp), %rdi leaq 0xd8(%rsp), %rsi leaq 0x18(%rsp), %rdx movq %rbp, %rcx vmovss %xmm0, (%rsp) callq 0x2e55e4 testl %eax, %eax je 0x2e51d6 vcvtsi2sd %r12d, %xmm4, %xmm0 vucomisd 0xa0(%r13), %xmm0 jbe 0x2e51bb vcvtsi2ss %eax, %xmm4, %xmm0 vcvtsi2ssl 0x160(%rsp), %xmm4, %xmm1 vdivss %xmm1, %xmm0, %xmm0 vmovss 0x40(%rsp), %xmm1 callq 0xb530 vmovss (%rsp), %xmm1 vfmadd231ss 0x34(%rsp), %xmm0, %xmm1 # xmm1 = (xmm0 * mem) + xmm1 vmovaps %xmm1, %xmm0 jmp 0x2e51c0 vmovss (%rsp), %xmm0 vucomiss 0x48(%rsp), %xmm0 ja 0x2e51cf incl %ebx jmp 0x2e512f leaq 0x18(%rsp), %rbx jmp 0x2e5219 vmovsd 0x58(%rsp), %xmm0 movq 0x50(%rsp), %rdi leaq 0x18(%rsp), %r12 movq %r12, %rsi callq 0x45118 movl 0xc0(%r13), %eax movq 0x28(%rsp), %rdi movq %rbp, %rsi movq %r12, %rdx imull 0x38(%rsp), %eax subl %ebx, %eax movq %r12, %rbx movslq %eax, %rcx callq 0x5b5c4 movb $0x1, %al movq %rax, 0x10(%rsp) decl 0x38(%rsp) incl %r14d jmp 0x2e509b movq 0xb0(%r13), %r8 movq 0x28(%rsp), %rdi leaq 0x18(%rsp), %rdx movq %rbp, %rsi movl %ebx, %ecx callq 0x5b64c jmp 0x2e4f34 leaq 0xa0(%rsp), %rdi callq 0x393ac leaq 0xd8(%rsp), %rdi callq 0x2e5a44 leaq 0x80(%rsp), %rdi callq 0x4495c leaq 0x20(%rsp), %rdi callq 0x357fe addq $0x168, %rsp # imm = 0x168 popq %rbx popq %r12 popq %r13 popq %r14 popq %r15 popq %rbp retq jmp 0x2e528f movq %rax, %rbx jmp 0x2e52b1 movq %rax, %rbx jmp 0x2e52be movq %rax, %rbx jmp 0x2e52cb jmp 0x2e52a1 jmp 0x2e52a1 jmp 0x2e52a1 jmp 0x2e52a1 leaq 0xa0(%rsp), %rdi movq %rax, %rbx callq 0x393ac leaq 0xd8(%rsp), %rdi callq 0x2e5a44 leaq 0x80(%rsp), %rdi callq 0x4495c leaq 0x20(%rsp), %rdi callq 0x357fe movq %rbx, %rdi callq 0xb3d0 nop
/quesnel[P]baryonyx/lib/src/itm-optimizer-common.hpp
baryonyx::itm::solver_inequalities_101coeff<float, baryonyx::itm::minimize_tag, baryonyx::itm::quadratic_cost_type<float>, true>::solver_inequalities_101coeff(std::linear_congruential_engine<unsigned long, 16807ul, 0ul, 2147483647ul>&, int, int, baryonyx::itm::quadratic_cost_type<float> const&, std::vector<baryonyx::itm::merged_constraint, std::allocator<baryonyx::itm::merged_constraint>> const&)
solver_inequalities_101coeff(random_engine& rng_, int m_, int n_, const cost_type& c_, const std::vector<merged_constraint>& csts) : logger("solver_inequalities_101coeff") , rng(rng_) , ap(csts, m_, n_) , P(std::make_unique<Float[]>(ap.size())) , A(static_cast<int>(ap.size())) , R(std::make_unique<rc_data[]>(compute_reduced_costs_vector_size(csts))) , b(std::make_unique<bound_factor[]>(m_)) , pi(std::make_unique<Float[]>(m_)) , c(c_) , m(m_) , n(n_) { int id = 0; for (int i = 0, e = length(csts); i != e; ++i) { int lower = 0, upper = 0; for (const auto& cst : csts[i].elements) { bx_ensures(std::abs(cst.factor) == 1); if (cst.factor > 0) A.set(id++); else A.unset(id++); if (cst.factor > 0) upper++; else lower++; } if (csts[i].min == csts[i].max) { b[i].min = csts[i].min; b[i].max = csts[i].max; } else { b[i].min = std::max(-lower, csts[i].min); b[i].max = std::min(upper, csts[i].max); } b[i].negative_factor = lower; bx_ensures(b[i].min <= b[i].max); bx_ensures(upper + lower == length(csts[i].elements)); } }
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x28, %rsp movl %edx, %ebp leaq 0x17e386(%rip), %rdx # 0x46367b movq %r9, %rbx movq %r8, %r15 movl %ecx, %r13d movq %rsi, %r14 movq %rdi, %r12 pushq $0x1c popq %rsi callq 0xb8418 movq %r14, 0x8(%r12) leaq 0x10(%r12), %r14 movq %r14, %rdi movq %rbx, 0x18(%rsp) movq %rbx, %rsi movl %ebp, %edx movl %r13d, %ecx callq 0x3a4ca movl 0x38(%r12), %esi leaq 0x50(%r12), %rbx movq %rbx, %rdi movq %r14, 0x20(%rsp) callq 0x35bd8 movl 0x38(%r12), %esi leaq 0x58(%r12), %r14 movq %r15, (%rsp) movq %r14, %rdi movq %rbx, 0x10(%rsp) callq 0x368ae movq 0x18(%rsp), %rdi movl %r13d, %r15d leaq 0x68(%r12), %r13 callq 0x3a89d movq %r13, %rbx movq %r13, %rdi movq %rax, %rsi callq 0x2e5a87 movslq %ebp, %r13 leaq 0x70(%r12), %rdi movq %r13, %rsi movq %rdi, 0x8(%rsp) callq 0x2e5ac4 leaq 0x78(%r12), %rdi movq %r13, %rsi callq 0x35bd8 movq (%rsp), %rax movq %rax, 0x80(%r12) movq 0x18(%rsp), %rax movl %ebp, 0x88(%r12) movl %r15d, 0x8c(%r12) movq (%rax), %rcx movq 0x8(%rax), %rax subq %rcx, %rax pushq $0x28 popq %rsi cqto xorl %r9d, %r9d xorl %ebp, %ebp idivq %rsi movl %eax, %eax movq %rax, 0x10(%rsp) cmpq 0x10(%rsp), %r9 je 0x2e54f4 imulq $0x28, %r9, %rax movq %r9, (%rsp) xorl %r15d, %r15d xorl %ebx, %ebx movq (%rcx,%rax), %r13 movq 0x8(%rcx,%rax), %r12 movq %rax, 0x20(%rsp) cmpq %r12, %r13 