9 #include "snapshot-interface.h"
11 #include "clockvector.h"
12 #include "cyclegraph.h"
15 #include "threads-model.h"
17 #define INITIAL_THREAD_ID 0
22 bug_message(const char *str) {
23 const char *fmt = " [BUG] %s\n";
24 msg = (char *)snapshot_malloc(strlen(fmt) + strlen(str));
25 sprintf(msg, fmt, str);
27 ~bug_message() { if (msg) snapshot_free(msg); }
30 void print() { printf("%s", msg); }
34 * Structure for holding small ModelChecker members that should be snapshotted
36 struct model_snapshot_members {
37 ModelAction *current_action;
38 unsigned int next_thread_id;
39 modelclock_t used_sequence_numbers;
41 ModelAction *next_backtrack;
42 std::vector< bug_message *, SnapshotAlloc<bug_message *> > bugs;
45 /** @brief Constructor */
46 ModelChecker::ModelChecker(struct model_params params) :
47 /* Initialize default scheduler */
49 scheduler(new Scheduler()),
51 num_feasible_executions(0),
53 earliest_diverge(NULL),
54 action_trace(new action_list_t()),
55 thread_map(new HashTable<int, Thread *, int>()),
56 obj_map(new HashTable<const void *, action_list_t *, uintptr_t, 4>()),
57 lock_waiters_map(new HashTable<const void *, action_list_t *, uintptr_t, 4>()),
58 condvar_waiters_map(new HashTable<const void *, action_list_t *, uintptr_t, 4>()),
59 obj_thrd_map(new HashTable<void *, std::vector<action_list_t> *, uintptr_t, 4 >()),
60 promises(new std::vector< Promise *, SnapshotAlloc<Promise *> >()),
61 futurevalues(new std::vector< struct PendingFutureValue, SnapshotAlloc<struct PendingFutureValue> >()),
62 pending_rel_seqs(new std::vector< struct release_seq *, SnapshotAlloc<struct release_seq *> >()),
63 thrd_last_action(new std::vector< ModelAction *, SnapshotAlloc<ModelAction *> >(1)),
64 node_stack(new NodeStack()),
65 mo_graph(new CycleGraph()),
66 failed_promise(false),
67 too_many_reads(false),
69 bad_synchronization(false)
71 /* Allocate this "size" on the snapshotting heap */
72 priv = (struct model_snapshot_members *)snapshot_calloc(1, sizeof(*priv));
73 /* First thread created will have id INITIAL_THREAD_ID */
74 priv->next_thread_id = INITIAL_THREAD_ID;
76 /* Initialize a model-checker thread, for special ModelActions */
77 model_thread = new Thread(get_next_id());
78 thread_map->put(id_to_int(model_thread->get_id()), model_thread);
81 /** @brief Destructor */
82 ModelChecker::~ModelChecker()
84 for (unsigned int i = 0; i < get_num_threads(); i++)
85 delete thread_map->get(i);
90 delete lock_waiters_map;
91 delete condvar_waiters_map;
94 for (unsigned int i = 0; i < promises->size(); i++)
95 delete (*promises)[i];
98 delete pending_rel_seqs;
100 delete thrd_last_action;
105 for (unsigned int i = 0; i < priv->bugs.size(); i++)
106 delete priv->bugs[i];
111 static action_list_t * get_safe_ptr_action(HashTable<const void *, action_list_t *, uintptr_t, 4> * hash, void * ptr) {
112 action_list_t * tmp=hash->get(ptr);
114 tmp=new action_list_t();
120 static std::vector<action_list_t> * get_safe_ptr_vect_action(HashTable<void *, std::vector<action_list_t> *, uintptr_t, 4> * hash, void * ptr) {
121 std::vector<action_list_t> * tmp=hash->get(ptr);
123 tmp=new std::vector<action_list_t>();
130 * Restores user program to initial state and resets all model-checker data
133 void ModelChecker::reset_to_initial_state()
135 DEBUG("+++ Resetting to initial state +++\n");
136 node_stack->reset_execution();
137 failed_promise = false;
138 too_many_reads = false;
139 bad_synchronization = false;
141 snapshotObject->backTrackBeforeStep(0);
144 /** @return a thread ID for a new Thread */
145 thread_id_t ModelChecker::get_next_id()
147 return priv->next_thread_id++;
150 /** @return the number of user threads created during this execution */
151 unsigned int ModelChecker::get_num_threads() const
153 return priv->next_thread_id;
156 /** @return The currently executing Thread. */
157 Thread * ModelChecker::get_current_thread()
159 return scheduler->get_current_thread();
162 /** @return a sequence number for a new ModelAction */
163 modelclock_t ModelChecker::get_next_seq_num()
165 return ++priv->used_sequence_numbers;
168 Node * ModelChecker::get_curr_node() {
169 return node_stack->get_head();
173 * @brief Choose the next thread to execute.
175 * This function chooses the next thread that should execute. It can force the
176 * adjacency of read/write portions of a RMW action, force THREAD_CREATE to be
177 * followed by a THREAD_START, or it can enforce execution replay/backtracking.
178 * The model-checker may have no preference regarding the next thread (i.e.,
179 * when exploring a new execution ordering), in which case this will return
181 * @param curr The current ModelAction. This action might guide the choice of
183 * @return The next thread to run. If the model-checker has no preference, NULL.
185 Thread * ModelChecker::get_next_thread(ModelAction *curr)
190 /* Do not split atomic actions. */
192 return thread_current();
193 /* The THREAD_CREATE action points to the created Thread */
194 else if (curr->get_type() == THREAD_CREATE)
195 return (Thread *)curr->get_location();
198 /* Have we completed exploring the preselected path? */
202 /* Else, we are trying to replay an execution */
203 ModelAction *next = node_stack->get_next()->get_action();
205 if (next == diverge) {
206 if (earliest_diverge == NULL || *diverge < *earliest_diverge)
207 earliest_diverge=diverge;
209 Node *nextnode = next->get_node();
210 Node *prevnode = nextnode->get_parent();
211 scheduler->update_sleep_set(prevnode);
213 /* Reached divergence point */
214 if (nextnode->increment_misc()) {
215 /* The next node will try to satisfy a different misc_index values. */
216 tid = next->get_tid();
217 node_stack->pop_restofstack(2);
218 } else if (nextnode->increment_promise()) {
219 /* The next node will try to satisfy a different set of promises. */
220 tid = next->get_tid();
221 node_stack->pop_restofstack(2);
222 } else if (nextnode->increment_read_from()) {
223 /* The next node will read from a different value. */
224 tid = next->get_tid();
225 node_stack->pop_restofstack(2);
226 } else if (nextnode->increment_future_value()) {
227 /* The next node will try to read from a different future value. */
228 tid = next->get_tid();
229 node_stack->pop_restofstack(2);
230 } else if (nextnode->increment_relseq_break()) {
231 /* The next node will try to resolve a release sequence differently */
232 tid = next->get_tid();
233 node_stack->pop_restofstack(2);
235 /* Make a different thread execute for next step */
236 scheduler->add_sleep(thread_map->get(id_to_int(next->get_tid())));
237 tid = prevnode->get_next_backtrack();
238 /* Make sure the backtracked thread isn't sleeping. */
239 node_stack->pop_restofstack(1);
240 if (diverge==earliest_diverge) {
241 earliest_diverge=prevnode->get_action();
244 /* The correct sleep set is in the parent node. */
247 DEBUG("*** Divergence point ***\n");
251 tid = next->get_tid();
253 DEBUG("*** ModelChecker chose next thread = %d ***\n", id_to_int(tid));
254 ASSERT(tid != THREAD_ID_T_NONE);
255 return thread_map->get(id_to_int(tid));
259 * We need to know what the next actions of all threads in the sleep
260 * set will be. This method computes them and stores the actions at
261 * the corresponding thread object's pending action.