je 0x2e544d movl (%r13), %eax leal 0x1(%rax), %ecx testl $0xfffffffd, %ecx # imm = 0xFFFFFFFD jne 0x2e5503 movq %r14, %rdi movl %ebp, %esi testl %eax, %eax jle 0x2e542c callq 0x369c8 jmp 0x2e5431 callq 0x36a06 incl %ebp xorl %eax, %eax xorl %ecx, %ecx cmpl $0x0, (%r13) setle %al setg %cl addq $0x8, %r13 addl %ecx, %r15d addl %eax, %ebx jmp 0x2e5404 movq 0x18(%rsp), %rax movq 0x20(%rsp), %rsi movq (%rax), %rcx movl 0x18(%rcx,%rsi), %edx leaq (%rcx,%rsi), %rax cmpl 0x1c(%rcx,%rsi), %edx jne 0x2e5485 movq 0x8(%rsp), %rsi movq (%rsp), %r9 movq (%rsi), %rsi imulq $0xc, %r9, %rdi movl %edx, (%rsi,%rdi) movl %edx, 0x4(%rsi,%rdi) movl %ebx, 0x8(%rsi,%rdi) jmp 0x2e54ba movq 0x8(%rsp), %rdi movq (%rsp), %r9 movl %ebx, %esi negl %esi cmpl %esi, %edx cmovgl %edx, %esi movq (%rdi), %rdx imulq $0xc, %r9, %rdi movl %esi, (%rdx,%rdi) movl 0x1c(%rax), %r8d cmpl %r15d, %r8d cmovgel %r15d, %r8d movl %r8d, 0x4(%rdx,%rdi) movl %ebx, 0x8(%rdx,%rdi) cmpl %r8d, %esi jg 0x2e5524 movq 0x8(%rax), %rdx addl %r15d, %ebx incq %r9 subq (%rax), %rdx shrq $0x3, %rdx cmpl %edx, %ebx je 0x2e53de leaq 0x172f50(%rip), %rdi # 0x45842a leaq 0x17e22c(%rip), %rsi # 0x46370d leaq 0x17e1b0(%rip), %rdx # 0x463698 leaq 0x17e248(%rip), %rcx # 0x463737 callq 0x2813f addq $0x28, %rsp popq %rbx popq %r12 popq %r13 popq %r14 popq %r15 popq %rbp retq leaq 0x172f20(%rip), %rdi # 0x45842a leaq 0x174d1c(%rip), %rsi # 0x45a22d leaq 0x17e180(%rip), %rdx # 0x463698 leaq 0x17e1e6(%rip), %rcx # 0x463705 callq 0x2813f leaq 0x172eff(%rip), %rdi # 0x45842a leaq 0x174d77(%rip), %rsi # 0x45a2a9 leaq 0x17e15f(%rip), %rdx # 0x463698 leaq 0x17e1c9(%rip), %rcx # 0x463709 callq 0x2813f movq 0x8(%rsp), %rdi movq %rax, %r14 callq 0x2e5b02 jmp 0x2e5557 movq %rax, %r14 movq %rbx, %rdi callq 0x2e5b1c jmp 0x2e5564 movq %rax, %r14 leaq 0x60(%r12), %rdi callq 0x357fe jmp 0x2e5573 movq %rax, %r14 movq 0x10(%rsp), %rdi callq 0x35c1a jmp 0x2e5582 movq %rax, %r14 movq 0x20(%rsp), %rdi callq 0x3a954 movq %r12, %rdi callq 0xb853a movq %r14, %rdi callq 0xb3d0 movq %rax, %r14 jmp 0x2e558c nop
/quesnel[P]baryonyx/lib/src/itm-solver-inequalities-101.cpp
void baryonyx::itm::compute_order::init<baryonyx::itm::solver_inequalities_101coeff<float, baryonyx::itm::minimize_tag, baryonyx::itm::quadratic_cost_type<float>, true>, baryonyx::bit_array>(baryonyx::itm::solver_inequalities_101coeff<float, baryonyx::itm::minimize_tag, baryonyx::itm::quadratic_cost_type<float>, true> const&, baryonyx::bit_array const&)
void init(const Solver& s, const Xtype& x) { switch (order) { case solver_parameters::constraint_order::infeasibility_decr: case solver_parameters::constraint_order::infeasibility_incr: infeasibility_local_compute_violated_constraints(s, x); break; case solver_parameters::constraint_order::pi_sign_change: std::iota(R.begin(), R.end(), 0); break; case solver_parameters::constraint_order::none: case solver_parameters::constraint_order::reversing: case solver_parameters::constraint_order::random_sorting: case solver_parameters::constraint_order::lagrangian_decr: case solver_parameters::constraint_order::lagrangian_incr: default: compute_violated_constraints(s, x, R); break; } }
movl 0x30(%rdi), %ecx movq %rdi, %rax leal -0x3(%rcx), %edi cmpl $0x2, %edi jae 0x2e55b8 movq %rax, %rdi jmp 0x2e5b52 cmpl $0x7, %ecx jne 0x2e55d5 movq (%rax), %rcx movq 0x8(%rax), %rax xorl %edx, %edx cmpq %rax, %rcx je 0x2e55e3 movl %edx, (%rcx) incl %edx addq $0x4, %rcx jmp 0x2e55c6 movq %rsi, %rdi movq %rdx, %rsi movq %rax, %rdx jmp 0x2e5c6d retq
/quesnel[P]baryonyx/lib/src/itm-common.hpp
int baryonyx::itm::compute_order::run<baryonyx::itm::solver_inequalities_101coeff<float, baryonyx::itm::minimize_tag, baryonyx::itm::quadratic_cost_type<float>, true>, baryonyx::bit_array, float>(baryonyx::itm::solver_inequalities_101coeff<float, baryonyx::itm::minimize_tag, baryonyx::itm::quadratic_cost_type<float>, true>&, baryonyx::bit_array&, std::linear_congruential_engine<unsigned long, 16807ul, 0ul, 2147483647ul>&, float, float, float)