264 void ModelChecker::execute_sleep_set() {
265 for(unsigned int i=0;i<get_num_threads();i++) {
266 thread_id_t tid=int_to_id(i);
267 Thread *thr=get_thread(tid);
268 if ( scheduler->get_enabled(thr) == THREAD_SLEEP_SET &&
269 thr->get_pending() == NULL ) {
270 thr->set_state(THREAD_RUNNING);
271 scheduler->next_thread(thr);
272 Thread::swap(&system_context, thr);
273 priv->current_action->set_sleep_flag();
274 thr->set_pending(priv->current_action);
277 priv->current_action = NULL;
280 void ModelChecker::wake_up_sleeping_actions(ModelAction * curr) {
281 for(unsigned int i=0;i<get_num_threads();i++) {
282 thread_id_t tid=int_to_id(i);
283 Thread *thr=get_thread(tid);
284 if ( scheduler->get_enabled(thr) == THREAD_SLEEP_SET ) {
285 ModelAction *pending_act=thr->get_pending();
286 if ((!curr->is_rmwr())&&pending_act->could_synchronize_with(curr)) {
287 //Remove this thread from sleep set
288 scheduler->remove_sleep(thr);
295 * Check if we are in a deadlock. Should only be called at the end of an
296 * execution, although it should not give false positives in the middle of an
297 * execution (there should be some ENABLED thread).
299 * @return True if program is in a deadlock; false otherwise
301 bool ModelChecker::is_deadlocked() const
303 bool blocking_threads = false;
304 for (unsigned int i = 0; i < get_num_threads(); i++) {
305 thread_id_t tid = int_to_id(i);
308 Thread *t = get_thread(tid);
309 if (!t->is_model_thread() && t->get_pending())
310 blocking_threads = true;
312 return blocking_threads;
316 * Check if this is a complete execution. That is, have all thread completed
317 * execution (rather than exiting because sleep sets have forced a redundant
320 * @return True if the execution is complete.
322 bool ModelChecker::is_complete_execution() const
324 for (unsigned int i = 0; i < get_num_threads(); i++)
325 if (is_enabled(int_to_id(i)))
331 * @brief Assert a bug in the executing program.
333 * Use this function to assert any sort of bug in the user program. If the
334 * current trace is feasible (actually, a prefix of some feasible execution),
335 * then this execution will be aborted, printing the appropriate message. If
336 * the current trace is not yet feasible, the error message will be stashed and
337 * printed if the execution ever becomes feasible.
339 * This function can also be used to immediately trigger the bug; that is, we
340 * don't wait for a feasible execution before aborting. Only use the
341 * "immediate" option when you know that the infeasibility is justified (e.g.,
342 * pending release sequences are not a problem)
344 * @param msg Descriptive message for the bug (do not include newline char)
345 * @param user_thread Was this assertion triggered from a user thread?
346 * @param immediate Should this bug be triggered immediately?
348 void ModelChecker::assert_bug(const char *msg, bool user_thread, bool immediate)
350 priv->bugs.push_back(new bug_message(msg));
352 if (immediate || isfeasibleprefix()) {
355 switch_to_master(NULL);
360 * @brief Assert a bug in the executing program, with a default message
361 * @see ModelChecker::assert_bug
362 * @param user_thread Was this assertion triggered from a user thread?
364 void ModelChecker::assert_bug(bool user_thread)
366 assert_bug("bug detected", user_thread);
370 * @brief Assert a bug in the executing program immediately
371 * @see ModelChecker::assert_bug
372 * @param msg Descriptive message for the bug (do not include newline char)
374 void ModelChecker::assert_bug_immediate(const char *msg)
376 printf("Feasible: %s\n", isfeasibleprefix() ? "yes" : "no");
377 assert_bug(msg, false, true);
380 /** @return True, if any bugs have been reported for this execution */
381 bool ModelChecker::have_bug_reports() const
383 return priv->bugs.size() != 0;
386 /** @brief Print bug report listing for this execution (if any bugs exist) */
387 void ModelChecker::print_bugs() const
389 if (have_bug_reports()) {
390 printf("Bug report: %zu bugs detected\n", priv->bugs.size());
391 for (unsigned int i = 0; i < priv->bugs.size(); i++)
392 priv->bugs[i]->print();
397 * Queries the model-checker for more executions to explore and, if one
398 * exists, resets the model-checker state to execute a new execution.
400 * @return If there are more executions to explore, return true. Otherwise,
403 bool ModelChecker::next_execution()
410 printf("ERROR: DEADLOCK\n");
411 if (isfinalfeasible()) {
412 printf("Earliest divergence point since last feasible execution:\n");
413 if (earliest_diverge)
414 earliest_diverge->print();
416 printf("(Not set)\n");
418 earliest_diverge = NULL;
419 num_feasible_executions++;
423 if (isfinalfeasible() || DBG_ENABLED()) {
428 if ((diverge = get_next_backtrack()) == NULL)
432 printf("Next execution will diverge at:\n");
436 reset_to_initial_state();
440 ModelAction * ModelChecker::get_last_conflict(ModelAction *act)
442 switch (act->get_type()) {
446 /* linear search: from most recent to oldest */
447 action_list_t *list = get_safe_ptr_action(obj_map, act->get_location());
448 action_list_t::reverse_iterator rit;
449 for (rit = list->rbegin(); rit != list->rend(); rit++) {
450 ModelAction *prev = *rit;
451 if (prev->could_synchronize_with(act))
457 case ATOMIC_TRYLOCK: {
458 /* linear search: from most recent to oldest */
459 action_list_t *list = get_safe_ptr_action(obj_map, act->get_location());
460 action_list_t::reverse_iterator rit;
461 for (rit = list->rbegin(); rit != list->rend(); rit++) {
462 ModelAction *prev = *rit;
463 if (act->is_conflicting_lock(prev))
468 case ATOMIC_UNLOCK: {
469 /* linear search: from most recent to oldest */
470 action_list_t *list = get_safe_ptr_action(obj_map, act->get_location());
471 action_list_t::reverse_iterator rit;
472 for (rit = list->rbegin(); rit != list->rend(); rit++) {
473 ModelAction *prev = *rit;
474 if (!act->same_thread(prev)&&prev->is_failed_trylock())
480 /* linear search: from most recent to oldest */
481 action_list_t *list = get_safe_ptr_action(obj_map, act->get_location());
482 action_list_t::reverse_iterator rit;
483 for (rit = list->rbegin(); rit != list->rend(); rit++) {
484 ModelAction *prev = *rit;
485 if (!act->same_thread(prev)&&prev->is_failed_trylock())
487 if (!act->same_thread(prev)&&prev->is_notify())
493 case ATOMIC_NOTIFY_ALL:
494 case ATOMIC_NOTIFY_ONE: {
495 /* linear search: from most recent to oldest */
496 action_list_t *list = get_safe_ptr_action(obj_map, act->get_location());
497 action_list_t::reverse_iterator rit;
498 for (rit = list->rbegin(); rit != list->rend(); rit++) {
499 ModelAction *prev = *rit;
500 if (!act->same_thread(prev)&&prev->is_wait())
511 /** This method finds backtracking points where we should try to
512 * reorder the parameter ModelAction against.
514 * @param the ModelAction to find backtracking points for.
516 void ModelChecker::set_backtracking(ModelAction *act)
518 Thread *t = get_thread(act);
519 ModelAction * prev = get_last_conflict(act);
523 Node * node = prev->get_node()->get_parent();
525 int low_tid, high_tid;
526 if (node->is_enabled(t)) {
527 low_tid = id_to_int(act->get_tid());
528 high_tid = low_tid+1;
531 high_tid = get_num_threads();
534 for(int i = low_tid; i < high_tid; i++) {
535 thread_id_t tid = int_to_id(i);
537 /* Make sure this thread can be enabled here. */
538 if (i >= node->get_num_threads())
541 /* Don't backtrack into a point where the thread is disabled or sleeping. */
542 if (node->enabled_status(tid)!=THREAD_ENABLED)
545 /* Check if this has been explored already */
546 if (node->has_been_explored(tid))
549 /* See if fairness allows */
550 if (model->params.fairwindow != 0 && !node->has_priority(tid)) {
552 for(int t=0;t<node->get_num_threads();t++) {
553 thread_id_t tother=int_to_id(t);
554 if (node->is_enabled(tother) && node->has_priority(tother)) {
562 /* Cache the latest backtracking point */
563 if (!priv->next_backtrack || *prev > *priv->next_backtrack)
564 priv->next_backtrack = prev;
566 /* If this is a new backtracking point, mark the tree */
567 if (!node->set_backtrack(tid))
569 DEBUG("Setting backtrack: conflict = %d, instead tid = %d\n",
570 id_to_int(prev->get_tid()),
571 id_to_int(t->get_id()));
580 * Returns last backtracking point. The model checker will explore a different
581 * path for this point in the next execution.
582 * @return The ModelAction at which the next execution should diverge.
584 ModelAction * ModelChecker::get_next_backtrack()
586 ModelAction *next = priv->next_backtrack;
587 priv->next_backtrack = NULL;
592 * Processes a read or rmw model action.
593 * @param curr is the read model action to process.
594 * @param second_part_of_rmw is boolean that is true is this is the second action of a rmw.
595 * @return True if processing this read updates the mo_graph.
597 bool ModelChecker::process_read(ModelAction *curr, bool second_part_of_rmw)
599 uint64_t value = VALUE_NONE;
600 bool updated = false;
602 const ModelAction *reads_from = curr->get_node()->get_read_from();
603 if (reads_from != NULL) {
604 mo_graph->startChanges();
606 value = reads_from->get_value();
607 bool r_status = false;
609 if (!second_part_of_rmw) {
610 check_recency(curr, reads_from);
611 r_status = r_modification_order(curr, reads_from);
615 if (!second_part_of_rmw&&!isfeasible()&&(curr->get_node()->increment_read_from()||curr->get_node()->increment_future_value())) {
616 mo_graph->rollbackChanges();
617 too_many_reads = false;
621 curr->read_from(reads_from);
622 mo_graph->commitChanges();
623 mo_check_promises(curr->get_tid(), reads_from);
626 } else if (!second_part_of_rmw) {
627 /* Read from future value */
628 value = curr->get_node()->get_future_value();
629 modelclock_t expiration = curr->get_node()->get_future_value_expiration();
630 curr->read_from(NULL);
631 Promise *valuepromise = new Promise(curr, value, expiration);
632 promises->push_back(valuepromise);
634 get_thread(curr)->set_return_value(value);
640 * Processes a lock, trylock, or unlock model action. @param curr is
641 * the read model action to process.
643 * The try lock operation checks whether the lock is taken. If not,
644 * it falls to the normal lock operation case. If so, it returns
647 * The lock operation has already been checked that it is enabled, so
648 * it just grabs the lock and synchronizes with the previous unlock.
650 * The unlock operation has to re-enable all of the threads that are
651 * waiting on the lock.
653 * @return True if synchronization was updated; false otherwise
655 bool ModelChecker::process_mutex(ModelAction *curr) {
656 std::mutex *mutex=NULL;
657 struct std::mutex_state *state=NULL;
659 if (curr->is_trylock() || curr->is_lock() || curr->is_unlock()) {
660 mutex = (std::mutex *)curr->get_location();
661 state = mutex->get_state();
662 } else if(curr->is_wait()) {
663 mutex = (std::mutex *)curr->get_value();
664 state = mutex->get_state();
667 switch (curr->get_type()) {
668 case ATOMIC_TRYLOCK: {
669 bool success = !state->islocked;
670 curr->set_try_lock(success);
672 get_thread(curr)->set_return_value(0);
675 get_thread(curr)->set_return_value(1);
677 //otherwise fall into the lock case
679 if (curr->get_cv()->getClock(state->alloc_tid) <= state->alloc_clock) {
680 printf("Lock access before initialization\n");
683 state->islocked = true;
684 ModelAction *unlock = get_last_unlock(curr);
685 //synchronize with the previous unlock statement
686 if (unlock != NULL) {
687 curr->synchronize_with(unlock);
692 case ATOMIC_UNLOCK: {
694 state->islocked = false;
695 //wake up the other threads
696 action_list_t *waiters = get_safe_ptr_action(lock_waiters_map, curr->get_location());
697 //activate all the waiting threads
698 for (action_list_t::iterator rit = waiters->begin(); rit != waiters->end(); rit++) {
699 scheduler->wake(get_thread(*rit));
706 state->islocked = false;
707 //wake up the other threads
708 action_list_t *waiters = get_safe_ptr_action(lock_waiters_map, (void *) curr->get_value());
709 //activate all the waiting threads
710 for (action_list_t::iterator rit = waiters->begin(); rit != waiters->end(); rit++) {
711 scheduler->wake(get_thread(*rit));
714 //check whether we should go to sleep or not...simulate spurious failures
715 if (curr->get_node()->get_misc()==0) {
716 get_safe_ptr_action(condvar_waiters_map, curr->get_location())->push_back(curr);
718 scheduler->sleep(get_current_thread());
722 case ATOMIC_NOTIFY_ALL: {
723 action_list_t *waiters = get_safe_ptr_action(condvar_waiters_map, curr->get_location());
724 //activate all the waiting threads
725 for (action_list_t::iterator rit = waiters->begin(); rit != waiters->end(); rit++) {
726 scheduler->wake(get_thread(*rit));
731 case ATOMIC_NOTIFY_ONE: {
732 action_list_t *waiters = get_safe_ptr_action(condvar_waiters_map, curr->get_location());
733 int wakeupthread=curr->get_node()->get_misc();
734 action_list_t::iterator it = waiters->begin();
735 advance(it, wakeupthread);
736 scheduler->wake(get_thread(*it));
748 * Process a write ModelAction
749 * @param curr The ModelAction to process
750 * @return True if the mo_graph was updated or promises were resolved
752 bool ModelChecker::process_write(ModelAction *curr)
754 bool updated_mod_order = w_modification_order(curr);
755 bool updated_promises = resolve_promises(curr);
757 if (promises->size() == 0) {
758 for (unsigned int i = 0; i < futurevalues->size(); i++) {
759 struct PendingFutureValue pfv = (*futurevalues)[i];
760 //Do more ambitious checks now that mo is more complete
761 if (mo_may_allow(pfv.writer, pfv.act)&&
762 pfv.act->get_node()->add_future_value(pfv.writer->get_value(), pfv.writer->get_seq_number()+params.maxfuturedelay) &&
763 (!priv->next_backtrack || *pfv.act > *priv->next_backtrack))
764 priv->next_backtrack = pfv.act;
766 futurevalues->resize(0);
769 mo_graph->commitChanges();
770 mo_check_promises(curr->get_tid(), curr);
772 get_thread(curr)->set_return_value(VALUE_NONE);
773 return updated_mod_order || updated_promises;
777 * @brief Process the current action for thread-related activity
779 * Performs current-action processing for a THREAD_* ModelAction. Proccesses
780 * may include setting Thread status, completing THREAD_FINISH/THREAD_JOIN
781 * synchronization, etc. This function is a no-op for non-THREAD actions
782 * (e.g., ATOMIC_{READ,WRITE,RMW,LOCK}, etc.)