int run(Solver& solver, Xtype& x, random_engine& rng, Float kappa, Float delta, Float theta) { bool pi_changed = false; int remaining = 0; switch (order) { case solver_parameters::constraint_order::reversing: solver.compute_update_row( x, R.crbegin(), R.crend(), kappa, delta, theta); return compute_violated_constraints(solver, x, R); case solver_parameters::constraint_order::random_sorting: std::shuffle(R.begin(), R.end(), rng); solver.compute_update_row( x, R.begin(), R.end(), kappa, delta, theta); return compute_violated_constraints(solver, x, R); case solver_parameters::constraint_order::infeasibility_decr: std::sort(m_order.begin(), m_order.end(), [](const auto& lhs, const auto& rhs) { return rhs.second < lhs.second; }); solver.compute_update_row( x, m_order.cbegin(), m_order.cend(), kappa, delta, theta); return infeasibility_local_compute_violated_constraints(solver, x); case solver_parameters::constraint_order::infeasibility_incr: std::sort(m_order.begin(), m_order.end(), [](const auto& lhs, const auto& rhs) { return lhs.second < rhs.second; }); solver.compute_update_row( x, m_order.cbegin(), m_order.cend(), kappa, delta, theta); return infeasibility_local_compute_violated_constraints(solver, x); case solver_parameters::constraint_order::lagrangian_decr: std::sort( R.begin(), R.end(), [&solver](const auto lhs, const auto rhs) { return solver.pi[rhs] < solver.pi[lhs]; }); solver.compute_update_row( x, R.cbegin(), R.cend(), kappa, delta, theta); return compute_violated_constraints(solver, x, R); case solver_parameters::constraint_order::lagrangian_incr: std::sort( R.begin(), R.end(), [&solver](const auto lhs, const auto rhs) { return solver.pi[lhs] < solver.pi[rhs]; }); solver.compute_update_row( x, R.cbegin(), R.cend(), kappa, delta, theta); return compute_violated_constraints(solver, x, R); case solver_parameters::constraint_order::pi_sign_change: std::shuffle(R.begin(), R.end(), rng); pi_changed = solver.compute_update_row( x, R.begin(), R.end(), kappa, delta, theta); remaining = local_compute_violated_constraints(solver, x); if (!pi_changed && remaining == 0) return 0; return remaining; case solver_parameters::constraint_order::none: default: solver.compute_update_row( x, R.begin(), R.end(), kappa, delta, theta); return compute_violated_constraints(solver, x, R); } }
pushq %r15 pushq %r14 pushq %rbx subq $0x20, %rsp movl 0x30(%rdi), %eax movq %rdx, %rbx movq %rsi, %r15 movq %rdi, %r14 decl %eax cmpl $0x6, %eax ja 0x2e5723 leaq 0x17dc45(%rip), %rdx # 0x463250 movslq (%rdx,%rax,4), %rax addq %rdx, %rax jmpq *%rax movq 0x8(%r14), %rax leaq 0x18(%rsp), %rdx leaq 0x10(%rsp), %rcx movq %r15, %rdi movq %rbx, %rsi movq %rax, (%rdx) movq (%r14), %rax movq %rax, (%rcx) callq 0x2e5d74 movq %r15, %rdi movq %rbx, %rsi movq %r14, %rdx callq 0x2e5c6d addq $0x20, %rsp popq %rbx popq %r14 popq %r15 retq movq (%r14), %rdi movq 0x8(%r14), %rsi movq %r15, %rdx vmovss %xmm2, 0xc(%rsp) vmovss %xmm1, 0x8(%rsp) vmovss %xmm0, 0x4(%rsp) callq 0x2e7c08 jmp 0x2e5758 movq 0x18(%r14), %rdi movq 0x20(%r14), %rsi vmovss %xmm2, 0xc(%rsp) vmovss %xmm1, 0x8(%rsp) vmovss %xmm0, 0x4(%rsp) callq 0x2e7332 jmp 0x2e56b4 movq 0x18(%r14), %rdi movq 0x20(%r14), %rsi vmovss %xmm2, 0xc(%rsp) vmovss %xmm1, 0x8(%rsp) vmovss %xmm0, 0x4(%rsp) callq 0x2e779d vmovss 0x4(%rsp), %xmm0 vmovss 0x8(%rsp), %xmm1 vmovss 0xc(%rsp), %xmm2 movq 0x18(%r14), %rdx movq 0x20(%r14), %rcx movq %r15, %rdi movq %rbx, %rsi callq 0x2e6148 movq %r14, %rdi movq %r15, %rsi movq %rbx, %rdx addq $0x20, %rsp popq %rbx popq %r14 popq %r15 jmp 0x2e632c movq (%r14), %rdi movq 0x8(%r14), %rsi movq %rcx, %rdx vmovss %xmm2, 0xc(%rsp) vmovss %xmm1, 0x8(%rsp) vmovss %xmm0, 0x4(%rsp) callq 0x3ceed vmovss 0x4(%rsp), %xmm0 vmovss 0x8(%rsp), %xmm1 vmovss 0xc(%rsp), %xmm2 movq (%r14), %rdx movq 0x8(%r14), %rcx movq %r15, %rdi movq %rbx, %rsi callq 0x2e5f64 jmp 0x2e577c movq (%r14), %rdi movq 0x8(%r14), %rsi movq %r15, %rdx vmovss %xmm2, 0xc(%rsp) vmovss %xmm1, 0x8(%rsp) vmovss %xmm0, 0x4(%rsp) callq 0x2e80fa vmovss 0x4(%rsp), %xmm0 vmovss 0x8(%rsp), %xmm1 vmovss 0xc(%rsp), %xmm2 movq (%r14), %rdx movq 0x8(%r14), %rcx movq %r15, %rdi movq %rbx, %rsi callq 0x2e6448 movq %r15, %rdi movq %rbx, %rsi movq %r14, %rdx addq $0x20, %rsp popq %rbx popq %r14 popq %r15 jmp 0x2e5c6d movq (%r14), %rdi movq 0x8(%r14), %rsi movq %rcx, %rdx vmovss %xmm2, 0xc(%rsp) vmovss %xmm1, 0x8(%rsp) vmovss %xmm0, 0x4(%rsp) callq 0x3ceed vmovss 0x4(%rsp), %xmm0 vmovss 0x8(%rsp), %xmm1 vmovss 0xc(%rsp), %xmm2 movq (%r14), %rdx movq 0x8(%r14), %rcx movq %r15, %rdi movq %rbx, %rsi callq 0x2e5f64 movq %r14, %rdi movq %r15, %rsi movq %rbx, %rdx addq $0x20, %rsp popq %rbx popq %r14 popq %r15 jmp 0x2e662c nop
/quesnel[P]baryonyx/lib/src/itm-common.hpp