784 * @param curr The current action
785 * @return True if synchronization was updated or a thread completed
787 bool ModelChecker::process_thread_action(ModelAction *curr)
789 bool updated = false;
791 switch (curr->get_type()) {
792 case THREAD_CREATE: {
793 Thread *th = (Thread *)curr->get_location();
794 th->set_creation(curr);
798 Thread *blocking = (Thread *)curr->get_location();
799 ModelAction *act = get_last_action(blocking->get_id());
800 curr->synchronize_with(act);
801 updated = true; /* trigger rel-seq checks */
804 case THREAD_FINISH: {
805 Thread *th = get_thread(curr);
806 while (!th->wait_list_empty()) {
807 ModelAction *act = th->pop_wait_list();
808 scheduler->wake(get_thread(act));
811 updated = true; /* trigger rel-seq checks */
815 check_promises(curr->get_tid(), NULL, curr->get_cv());
826 * @brief Process the current action for release sequence fixup activity
828 * Performs model-checker release sequence fixups for the current action,
829 * forcing a single pending release sequence to break (with a given, potential
830 * "loose" write) or to complete (i.e., synchronize). If a pending release
831 * sequence forms a complete release sequence, then we must perform the fixup
832 * synchronization, mo_graph additions, etc.
834 * @param curr The current action; must be a release sequence fixup action
835 * @param work_queue The work queue to which to add work items as they are
838 void ModelChecker::process_relseq_fixup(ModelAction *curr, work_queue_t *work_queue)
840 const ModelAction *write = curr->get_node()->get_relseq_break();
841 struct release_seq *sequence = pending_rel_seqs->back();
842 pending_rel_seqs->pop_back();
844 ModelAction *acquire = sequence->acquire;
845 const ModelAction *rf = sequence->rf;
846 const ModelAction *release = sequence->release;
850 ASSERT(release->same_thread(rf));
854 * @todo Forcing a synchronization requires that we set
855 * modification order constraints. For instance, we can't allow
856 * a fixup sequence in which two separate read-acquire
857 * operations read from the same sequence, where the first one
858 * synchronizes and the other doesn't. Essentially, we can't
859 * allow any writes to insert themselves between 'release' and
863 /* Must synchronize */
864 if (!acquire->synchronize_with(release)) {
865 set_bad_synchronization();
868 /* Re-check all pending release sequences */
869 work_queue->push_back(CheckRelSeqWorkEntry(NULL));
870 /* Re-check act for mo_graph edges */
871 work_queue->push_back(MOEdgeWorkEntry(acquire));
873 /* propagate synchronization to later actions */
874 action_list_t::reverse_iterator rit = action_trace->rbegin();
875 for (; (*rit) != acquire; rit++) {
876 ModelAction *propagate = *rit;
877 if (acquire->happens_before(propagate)) {
878 propagate->synchronize_with(acquire);
879 /* Re-check 'propagate' for mo_graph edges */
880 work_queue->push_back(MOEdgeWorkEntry(propagate));
884 /* Break release sequence with new edges:
885 * release --mo--> write --mo--> rf */
886 mo_graph->addEdge(release, write);
887 mo_graph->addEdge(write, rf);
890 /* See if we have realized a data race */
891 if (checkDataRaces())
896 * Initialize the current action by performing one or more of the following
897 * actions, as appropriate: merging RMWR and RMWC/RMW actions, stepping forward
898 * in the NodeStack, manipulating backtracking sets, allocating and
899 * initializing clock vectors, and computing the promises to fulfill.
901 * @param curr The current action, as passed from the user context; may be
902 * freed/invalidated after the execution of this function, with a different
903 * action "returned" its place (pass-by-reference)
904 * @return True if curr is a newly-explored action; false otherwise
906 bool ModelChecker::initialize_curr_action(ModelAction **curr)
908 ModelAction *newcurr;
910 if ((*curr)->is_rmwc() || (*curr)->is_rmw()) {
911 newcurr = process_rmw(*curr);
914 if (newcurr->is_rmw())
915 compute_promises(newcurr);
921 (*curr)->set_seq_number(get_next_seq_num());
923 newcurr = node_stack->explore_action(*curr, scheduler->get_enabled());
925 /* First restore type and order in case of RMW operation */
926 if ((*curr)->is_rmwr())
927 newcurr->copy_typeandorder(*curr);
929 ASSERT((*curr)->get_location() == newcurr->get_location());
930 newcurr->copy_from_new(*curr);
932 /* Discard duplicate ModelAction; use action from NodeStack */
935 /* Always compute new clock vector */
936 newcurr->create_cv(get_parent_action(newcurr->get_tid()));
939 return false; /* Action was explored previously */
943 /* Always compute new clock vector */
944 newcurr->create_cv(get_parent_action(newcurr->get_tid()));
946 * Perform one-time actions when pushing new ModelAction onto
949 if (newcurr->is_write())
950 compute_promises(newcurr);
951 else if (newcurr->is_relseq_fixup())
952 compute_relseq_breakwrites(newcurr);
953 else if (newcurr->is_wait())
954 newcurr->get_node()->set_misc_max(2);
955 else if (newcurr->is_notify_one()) {
956 newcurr->get_node()->set_misc_max(get_safe_ptr_action(condvar_waiters_map, newcurr->get_location())->size());
958 return true; /* This was a new ModelAction */
963 * @brief Check whether a model action is enabled.
965 * Checks whether a lock or join operation would be successful (i.e., is the
966 * lock already locked, or is the joined thread already complete). If not, put
967 * the action in a waiter list.
969 * @param curr is the ModelAction to check whether it is enabled.
970 * @return a bool that indicates whether the action is enabled.
972 bool ModelChecker::check_action_enabled(ModelAction *curr) {
973 if (curr->is_lock()) {
974 std::mutex * lock = (std::mutex *)curr->get_location();
975 struct std::mutex_state * state = lock->get_state();
976 if (state->islocked) {
977 //Stick the action in the appropriate waiting queue
978 get_safe_ptr_action(lock_waiters_map, curr->get_location())->push_back(curr);
981 } else if (curr->get_type() == THREAD_JOIN) {
982 Thread *blocking = (Thread *)curr->get_location();
983 if (!blocking->is_complete()) {
984 blocking->push_wait_list(curr);
993 * Stores the ModelAction for the current thread action. Call this
994 * immediately before switching from user- to system-context to pass
996 * @param act The ModelAction created by the user-thread action
998 void ModelChecker::set_current_action(ModelAction *act) {
999 priv->current_action = act;
1003 * This is the heart of the model checker routine. It performs model-checking
1004 * actions corresponding to a given "current action." Among other processes, it
1005 * calculates reads-from relationships, updates synchronization clock vectors,
1006 * forms a memory_order constraints graph, and handles replay/backtrack
1007 * execution when running permutations of previously-observed executions.
1009 * @param curr The current action to process
1010 * @return The next Thread that must be executed. May be NULL if ModelChecker
1011 * makes no choice (e.g., according to replay execution, combining RMW actions,
1014 Thread * ModelChecker::check_current_action(ModelAction *curr)
1017 bool second_part_of_rmw = curr->is_rmwc() || curr->is_rmw();
1019 if (!check_action_enabled(curr)) {
1020 /* Make the execution look like we chose to run this action
1021 * much later, when a lock/join can succeed */
1022 get_current_thread()->set_pending(curr);
1023 scheduler->sleep(get_current_thread());
1024 return get_next_thread(NULL);
1027 bool newly_explored = initialize_curr_action(&curr);
1029 wake_up_sleeping_actions(curr);
1031 /* Add the action to lists before any other model-checking tasks */
1032 if (!second_part_of_rmw)
1033 add_action_to_lists(curr);
1035 /* Build may_read_from set for newly-created actions */
1036 if (newly_explored && curr->is_read())
1037 build_reads_from_past(curr);
1039 /* Initialize work_queue with the "current action" work */
1040 work_queue_t work_queue(1, CheckCurrWorkEntry(curr));
1041 while (!work_queue.empty()) {
1042 WorkQueueEntry work = work_queue.front();
1043 work_queue.pop_front();
1045 switch (work.type) {
1046 case WORK_CHECK_CURR_ACTION: {
1047 ModelAction *act = work.action;
1048 bool update = false; /* update this location's release seq's */
1049 bool update_all = false; /* update all release seq's */
1051 if (process_thread_action(curr))
1054 if (act->is_read() && process_read(act, second_part_of_rmw))
1057 if (act->is_write() && process_write(act))
1060 if (act->is_mutex_op() && process_mutex(act))
1063 if (act->is_relseq_fixup())
1064 process_relseq_fixup(curr, &work_queue);
1067 work_queue.push_back(CheckRelSeqWorkEntry(NULL));
1069 work_queue.push_back(CheckRelSeqWorkEntry(act->get_location()));
1072 case WORK_CHECK_RELEASE_SEQ:
1073 resolve_release_sequences(work.location, &work_queue);
1075 case WORK_CHECK_MO_EDGES: {
1076 /** @todo Complete verification of work_queue */
1077 ModelAction *act = work.action;
1078 bool updated = false;
1080 if (act->is_read()) {
1081 const ModelAction *rf = act->get_reads_from();
1082 if (rf != NULL && r_modification_order(act, rf))
1085 if (act->is_write()) {
1086 if (w_modification_order(act))
1089 mo_graph->commitChanges();
1092 work_queue.push_back(CheckRelSeqWorkEntry(act->get_location()));
1101 check_curr_backtracking(curr);
1102 set_backtracking(curr);
1103 return get_next_thread(curr);
1106 void ModelChecker::check_curr_backtracking(ModelAction * curr) {
1107 Node *currnode = curr->get_node();
1108 Node *parnode = currnode->get_parent();
1110 if ((!parnode->backtrack_empty() ||
1111 !currnode->misc_empty() ||
1112 !currnode->read_from_empty() ||
1113 !currnode->future_value_empty() ||
1114 !currnode->promise_empty() ||
1115 !currnode->relseq_break_empty())
1116 && (!priv->next_backtrack ||
1117 *curr > *priv->next_backtrack)) {
1118 priv->next_backtrack = curr;
1122 bool ModelChecker::promises_expired() const
1124 for (unsigned int promise_index = 0; promise_index < promises->size(); promise_index++) {
1125 Promise *promise = (*promises)[promise_index];
1126 if (promise->get_expiration()<priv->used_sequence_numbers) {
1133 /** @return whether the current partial trace must be a prefix of a
1134 * feasible trace. */
1135 bool ModelChecker::isfeasibleprefix() const
1137 return promises->size() == 0 && pending_rel_seqs->size() == 0 && isfeasible();
1140 /** @return whether the current partial trace is feasible. */
1141 bool ModelChecker::isfeasible() const
1143 if (DBG_ENABLED() && mo_graph->checkForRMWViolation())
1144 DEBUG("Infeasible: RMW violation\n");
1146 return !mo_graph->checkForRMWViolation() && isfeasibleotherthanRMW();
1149 /** @return whether the current partial trace is feasible other than
1150 * multiple RMW reading from the same store. */
1151 bool ModelChecker::isfeasibleotherthanRMW() const
1153 if (DBG_ENABLED()) {
1154 if (mo_graph->checkForCycles())
1155 DEBUG("Infeasible: modification order cycles\n");
1157 DEBUG("Infeasible: failed promise\n");
1159 DEBUG("Infeasible: too many reads\n");
1160 if (bad_synchronization)
1161 DEBUG("Infeasible: bad synchronization ordering\n");
1162 if (promises_expired())
1163 DEBUG("Infeasible: promises expired\n");
1165 return !mo_graph->checkForCycles() && !failed_promise && !too_many_reads && !bad_synchronization && !promises_expired();
1168 /** Returns whether the current completed trace is feasible. */
1169 bool ModelChecker::isfinalfeasible() const
1171 if (DBG_ENABLED() && promises->size() != 0)
1172 DEBUG("Infeasible: unrevolved promises\n");
1174 return isfeasible() && promises->size() == 0;
1177 /** Close out a RMWR by converting previous RMWR into a RMW or READ. */
1178 ModelAction * ModelChecker::process_rmw(ModelAction *act) {
1179 ModelAction *lastread = get_last_action(act->get_tid());
1180 lastread->process_rmw(act);
1181 if (act->is_rmw() && lastread->get_reads_from()!=NULL) {
1182 mo_graph->addRMWEdge(lastread->get_reads_from(), lastread);
1183 mo_graph->commitChanges();
1189 * Checks whether a thread has read from the same write for too many times
1190 * without seeing the effects of a later write.
1193 * 1) there must a different write that we could read from that would satisfy the modification order,
1194 * 2) we must have read from the same value in excess of maxreads times, and
1195 * 3) that other write must have been in the reads_from set for maxreads times.
1197 * If so, we decide that the execution is no longer feasible.
1199 void ModelChecker::check_recency(ModelAction *curr, const ModelAction *rf) {
1200 if (params.maxreads != 0) {
1202 if (curr->get_node()->get_read_from_size() <= 1)
1204 //Must make sure that execution is currently feasible... We could
1205 //accidentally clear by rolling back
1208 std::vector<action_list_t> *thrd_lists = get_safe_ptr_vect_action(obj_thrd_map, curr->get_location());
1209 int tid = id_to_int(curr->get_tid());
1212 if ((int)thrd_lists->size() <= tid)
1214 action_list_t *list = &(*thrd_lists)[tid];
1216 action_list_t::reverse_iterator rit = list->rbegin();
1217 /* Skip past curr */
1218 for (; (*rit) != curr; rit++)
1220 /* go past curr now */
1223 action_list_t::reverse_iterator ritcopy = rit;
1224 //See if we have enough reads from the same value
1226 for (; count < params.maxreads; rit++,count++) {
1227 if (rit==list->rend())
1229 ModelAction *act = *rit;
1230 if (!act->is_read())
1233 if (act->get_reads_from() != rf)
1235 if (act->get_node()->get_read_from_size() <= 1)
1238 for (int i = 0; i<curr->get_node()->get_read_from_size(); i++) {
1240 const ModelAction * write = curr->get_node()->get_read_from_at(i);
1242 //Need a different write
1246 /* Test to see whether this is a feasible write to read from*/
1247 mo_graph->startChanges();
1248 r_modification_order(curr, write);
1249 bool feasiblereadfrom = isfeasible();
1250 mo_graph->rollbackChanges();
1252 if (!feasiblereadfrom)
1256 bool feasiblewrite = true;
1257 //new we need to see if this write works for everyone
1259 for (int loop = count; loop>0; loop--,rit++) {
1260 ModelAction *act=*rit;
1261 bool foundvalue = false;
1262 for (int j = 0; j<act->get_node()->get_read_from_size(); j++) {
1263 if (act->get_node()->get_read_from_at(j)==write) {
1269 feasiblewrite = false;
1273 if (feasiblewrite) {
1274 too_many_reads = true;
1282 * Updates the mo_graph with the constraints imposed from the current
1285 * Basic idea is the following: Go through each other thread and find
1286 * the lastest action that happened before our read. Two cases:
1288 * (1) The action is a write => that write must either occur before
1289 * the write we read from or be the write we read from.
1291 * (2) The action is a read => the write that that action read from
1292 * must occur before the write we read from or be the same write.
1294 * @param curr The current action. Must be a read.