int baryonyx::itm::compute_order::push_and_run<baryonyx::itm::solver_inequalities_101coeff<float, baryonyx::itm::minimize_tag, baryonyx::itm::quadratic_cost_type<float>, true>, baryonyx::bit_array, float>(baryonyx::itm::solver_inequalities_101coeff<float, baryonyx::itm::minimize_tag, baryonyx::itm::quadratic_cost_type<float>, true>&, baryonyx::bit_array&, std::linear_congruential_engine<unsigned long, 16807ul, 0ul, 2147483647ul>&, float, float, float, float)
int push_and_run(Solver& solver, Xtype& x, random_engine& rng, Float kappa, Float delta, Float theta, Float objective_amplifier) { bool pi_changed = 0; int remaining = 0; if (use_cycle) order = next_state(order); switch (order) { case solver_parameters::constraint_order::reversing: solver.push_and_compute_update_row(x, R.crbegin(), R.crend(), kappa, delta, theta, objective_amplifier); return compute_violated_constraints(solver, x, R); case solver_parameters::constraint_order::random_sorting: std::shuffle(R.begin(), R.end(), rng); solver.push_and_compute_update_row( x, R.begin(), R.end(), kappa, delta, theta, objective_amplifier); return compute_violated_constraints(solver, x, R); case solver_parameters::constraint_order::infeasibility_decr: std::sort(m_order.begin(), m_order.end(), [](const auto& lhs, const auto& rhs) { return rhs.second < lhs.second; }); solver.push_and_compute_update_row(x, m_order.cbegin(), m_order.cend(), kappa, delta, theta, objective_amplifier); return infeasibility_local_compute_violated_constraints(solver, x); case solver_parameters::constraint_order::infeasibility_incr: std::sort(m_order.begin(), m_order.end(), [](const auto& lhs, const auto& rhs) { return lhs.second < rhs.second; }); solver.push_and_compute_update_row(x, m_order.cbegin(), m_order.cend(), kappa, delta, theta, objective_amplifier); return infeasibility_local_compute_violated_constraints(solver, x); case solver_parameters::constraint_order::lagrangian_decr: std::sort( R.begin(), R.end(), [&solver](const auto lhs, const auto rhs) { return solver.pi[rhs] < solver.pi[lhs]; }); solver.push_and_compute_update_row(x, R.cbegin(), R.cend(), kappa, delta, theta, objective_amplifier); return compute_violated_constraints(solver, x, R); case solver_parameters::constraint_order::lagrangian_incr: std::sort( R.begin(), R.end(), [&solver](const auto lhs, const auto rhs) { return solver.pi[lhs] < solver.pi[rhs]; }); solver.push_and_compute_update_row(x, R.cbegin(), R.cend(), kappa, delta, theta, objective_amplifier); return compute_violated_constraints(solver, x, R); case solver_parameters::constraint_order::pi_sign_change: std::shuffle(R.begin(), R.end(), rng); pi_changed = solver.push_and_compute_update_row( x, R.begin(), R.end(), kappa, delta, theta, objective_amplifier); remaining = local_compute_violated_constraints(solver, x); if (!pi_changed && remaining == 0) return 0; return remaining; case solver_parameters::constraint_order::none: default: solver.push_and_compute_update_row(x, R.cbegin(), R.cend(), kappa, delta, theta, objective_amplifier); return compute_violated_constraints(solver, x, R); } }
pushq %r15 pushq %r14 pushq %rbx subq $0x20, %rsp movl 0x30(%rdi), %eax cmpb $0x1, 0x34(%rdi) movq %rdx, %rbx movq %rsi, %r15 movq %rdi, %r14 jne 0x2e581c xorl %esi, %esi leal 0x1(%rax), %edx cmpl $0x7, %eax cmovll %edx, %esi movl %esi, %eax movl %esi, 0x30(%r14) decl %eax cmpl $0x6, %eax ja 0x2e59b5 leaq 0x17da3e(%rip), %rdx # 0x46326c movslq (%rdx,%rax,4), %rax addq %rdx, %rax jmpq *%rax movq 0x8(%r14), %rax leaq 0x18(%rsp), %rdx leaq 0x10(%rsp), %rcx movq %r15, %rdi movq %rbx, %rsi movq %rax, (%rdx) movq (%r14), %rax movq %rax, (%rcx) callq 0x2e85f0 movq %r15, %rdi movq %rbx, %rsi movq %r14, %rdx callq 0x2e5c6d addq $0x20, %rsp popq %rbx popq %r14 popq %r15 retq movq (%r14), %rdi movq 0x8(%r14), %rsi movq %r15, %rdx vmovss %xmm3, 0xc(%rsp) vmovss %xmm2, 0x8(%rsp) vmovss %xmm1, 0x4(%rsp) vmovss %xmm0, (%rsp) callq 0x2e9853 jmp 0x2e599e movq 0x18(%r14), %rdi movq 0x20(%r14), %rsi vmovss %xmm3, 0xc(%rsp) vmovss %xmm2, 0x8(%rsp) vmovss %xmm1, 0x4(%rsp) vmovss %xmm0, (%rsp) callq 0x2e8f7d jmp 0x2e58e6 movq 0x18(%r14), %rdi movq 0x20(%r14), %rsi vmovss %xmm3, 0xc(%rsp) vmovss %xmm2, 0x8(%rsp) vmovss %xmm1, 0x4(%rsp) vmovss %xmm0, (%rsp) callq 0x2e93e8 vmovss (%rsp), %xmm0 vmovss 0x4(%rsp), %xmm1 vmovss 0x8(%rsp), %xmm2 vmovss 0xc(%rsp), %xmm3 movq 0x18(%r14), %rdx movq 0x20(%r14), %rcx movq %r15, %rdi movq %rbx, %rsi callq 0x2e8abe movq %r14, %rdi movq %r15, %rsi movq %rbx, %rdx addq $0x20, %rsp popq %rbx popq %r14 popq %r15 jmp 0x2e632c movq (%r14), %rdi movq 0x8(%r14), %rsi movq %rcx, %rdx vmovss %xmm3, 0xc(%rsp) vmovss %xmm2, 0x8(%rsp) vmovss %xmm1, 0x4(%rsp) vmovss %xmm0, (%rsp) callq 0x3ceed vmovss (%rsp), %xmm0 vmovss 0x4(%rsp), %xmm1 vmovss 0x8(%rsp), %xmm2 vmovss 0xc(%rsp), %xmm3 movq (%r14), %rdx movq 0x8(%r14), %rcx movq %r15, %rdi movq %rbx, %rsi callq 0x2e885e jmp 0x2e59c7 movq (%r14), %rdi movq 0x8(%r14), %rsi movq %r15, %rdx vmovss %xmm3, 0xc(%rsp) vmovss %xmm2, 0x8(%rsp) vmovss %xmm1, 0x4(%rsp) vmovss %xmm0, (%rsp) callq 0x2e9d45 vmovss (%rsp), %xmm0 vmovss 0x4(%rsp), %xmm1 vmovss 0x8(%rsp), %xmm2 vmovss 0xc(%rsp), %xmm3 movq (%r14), %rdx movq 0x8(%r14), %rcx movq %r15, %rdi movq %rbx, %rsi callq 0x2e8d1e movq %r15, %rdi movq %rbx, %rsi movq %r14, %rdx addq $0x20, %rsp popq %rbx popq %r14 popq %r15 jmp 0x2e5c6d movq (%r14), %rdi movq 0x8(%r14), %rsi movq %rcx, %rdx vmovss %xmm3, 0xc(%rsp) vmovss %xmm2, 0x8(%rsp) vmovss %xmm1, 0x4(%rsp) vmovss %xmm0, (%rsp) callq 0x3ceed vmovss (%rsp), %xmm0 vmovss 0x4(%rsp), %xmm1 vmovss 0x8(%rsp), %xmm2 vmovss 0xc(%rsp), %xmm3 movq (%r14), %rdx movq 0x8(%r14), %rcx movq %r15, %rdi movq %rbx, %rsi callq 0x2e885e movq %r14, %rdi movq %r15, %rsi movq %rbx, %rdx addq $0x20, %rsp popq %rbx popq %r14 popq %r15 jmp 0x2e662c
/quesnel[P]baryonyx/lib/src/itm-common.hpp
int baryonyx::itm::compute_order::infeasibility_local_compute_violated_constraints<baryonyx::itm::solver_inequalities_101coeff<float, baryonyx::itm::minimize_tag, baryonyx::itm::quadratic_cost_type<float>, true> const, baryonyx::bit_array>(baryonyx::itm::solver_inequalities_101coeff<float, baryonyx::itm::minimize_tag, baryonyx::itm::quadratic_cost_type<float>, true> const&, baryonyx::bit_array const&)
int infeasibility_local_compute_violated_constraints(Solver& solver, const Xtype& x) { m_order.clear(); for (int k = 0, e = solver.m; k != e; ++k) { sparse_matrix<int>::const_row_iterator it, et; std::tie(it, et) = solver.ap.row(k); int v = 0; for (; it != et; ++it) v += solver.factor(it->value) * x[it->column]; if (solver.bound_min(k) > v) m_order.push_back(std::make_pair(k, solver.bound_min(k) - v)); else if (solver.bound_max(k) < v) m_order.push_back(std::make_pair(k, v - solver.bound_max(k))); } return length(m_order); }
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x48, %rsp movq 0x18(%rdi), %rax movq %rdx, %r14 movq %rsi, 0x20(%rsp) cmpq %rax, 0x20(%rdi) je 0x2e5b76 movq %rax, 0x20(%rdi) movq 0x20(%rsp), %rsi leaq 0x18(%rdi), %rax movq %rdi, 0x28(%rsp) xorl %ebx, %ebx movq %rax, 0x18(%rsp) movl 0x88(%rsi), %ecx leaq 0x10(%rsi), %rdx leaq 0x58(%rsi), %rbp movq %rdx, 0x30(%rsp) movq %rcx, 0x38(%rsp) cmpq 0x38(%rsp), %rbx je 0x2e5c4d movq 0x30(%rsp), %rsi leaq 0x8(%rsp), %rdi movl %ebx, %edx movq %rbx, 0x40(%rsp) callq 0x3c868 movq 0x8(%rsp), %r15 movq 0x10(%rsp), %rbx xorl %r13d, %r13d cmpq %r15, %rbx je 0x2e5bfb movl (%rbx), %esi movq %rbp, %rdi callq 0x2b9c1a movl 0x4(%rbx), %esi movq %r14, %rdi movl %eax, %r12d callq 0x3664a imull %r12d, %eax addq $0x8, %rbx addl %eax, %r13d jmp 0x2e5bd1 movq 0x20(%rsp), %rax movq 0x40(%rsp), %rbx movq 0x70(%rax), %rcx imulq $0xc, %rbx, %rdx movl (%rcx,%rdx), %eax subl %r13d, %eax jle 0x2e5c23 shlq $0x20, %rax orq %rbx, %rax movq %rax, 0x8(%rsp) jmp 0x2e5c36 subl 0x4(%rcx,%rdx), %r13d jle 0x2e5c45 shlq $0x20, %r13 orq %rbx, %r13 movq %r13, 0x8(%rsp) movq 0x18(%rsp), %rdi leaq 0x8(%rsp), %rsi callq 0x3c8c2 incq %rbx jmp 0x2e5ba3 movq 0x28(%rsp), %rcx movq 0x20(%rcx), %rax subq 0x18(%rcx), %rax shrq $0x3, %rax addq $0x48, %rsp popq %rbx popq %r12 popq %r13 popq %r14 popq %r15 popq %rbp retq
/quesnel[P]baryonyx/lib/src/itm-common.hpp
int baryonyx::itm::compute_violated_constraints<baryonyx::itm::solver_inequalities_101coeff<float, baryonyx::itm::minimize_tag, baryonyx::itm::quadratic_cost_type<float>, true>, baryonyx::bit_array>(baryonyx::itm::solver_inequalities_101coeff<float, baryonyx::itm::minimize_tag, baryonyx::itm::quadratic_cost_type<float>, true> const&, baryonyx::bit_array const&, std::vector<int, std::allocator<int>>&)
int compute_violated_constraints(const Solver& slv, const Xtype& x, std::vector<int>& out) { out.clear(); for (int k = 0; k != slv.m; ++k) if (!is_valid_constraint(slv, k, x)) out.emplace_back(k); return length(out); }