1295 * @param rf The action that curr reads from. Must be a write.
1296 * @return True if modification order edges were added; false otherwise
1298 bool ModelChecker::r_modification_order(ModelAction *curr, const ModelAction *rf)
1300 std::vector<action_list_t> *thrd_lists = get_safe_ptr_vect_action(obj_thrd_map, curr->get_location());
1303 ASSERT(curr->is_read());
1305 /* Iterate over all threads */
1306 for (i = 0; i < thrd_lists->size(); i++) {
1307 /* Iterate over actions in thread, starting from most recent */
1308 action_list_t *list = &(*thrd_lists)[i];
1309 action_list_t::reverse_iterator rit;
1310 for (rit = list->rbegin(); rit != list->rend(); rit++) {
1311 ModelAction *act = *rit;
1314 * Include at most one act per-thread that "happens
1315 * before" curr. Don't consider reflexively.
1317 if (act->happens_before(curr) && act != curr) {
1318 if (act->is_write()) {
1320 mo_graph->addEdge(act, rf);
1324 const ModelAction *prevreadfrom = act->get_reads_from();
1325 //if the previous read is unresolved, keep going...
1326 if (prevreadfrom == NULL)
1329 if (rf != prevreadfrom) {
1330 mo_graph->addEdge(prevreadfrom, rf);
1342 /** This method fixes up the modification order when we resolve a
1343 * promises. The basic problem is that actions that occur after the
1344 * read curr could not property add items to the modification order
1347 * So for each thread, we find the earliest item that happens after
1348 * the read curr. This is the item we have to fix up with additional
1349 * constraints. If that action is write, we add a MO edge between
1350 * the Action rf and that action. If the action is a read, we add a
1351 * MO edge between the Action rf, and whatever the read accessed.
1353 * @param curr is the read ModelAction that we are fixing up MO edges for.
1354 * @param rf is the write ModelAction that curr reads from.
1357 void ModelChecker::post_r_modification_order(ModelAction *curr, const ModelAction *rf)
1359 std::vector<action_list_t> *thrd_lists = get_safe_ptr_vect_action(obj_thrd_map, curr->get_location());
1361 ASSERT(curr->is_read());
1363 /* Iterate over all threads */
1364 for (i = 0; i < thrd_lists->size(); i++) {
1365 /* Iterate over actions in thread, starting from most recent */
1366 action_list_t *list = &(*thrd_lists)[i];
1367 action_list_t::reverse_iterator rit;
1368 ModelAction *lastact = NULL;
1370 /* Find last action that happens after curr that is either not curr or a rmw */
1371 for (rit = list->rbegin(); rit != list->rend(); rit++) {
1372 ModelAction *act = *rit;
1373 if (curr->happens_before(act) && (curr != act || curr->is_rmw())) {
1379 /* Include at most one act per-thread that "happens before" curr */
1380 if (lastact != NULL) {
1381 if (lastact==curr) {
1382 //Case 1: The resolved read is a RMW, and we need to make sure
1383 //that the write portion of the RMW mod order after rf
1385 mo_graph->addEdge(rf, lastact);
1386 } else if (lastact->is_read()) {
1387 //Case 2: The resolved read is a normal read and the next
1388 //operation is a read, and we need to make sure the value read
1389 //is mod ordered after rf
1391 const ModelAction *postreadfrom = lastact->get_reads_from();
1392 if (postreadfrom != NULL&&rf != postreadfrom)
1393 mo_graph->addEdge(rf, postreadfrom);
1395 //Case 3: The resolved read is a normal read and the next
1396 //operation is a write, and we need to make sure that the
1397 //write is mod ordered after rf
1399 mo_graph->addEdge(rf, lastact);
1407 * Updates the mo_graph with the constraints imposed from the current write.
1409 * Basic idea is the following: Go through each other thread and find
1410 * the lastest action that happened before our write. Two cases:
1412 * (1) The action is a write => that write must occur before
1415 * (2) The action is a read => the write that that action read from
1416 * must occur before the current write.
1418 * This method also handles two other issues:
1420 * (I) Sequential Consistency: Making sure that if the current write is
1421 * seq_cst, that it occurs after the previous seq_cst write.
1423 * (II) Sending the write back to non-synchronizing reads.
1425 * @param curr The current action. Must be a write.
1426 * @return True if modification order edges were added; false otherwise
1428 bool ModelChecker::w_modification_order(ModelAction *curr)
1430 std::vector<action_list_t> *thrd_lists = get_safe_ptr_vect_action(obj_thrd_map, curr->get_location());
1433 ASSERT(curr->is_write());
1435 if (curr->is_seqcst()) {
1436 /* We have to at least see the last sequentially consistent write,
1437 so we are initialized. */
1438 ModelAction *last_seq_cst = get_last_seq_cst(curr);
1439 if (last_seq_cst != NULL) {
1440 mo_graph->addEdge(last_seq_cst, curr);
1445 /* Iterate over all threads */
1446 for (i = 0; i < thrd_lists->size(); i++) {
1447 /* Iterate over actions in thread, starting from most recent */
1448 action_list_t *list = &(*thrd_lists)[i];
1449 action_list_t::reverse_iterator rit;
1450 for (rit = list->rbegin(); rit != list->rend(); rit++) {
1451 ModelAction *act = *rit;
1454 * 1) If RMW and it actually read from something, then we
1455 * already have all relevant edges, so just skip to next
1458 * 2) If RMW and it didn't read from anything, we should
1459 * whatever edge we can get to speed up convergence.
1461 * 3) If normal write, we need to look at earlier actions, so
1462 * continue processing list.
1464 if (curr->is_rmw()) {
1465 if (curr->get_reads_from()!=NULL)
1474 * Include at most one act per-thread that "happens
1477 if (act->happens_before(curr)) {
1479 * Note: if act is RMW, just add edge:
1481 * The following edge should be handled elsewhere:
1482 * readfrom(act) --mo--> act
1484 if (act->is_write())
1485 mo_graph->addEdge(act, curr);
1486 else if (act->is_read()) {
1487 //if previous read accessed a null, just keep going
1488 if (act->get_reads_from() == NULL)
1490 mo_graph->addEdge(act->get_reads_from(), curr);
1494 } else if (act->is_read() && !act->could_synchronize_with(curr) &&
1495 !act->same_thread(curr)) {
1496 /* We have an action that:
1497 (1) did not happen before us
1498 (2) is a read and we are a write
1499 (3) cannot synchronize with us
1500 (4) is in a different thread
1502 that read could potentially read from our write. Note that
1503 these checks are overly conservative at this point, we'll
1504 do more checks before actually removing the
1508 if (thin_air_constraint_may_allow(curr, act)) {
1510 (curr->is_rmw() && act->is_rmw() && curr->get_reads_from() == act->get_reads_from() && isfeasibleotherthanRMW())) {
1511 struct PendingFutureValue pfv = {curr,act};
1512 futurevalues->push_back(pfv);
1522 /** Arbitrary reads from the future are not allowed. Section 29.3
1523 * part 9 places some constraints. This method checks one result of constraint
1524 * constraint. Others require compiler support. */
1525 bool ModelChecker::thin_air_constraint_may_allow(const ModelAction * writer, const ModelAction *reader) {
1526 if (!writer->is_rmw())
1529 if (!reader->is_rmw())
1532 for (const ModelAction *search = writer->get_reads_from(); search != NULL; search = search->get_reads_from()) {
1533 if (search == reader)
1535 if (search->get_tid() == reader->get_tid() &&
1536 search->happens_before(reader))
1544 * Arbitrary reads from the future are not allowed. Section 29.3 part 9 places
1545 * some constraints. This method checks one the following constraint (others
1546 * require compiler support):
1548 * If X --hb-> Y --mo-> Z, then X should not read from Z.