pushq %r15 pushq %r14 pushq %r12 pushq %rbx pushq %rax movq (%rdx), %rax movq %rdx, %rbx movq %rsi, %r14 movq %rdi, %r15 cmpq %rax, 0x8(%rdx) je 0x2e5c8b movq %rax, 0x8(%rbx) leaq 0x4(%rsp), %r12 xorl %esi, %esi movl %esi, 0x4(%rsp) cmpl 0x88(%r15), %esi je 0x2e5cc1 movq %r15, %rdi movq %r14, %rdx callq 0x2e5cd8 testb %al, %al jne 0x2e5cb9 movq %rbx, %rdi movq %r12, %rsi callq 0x31460 movl 0x4(%rsp), %esi incl %esi jmp 0x2e5c92 movq 0x8(%rbx), %rax subq (%rbx), %rax shrq $0x2, %rax addq $0x8, %rsp popq %rbx popq %r12 popq %r14 popq %r15 retq
/quesnel[P]baryonyx/lib/src/itm-common.hpp
bool baryonyx::itm::is_valid_constraint<baryonyx::itm::solver_inequalities_101coeff<float, baryonyx::itm::minimize_tag, baryonyx::itm::quadratic_cost_type<float>, true>, baryonyx::bit_array>(baryonyx::itm::solver_inequalities_101coeff<float, baryonyx::itm::minimize_tag, baryonyx::itm::quadratic_cost_type<float>, true> const&, int, baryonyx::bit_array const&)
bool is_valid_constraint(const Solver& slv, int k, const Xtype& x) { typename sparse_matrix<int>::const_row_iterator it, et; std::tie(it, et) = slv.ap.row(k); int v = 0; for (; it != et; ++it) v += slv.factor(it->value) * x[it->column]; return slv.bound_min(k) <= v && v <= slv.bound_max(k); }
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x28, %rsp movq %rdx, %r14 movl %esi, %edx leaq 0x10(%rdi), %rsi leaq 0x18(%rsp), %r15 movq %rdi, %r12 movl %edx, 0xc(%rsp) movq %r15, %rdi callq 0x3c868 movq (%r15), %rbp movq 0x8(%r15), %rbx leaq 0x58(%r12), %r15 xorl %r13d, %r13d movq %r12, 0x10(%rsp) cmpq %rbp, %rbx je 0x2e5d41 movl (%rbx), %esi movq %r15, %rdi callq 0x2b9c1a movl 0x4(%rbx), %esi movq %r14, %rdi movl %eax, %r12d callq 0x3664a imull %r12d, %eax addq $0x8, %rbx addl %eax, %r13d jmp 0x2e5d17 movq 0x10(%rsp), %rax movslq 0xc(%rsp), %rcx movq 0x70(%rax), %rax imulq $0xc, %rcx, %rcx cmpl %r13d, (%rax,%rcx) jle 0x2e5d5d xorl %eax, %eax jmp 0x2e5d65 cmpl 0x4(%rax,%rcx), %r13d setle %al addq $0x28, %rsp popq %rbx popq %r12 popq %r13 popq %r14 popq %r15 popq %rbp retq
/quesnel[P]baryonyx/lib/src/itm-common.hpp
bool baryonyx::itm::solver_inequalities_101coeff<float, baryonyx::itm::minimize_tag, baryonyx::itm::quadratic_cost_type<float>, true>::compute_update_row<baryonyx::bit_array, std::reverse_iterator<__gnu_cxx::__normal_iterator<int const*, std::vector<int, std::allocator<int>>>>>(baryonyx::bit_array&, std::reverse_iterator<__gnu_cxx::__normal_iterator<int const*, std::vector<int, std::allocator<int>>>>, std::reverse_iterator<__gnu_cxx::__normal_iterator<int const*, std::vector<int, std::allocator<int>>>>, float, float, float)
bool compute_update_row(Xtype& x, Iterator first, Iterator last, Float kappa, Float delta, Float theta) { auto at_least_one_pi_changed{ false }; logger::log("update-row {} {} {}\n", kappa, delta, theta); for (; first != last; ++first) { auto k = constraint(first); const auto it = ap.row(k); decrease_preference(std::get<0>(it), std::get<1>(it), theta); const auto sizes = compute_reduced_costs(std::get<0>(it), std::get<1>(it), x); calculator_sort<Mode>(R.get(), R.get() + sizes.r_size, rng); int selected = select_variables(sizes, b[k].min, b[k].max); logger::log("constraints {}: {} <= ", k, b[k].min); for (int i = 0; i < sizes.r_size; ++i) logger::log("{} ({} {}) ", R[i].f, R[i].value, R[i].id); logger::log("<= {} => Selected: {}\n", b[k].max, selected); auto pi_change = affect(*this, x, std::get<0>(it), k, selected, sizes.r_size, kappa, delta); at_least_one_pi_changed = at_least_one_pi_changed || pi_change; } return at_least_one_pi_changed; }
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x58, %rsp movq %rcx, 0x48(%rsp) movq %rdx, %rbx leaq 0x1c(%rsp), %rcx leaq 0x18(%rsp), %r8 leaq 0x14(%rsp), %r9 leaq 0x1765a4(%rip), %rdx # 0x45c344 movq %rsi, %r14 movq %rdi, %r12 vmovss %xmm0, (%rcx) vmovss %xmm1, (%r8) vmovss %xmm2, (%r9) pushq $0x14 popq %rsi callq 0xb907e leaq 0x10(%r12), %rax movq %rbx, 0x40(%rsp) xorl %edx, %edx movq %rax, 0x38(%rsp) movq (%rbx), %rax pushq $0x16 popq %r15 leaq 0x17d960(%rip), %rbp # 0x46373b pushq $0xb popq %rbx movq %r14, 0x30(%rsp) movq 0x48(%rsp), %rcx cmpq (%rcx), %rax je 0x2e5f4f movq %rdx, 0x50(%rsp) movl -0x4(%rax), %edx movq 0x38(%rsp), %rsi leaq 0x20(%rsp), %rdi movl %edx, (%rsp) callq 0x3de66 vmovss 0x14(%rsp), %xmm0 movq 0x20(%rsp), %rdx movq 0x28(%rsp), %rsi movq %r12, %rdi callq 0x2e666e movq 0x20(%rsp), %rdx movq 0x28(%rsp), %rsi movq %r12, %rdi movq %r14, %rcx callq 0x2e668c movq %rax, 0x8(%rsp) cltq movq 0x68(%r12), %rdi movq 0x8(%r12), %rdx imulq $0xc, %rax, %rsi addq %rdi, %rsi callq 0x2e67e5 movslq (%rsp), %rax movq 0x70(%r12), %rcx leaq 0x8(%rsp), %rsi movq %r12, %rdi imulq $0xc, %rax, %rax movl (%rcx,%rax), %edx movl 0x4(%rcx,%rax), %ecx callq 0x2e6818 movslq (%rsp), %rcx movl %eax, 0x4(%rsp) leaq 0x17b1c0(%rip), %rdx # 0x461045 movq %r12, %rdi movq %r15, %r13 movq %r15, %rsi imulq $0xc, %rcx, %r8 addq 0x70(%r12), %r8 movq %rsp, %rcx callq 0xb91c2 pushq $0x8 popq %r15 xorl %r14d, %r14d movslq 0x8(%rsp), %rax cmpq %rax, %r14 jge 0x2e5eda movq 0x68(%r12), %rax movq %r12, %rdi movq %rbx, %rsi movq %rbp, %rdx leaq -0x8(%rax,%r15), %r8 leaq -0x4(%rax,%r15), %rcx leaq (%rax,%r15), %r9 callq 0x2ba7d2 incq %r14 addq $0xc, %r15 jmp 0x2e5ea6 movslq (%rsp), %rax movq 0x70(%r12), %rcx leaq 0x17b172(%rip), %rdx # 0x46105c leaq 0x4(%rsp), %r8 movq %r12, %rdi movq %r13, %r15 movq %r13, %rsi imulq $0xc, %rax, %rax leaq 0x4(%rcx,%rax), %rcx callq 0xb91c2 vmovss 0x1c(%rsp), %xmm0 vmovss 0x18(%rsp), %xmm1 movq 0x30(%rsp), %r14 movq 0x28(%rsp), %rdx movl (%rsp), %ecx movl 0x4(%rsp), %r8d movl 0x8(%rsp), %r9d movq %r12, %rdi movq %r14, %rsi callq 0x2e6889 movq 0x50(%rsp), %rdx movq 0x40(%rsp), %rcx orb %al, %dl movq (%rcx), %rax addq $-0x4, %rax movq %rax, (%rcx) jmp 0x2e5de3 andb $0x1, %dl movl %edx, %eax addq $0x58, %rsp popq %rbx popq %r12 popq %r13 popq %r14 popq %r15 popq %rbp retq nop
/quesnel[P]baryonyx/lib/src/itm-solver-inequalities-101.cpp
bool baryonyx::itm::solver_inequalities_101coeff<float, baryonyx::itm::minimize_tag, baryonyx::itm::quadratic_cost_type<float>, true>::compute_update_row<baryonyx::bit_array, __gnu_cxx::__normal_iterator<std::pair<int, int> const*, std::vector<std::pair<int, int>, std::allocator<std::pair<int, int>>>>>(baryonyx::bit_array&, __gnu_cxx::__normal_iterator<std::pair<int, int> const*, std::vector<std::pair<int, int>, std::allocator<std::pair<int, int>>>>, __gnu_cxx::__normal_iterator<std::pair<int, int> const*, std::vector<std::pair<int, int>, std::allocator<std::pair<int, int>>>>, float, float, float)
bool compute_update_row(Xtype& x, Iterator first, Iterator last, Float kappa, Float delta, Float theta) { auto at_least_one_pi_changed{ false }; logger::log("update-row {} {} {}\n", kappa, delta, theta); for (; first != last; ++first) { auto k = constraint(first); const auto it = ap.row(k); decrease_preference(std::get<0>(it), std::get<1>(it), theta); const auto sizes = compute_reduced_costs(std::get<0>(it), std::get<1>(it), x); calculator_sort<Mode>(R.get(), R.get() + sizes.r_size, rng); int selected = select_variables(sizes, b[k].min, b[k].max); logger::log("constraints {}: {} <= ", k, b[k].min); for (int i = 0; i < sizes.r_size; ++i) logger::log("{} ({} {}) ", R[i].f, R[i].value, R[i].id); logger::log("<= {} => Selected: {}\n", b[k].max, selected); auto pi_change = affect(*this, x, std::get<0>(it), k, selected, sizes.r_size, kappa, delta); at_least_one_pi_changed = at_least_one_pi_changed || pi_change; } return at_least_one_pi_changed; }
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x58, %rsp movq %rcx, 0x40(%rsp) leaq 0x1c(%rsp), %rcx leaq 0x18(%rsp), %r8 leaq 0x14(%rsp), %r9 movq %rdx, %r14 movq %rsi, %r15 movq %rdi, %r12 leaq 0x1761ca(%rip), %rdx # 0x45c344 vmovss %xmm0, (%rcx) vmovss %xmm1, (%r8) vmovss %xmm2, (%r9) pushq $0x14 popq %rsi callq 0xb907e leaq 0x10(%r12), %rax movq %rax, 0x30(%rsp) xorl %eax, %eax pushq $0x16 popq %r13 leaq 0x17d594(%rip), %rbp # 0x46373b pushq $0xb popq %rbx movq %r15, 0x38(%rsp) cmpq 0x40(%rsp), %r14 je 0x2e631a movl (%r14), %edx movq 0x30(%rsp), %rsi leaq 0x20(%rsp), %rdi movq %rax, 0x50(%rsp) movq %r14, 0x48(%rsp) movl %edx, (%rsp) callq 0x3de66 vmovss 0x14(%rsp), %xmm0 movq 0x20(%rsp), %rdx movq 0x28(%rsp), %rsi movq %r12, %rdi callq 0x2e666e movq 0x20(%rsp), %rdx movq 0x28(%rsp), %rsi movq %r12, %rdi movq %r15, %rcx callq 0x2e668c movq %rax, 0x8(%rsp) cltq movq 0x68(%r12), %rdi movq 0x8(%r12), %rdx imulq $0xc, %rax, %rsi addq %rdi, %rsi callq 0x2e67e5 movslq (%rsp), %rax movq 0x70(%r12), %rcx leaq 0x8(%rsp), %rsi movq %r12, %rdi imulq $0xc, %rax, %rax movl (%rcx,%rax), %edx movl 0x4(%rcx,%rax), %ecx callq 0x2e6818 movslq (%rsp), %rcx movl %eax, 0x4(%rsp) leaq 0x17adf2(%rip), %rdx # 0x461045 movq %r12, %rdi movq %r13, %r14 movq %r13, %rsi imulq $0xc, %rcx, %r8 addq 0x70(%r12), %r8 movq %rsp, %rcx callq 0xb91c2 pushq $0x8 popq %r15 xorl %r13d, %r13d movslq 0x8(%rsp), %rax cmpq %rax, %r13 jge 0x2e62a8 movq 0x68(%r12), %rax movq %r12, %rdi movq %rbx, %rsi movq %rbp, %rdx leaq -0x8(%rax,%r15), %r8 leaq -0x4(%rax,%r15), %rcx leaq (%rax,%r15), %r9 callq 0x2ba7d2 incq %r13 addq $0xc, %r15 jmp 0x2e6274 movslq (%rsp), %rax movq 0x70(%r12), %rcx leaq 0x17ada4(%rip), %rdx # 0x46105c leaq 0x4(%rsp), %r8 movq %r12, %rdi movq %r14, %r13 movq %r14, %rsi imulq $0xc, %rax, %rax leaq 0x4(%rcx,%rax), %rcx callq 0xb91c2 vmovss 0x1c(%rsp), %xmm0 vmovss 0x18(%rsp), %xmm1 movq 0x38(%rsp), %r15 movq 0x28(%rsp), %rdx movl (%rsp), %ecx movl 0x4(%rsp), %r8d movl 0x8(%rsp), %r9d movq %r12, %rdi movq %r15, %rsi callq 0x2e6889 movq 0x50(%rsp), %rcx movq 0x48(%rsp), %r14 orb %al, %cl addq $0x8, %r14 movq %rcx, %rax jmp 0x2e61af andb $0x1, %al addq $0x58, %rsp popq %rbx popq %r12 popq %r13 popq %r14 popq %r15 popq %rbp retq nop
/quesnel[P]baryonyx/lib/src/itm-solver-inequalities-101.cpp
bool baryonyx::itm::solver_inequalities_101coeff<float, baryonyx::itm::minimize_tag, baryonyx::itm::quadratic_cost_type<float>, true>::compute_update_row<baryonyx::bit_array, __gnu_cxx::__normal_iterator<int const*, std::vector<int, std::allocator<int>>>>(baryonyx::bit_array&, __gnu_cxx::__normal_iterator<int const*, std::vector<int, std::allocator<int>>>, __gnu_cxx::__normal_iterator<int const*, std::vector<int, std::allocator<int>>>, float, float, float)
bool compute_update_row(Xtype& x, Iterator first, Iterator last, Float kappa, Float delta, Float theta) { auto at_least_one_pi_changed{ false }; logger::log("update-row {} {} {}\n", kappa, delta, theta); for (; first != last; ++first) { auto k = constraint(first); const auto it = ap.row(k); decrease_preference(std::get<0>(it), std::get<1>(it), theta); const auto sizes = compute_reduced_costs(std::get<0>(it), std::get<1>(it), x); calculator_sort<Mode>(R.get(), R.get() + sizes.r_size, rng); int selected = select_variables(sizes, b[k].min, b[k].max); logger::log("constraints {}: {} <= ", k, b[k].min); for (int i = 0; i < sizes.r_size; ++i) logger::log("{} ({} {}) ", R[i].f, R[i].value, R[i].id); logger::log("<= {} => Selected: {}\n", b[k].max, selected); auto pi_change = affect(*this, x, std::get<0>(it), k, selected, sizes.r_size, kappa, delta); at_least_one_pi_changed = at_least_one_pi_changed || pi_change; } return at_least_one_pi_changed; }
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x58, %rsp movq %rcx, 0x40(%rsp) leaq 0x1c(%rsp), %rcx leaq 0x18(%rsp), %r8 leaq 0x14(%rsp), %r9 movq %rdx, %r14 movq %rsi, %r15 movq %rdi, %r12 leaq 0x175eca(%rip), %rdx # 0x45c344 vmovss %xmm0, (%rcx) vmovss %xmm1, (%r8) vmovss %xmm2, (%r9) pushq $0x14 popq %rsi callq 0xb907e leaq 0x10(%r12), %rax movq %rax, 0x30(%rsp) xorl %eax, %eax pushq $0x16 popq %r13 leaq 0x17d294(%rip), %rbp # 0x46373b pushq $0xb popq %rbx movq %r15, 0x38(%rsp) cmpq 0x40(%rsp), %r14 je 0x2e661a movl (%r14), %edx movq 0x30(%rsp), %rsi leaq 0x20(%rsp), %rdi movq %rax, 0x50(%rsp) movq %r14, 0x48(%rsp) movl %edx, (%rsp) callq 0x3de66 vmovss 0x14(%rsp), %xmm0 movq 0x20(%rsp), %rdx movq 0x28(%rsp), %rsi movq %r12, %rdi callq 0x2e666e movq 0x20(%rsp), %rdx movq 0x28(%rsp), %rsi movq %r12, %rdi movq %r15, %rcx callq 0x2e668c movq %rax, 0x8(%rsp) cltq movq 0x68(%r12), %rdi movq 0x8(%r12), %rdx imulq $0xc, %rax, %rsi addq %rdi, %rsi callq 0x2e67e5 movslq (%rsp), %rax movq 0x70(%r12), %rcx leaq 0x8(%rsp), %rsi movq %r12, %rdi imulq $0xc, %rax, %rax movl (%rcx,%rax), %edx movl 0x4(%rcx,%rax), %ecx callq 0x2e6818 movslq (%rsp), %rcx movl %eax, 0x4(%rsp) leaq 0x17aaf2(%rip), %rdx # 0x461045 movq %r12, %rdi movq %r13, %r14 movq %r13, %rsi imulq $0xc, %rcx, %r8 addq 0x70(%r12), %r8 movq %rsp, %rcx callq 0xb91c2 pushq $0x8 popq %r15 xorl %r13d, %r13d movslq 0x8(%rsp), %rax cmpq %rax, %r13 jge 0x2e65a8 movq 0x68(%r12), %rax movq %r12, %rdi movq %rbx, %rsi movq %rbp, %rdx leaq -0x8(%rax,%r15), %r8 leaq -0x4(%rax,%r15), %rcx leaq (%rax,%r15), %r9 callq 0x2ba7d2 incq %r13 addq $0xc, %r15 jmp 0x2e6574 movslq (%rsp), %rax movq 0x70(%r12), %rcx leaq 0x17aaa4(%rip), %rdx # 0x46105c leaq 0x4(%rsp), %r8 movq %r12, %rdi movq %r14, %r13 movq %r14, %rsi imulq $0xc, %rax, %rax leaq 0x4(%rcx,%rax), %rcx callq 0xb91c2 vmovss 0x1c(%rsp), %xmm0 vmovss 0x18(%rsp), %xmm1 movq 0x38(%rsp), %r15 movq 0x28(%rsp), %rdx movl (%rsp), %ecx movl 0x4(%rsp), %r8d movl 0x8(%rsp), %r9d movq %r12, %rdi movq %r15, %rsi callq 0x2e6889 movq 0x50(%rsp), %rcx movq 0x48(%rsp), %r14 orb %al, %cl addq $0x4, %r14 movq %rcx, %rax jmp 0x2e64af andb $0x1, %al addq $0x58, %rsp popq %rbx popq %r12 popq %r13 popq %r14 popq %r15 popq %rbp retq nop
/quesnel[P]baryonyx/lib/src/itm-solver-inequalities-101.cpp