1550 bool ModelChecker::mo_may_allow(const ModelAction *writer, const ModelAction *reader)
1552 std::vector<action_list_t> *thrd_lists = get_safe_ptr_vect_action(obj_thrd_map, reader->get_location());
1554 /* Iterate over all threads */
1555 for (i = 0; i < thrd_lists->size(); i++) {
1556 const ModelAction *write_after_read = NULL;
1558 /* Iterate over actions in thread, starting from most recent */
1559 action_list_t *list = &(*thrd_lists)[i];
1560 action_list_t::reverse_iterator rit;
1561 for (rit = list->rbegin(); rit != list->rend(); rit++) {
1562 ModelAction *act = *rit;
1564 if (!reader->happens_before(act))
1566 else if (act->is_write())
1567 write_after_read = act;
1568 else if (act->is_read() && act->get_reads_from() != NULL && act != reader) {
1569 write_after_read = act->get_reads_from();
1573 if (write_after_read && write_after_read!=writer && mo_graph->checkReachable(write_after_read, writer))
1580 * Finds the head(s) of the release sequence(s) containing a given ModelAction.
1581 * The ModelAction under consideration is expected to be taking part in
1582 * release/acquire synchronization as an object of the "reads from" relation.
1583 * Note that this can only provide release sequence support for RMW chains
1584 * which do not read from the future, as those actions cannot be traced until
1585 * their "promise" is fulfilled. Similarly, we may not even establish the
1586 * presence of a release sequence with certainty, as some modification order
1587 * constraints may be decided further in the future. Thus, this function
1588 * "returns" two pieces of data: a pass-by-reference vector of @a release_heads
1589 * and a boolean representing certainty.
1591 * @param rf The action that might be part of a release sequence. Must be a
1593 * @param release_heads A pass-by-reference style return parameter. After
1594 * execution of this function, release_heads will contain the heads of all the
1595 * relevant release sequences, if any exists with certainty
1596 * @param pending A pass-by-reference style return parameter which is only used
1597 * when returning false (i.e., uncertain). Returns most information regarding
1598 * an uncertain release sequence, including any write operations that might
1599 * break the sequence.
1600 * @return true, if the ModelChecker is certain that release_heads is complete;
1603 bool ModelChecker::release_seq_heads(const ModelAction *rf,
1604 rel_heads_list_t *release_heads,
1605 struct release_seq *pending) const
1607 /* Only check for release sequences if there are no cycles */
1608 if (mo_graph->checkForCycles())
1612 ASSERT(rf->is_write());
1614 if (rf->is_release())
1615 release_heads->push_back(rf);
1617 break; /* End of RMW chain */
1619 /** @todo Need to be smarter here... In the linux lock
1620 * example, this will run to the beginning of the program for
1622 /** @todo The way to be smarter here is to keep going until 1
1623 * thread has a release preceded by an acquire and you've seen
1626 /* acq_rel RMW is a sufficient stopping condition */
1627 if (rf->is_acquire() && rf->is_release())
1628 return true; /* complete */
1630 rf = rf->get_reads_from();
1633 /* read from future: need to settle this later */
1635 return false; /* incomplete */
1638 if (rf->is_release())
1639 return true; /* complete */
1641 /* else relaxed write; check modification order for contiguous subsequence
1642 * -> rf must be same thread as release */
1643 int tid = id_to_int(rf->get_tid());
1644 std::vector<action_list_t> *thrd_lists = get_safe_ptr_vect_action(obj_thrd_map, rf->get_location());
1645 action_list_t *list = &(*thrd_lists)[tid];
1646 action_list_t::const_reverse_iterator rit;
1648 /* Find rf in the thread list */
1649 rit = std::find(list->rbegin(), list->rend(), rf);
1650 ASSERT(rit != list->rend());
1652 /* Find the last write/release */
1653 for (; rit != list->rend(); rit++)
1654 if ((*rit)->is_release())
1656 if (rit == list->rend()) {
1657 /* No write-release in this thread */
1658 return true; /* complete */
1660 ModelAction *release = *rit;
1662 ASSERT(rf->same_thread(release));
1664 pending->writes.clear();
1666 bool certain = true;
1667 for (unsigned int i = 0; i < thrd_lists->size(); i++) {
1668 if (id_to_int(rf->get_tid()) == (int)i)
1670 list = &(*thrd_lists)[i];
1672 /* Can we ensure no future writes from this thread may break
1673 * the release seq? */
1674 bool future_ordered = false;
1676 ModelAction *last = get_last_action(int_to_id(i));
1677 Thread *th = get_thread(int_to_id(i));
1678 if ((last && rf->happens_before(last)) ||
1681 future_ordered = true;
1683 ASSERT(!th->is_model_thread() || future_ordered);
1685 for (rit = list->rbegin(); rit != list->rend(); rit++) {
1686 const ModelAction *act = *rit;
1687 /* Reach synchronization -> this thread is complete */
1688 if (act->happens_before(release))
1690 if (rf->happens_before(act)) {
1691 future_ordered = true;
1695 /* Only non-RMW writes can break release sequences */
1696 if (!act->is_write() || act->is_rmw())
1699 /* Check modification order */
1700 if (mo_graph->checkReachable(rf, act)) {
1701 /* rf --mo--> act */
1702 future_ordered = true;
1705 if (mo_graph->checkReachable(act, release))
1706 /* act --mo--> release */
1708 if (mo_graph->checkReachable(release, act) &&
1709 mo_graph->checkReachable(act, rf)) {
1710 /* release --mo-> act --mo--> rf */
1711 return true; /* complete */
1713 /* act may break release sequence */
1714 pending->writes.push_back(act);
1717 if (!future_ordered)
1718 certain = false; /* This thread is uncertain */
1722 release_heads->push_back(release);
1723 pending->writes.clear();
1725 pending->release = release;
1732 * A public interface for getting the release sequence head(s) with which a
1733 * given ModelAction must synchronize. This function only returns a non-empty
1734 * result when it can locate a release sequence head with certainty. Otherwise,
1735 * it may mark the internal state of the ModelChecker so that it will handle
1736 * the release sequence at a later time, causing @a act to update its
1737 * synchronization at some later point in execution.
1738 * @param act The 'acquire' action that may read from a release sequence
1739 * @param release_heads A pass-by-reference return parameter. Will be filled
1740 * with the head(s) of the release sequence(s), if they exists with certainty.
1741 * @see ModelChecker::release_seq_heads
1743 void ModelChecker::get_release_seq_heads(ModelAction *act, rel_heads_list_t *release_heads)
1745 const ModelAction *rf = act->get_reads_from();
1746 struct release_seq *sequence = (struct release_seq *)snapshot_calloc(1, sizeof(struct release_seq));
1747 sequence->acquire = act;
1749 if (!release_seq_heads(rf, release_heads, sequence)) {
1750 /* add act to 'lazy checking' list */
1751 pending_rel_seqs->push_back(sequence);
1753 snapshot_free(sequence);
1758 * Attempt to resolve all stashed operations that might synchronize with a
1759 * release sequence for a given location. This implements the "lazy" portion of
1760 * determining whether or not a release sequence was contiguous, since not all
1761 * modification order information is present at the time an action occurs.
1763 * @param location The location/object that should be checked for release
1764 * sequence resolutions. A NULL value means to check all locations.
1765 * @param work_queue The work queue to which to add work items as they are
1767 * @return True if any updates occurred (new synchronization, new mo_graph
1770 bool ModelChecker::resolve_release_sequences(void *location, work_queue_t *work_queue)
1772 bool updated = false;
1773 std::vector< struct release_seq *, SnapshotAlloc<struct release_seq *> >::iterator it = pending_rel_seqs->begin();
1774 while (it != pending_rel_seqs->end()) {
1775 struct release_seq *pending = *it;
1776 ModelAction *act = pending->acquire;
1778 /* Only resolve sequences on the given location, if provided */
1779 if (location && act->get_location() != location) {
1784 const ModelAction *rf = act->get_reads_from();
1785 rel_heads_list_t release_heads;
1787 complete = release_seq_heads(rf, &release_heads, pending);
1788 for (unsigned int i = 0; i < release_heads.size(); i++) {
1789 if (!act->has_synchronized_with(release_heads[i])) {
1790 if (act->synchronize_with(release_heads[i]))
1793 set_bad_synchronization();
1798 /* Re-check all pending release sequences */
1799 work_queue->push_back(CheckRelSeqWorkEntry(NULL));
1800 /* Re-check act for mo_graph edges */
1801 work_queue->push_back(MOEdgeWorkEntry(act));
1803 /* propagate synchronization to later actions */
1804 action_list_t::reverse_iterator rit = action_trace->rbegin();
1805 for (; (*rit) != act; rit++) {
1806 ModelAction *propagate = *rit;
1807 if (act->happens_before(propagate)) {
1808 propagate->synchronize_with(act);
1809 /* Re-check 'propagate' for mo_graph edges */
1810 work_queue->push_back(MOEdgeWorkEntry(propagate));
1815 it = pending_rel_seqs->erase(it);
1816 snapshot_free(pending);
1822 // If we resolved promises or data races, see if we have realized a data race.
1823 if (checkDataRaces()) {
1831 * Performs various bookkeeping operations for the current ModelAction. For
1832 * instance, adds action to the per-object, per-thread action vector and to the
1833 * action trace list of all thread actions.
1835 * @param act is the ModelAction to add.
1837 void ModelChecker::add_action_to_lists(ModelAction *act)
1839 int tid = id_to_int(act->get_tid());
1840 action_trace->push_back(act);
1842 get_safe_ptr_action(obj_map, act->get_location())->push_back(act);
1844 std::vector<action_list_t> *vec = get_safe_ptr_vect_action(obj_thrd_map, act->get_location());
1845 if (tid >= (int)vec->size())
1846 vec->resize(priv->next_thread_id);
1847 (*vec)[tid].push_back(act);
1849 if ((int)thrd_last_action->size() <= tid)
1850 thrd_last_action->resize(get_num_threads());
1851 (*thrd_last_action)[tid] = act;
1853 if (act->is_wait()) {
1854 void *mutex_loc=(void *) act->get_value();
1855 get_safe_ptr_action(obj_map, mutex_loc)->push_back(act);
1857 std::vector<action_list_t> *vec = get_safe_ptr_vect_action(obj_thrd_map, mutex_loc);
1858 if (tid >= (int)vec->size())
1859 vec->resize(priv->next_thread_id);
1860 (*vec)[tid].push_back(act);
1862 if ((int)thrd_last_action->size() <= tid)
1863 thrd_last_action->resize(get_num_threads());
1864 (*thrd_last_action)[tid] = act;
1869 * @brief Get the last action performed by a particular Thread
1870 * @param tid The thread ID of the Thread in question
1871 * @return The last action in the thread
1873 ModelAction * ModelChecker::get_last_action(thread_id_t tid) const
1875 int threadid = id_to_int(tid);
1876 if (threadid < (int)thrd_last_action->size())
1877 return (*thrd_last_action)[id_to_int(tid)];
1883 * Gets the last memory_order_seq_cst write (in the total global sequence)
1884 * performed on a particular object (i.e., memory location), not including the
1886 * @param curr The current ModelAction; also denotes the object location to
1888 * @return The last seq_cst write
1890 ModelAction * ModelChecker::get_last_seq_cst(ModelAction *curr) const
1892 void *location = curr->get_location();
1893 action_list_t *list = get_safe_ptr_action(obj_map, location);
1894 /* Find: max({i in dom(S) | seq_cst(t_i) && isWrite(t_i) && samevar(t_i, t)}) */
1895 action_list_t::reverse_iterator rit;
1896 for (rit = list->rbegin(); rit != list->rend(); rit++)
1897 if ((*rit)->is_write() && (*rit)->is_seqcst() && (*rit) != curr)
1903 * Gets the last unlock operation performed on a particular mutex (i.e., memory
1904 * location). This function identifies the mutex according to the current
1905 * action, which is presumed to perform on the same mutex.
1906 * @param curr The current ModelAction; also denotes the object location to
1908 * @return The last unlock operation
1910 ModelAction * ModelChecker::get_last_unlock(ModelAction *curr) const
1912 void *location = curr->get_location();
1913 action_list_t *list = get_safe_ptr_action(obj_map, location);
1914 /* Find: max({i in dom(S) | isUnlock(t_i) && samevar(t_i, t)}) */
1915 action_list_t::reverse_iterator rit;
1916 for (rit = list->rbegin(); rit != list->rend(); rit++)
1917 if ((*rit)->is_unlock() || (*rit)->is_wait())
1922 ModelAction * ModelChecker::get_parent_action(thread_id_t tid)
1924 ModelAction *parent = get_last_action(tid);
1926 parent = get_thread(tid)->get_creation();
1931 * Returns the clock vector for a given thread.
1932 * @param tid The thread whose clock vector we want
1933 * @return Desired clock vector
1935 ClockVector * ModelChecker::get_cv(thread_id_t tid)
1937 return get_parent_action(tid)->get_cv();
1941 * Resolve a set of Promises with a current write. The set is provided in the
1942 * Node corresponding to @a write.
1943 * @param write The ModelAction that is fulfilling Promises
1944 * @return True if promises were resolved; false otherwise
1946 bool ModelChecker::resolve_promises(ModelAction *write)
1948 bool resolved = false;
1949 std::vector< thread_id_t, ModelAlloc<thread_id_t> > threads_to_check;
1951 for (unsigned int i = 0, promise_index = 0; promise_index < promises->size(); i++) {
1952 Promise *promise = (*promises)[promise_index];
1953 if (write->get_node()->get_promise(i)) {
1954 ModelAction *read = promise->get_action();
1955 if (read->is_rmw()) {
1956 mo_graph->addRMWEdge(write, read);
1958 read->read_from(write);
1959 //First fix up the modification order for actions that happened
1961 r_modification_order(read, write);
1962 //Next fix up the modification order for actions that happened
1964 post_r_modification_order(read, write);
1965 //Make sure the promise's value matches the write's value
1966 ASSERT(promise->get_value() == write->get_value());
1969 promises->erase(promises->begin() + promise_index);
1970 threads_to_check.push_back(read->get_tid());
1977 //Check whether reading these writes has made threads unable to
1980 for(unsigned int i=0;i<threads_to_check.size();i++)
1981 mo_check_promises(threads_to_check[i], write);
1987 * Compute the set of promises that could potentially be satisfied by this
1988 * action. Note that the set computation actually appears in the Node, not in
1990 * @param curr The ModelAction that may satisfy promises
1992 void ModelChecker::compute_promises(ModelAction *curr)
1994 for (unsigned int i = 0; i < promises->size(); i++) {
1995 Promise *promise = (*promises)[i];
1996 const ModelAction *act = promise->get_action();
1997 if (!act->happens_before(curr) &&
1999 !act->could_synchronize_with(curr) &&
2000 !act->same_thread(curr) &&
2001 act->get_location() == curr->get_location() &&
2002 promise->get_value() == curr->get_value()) {
2003 curr->get_node()->set_promise(i, act->is_rmw());
2008 /** Checks promises in response to change in ClockVector Threads. */
2009 void ModelChecker::check_promises(thread_id_t tid, ClockVector *old_cv, ClockVector *merge_cv)
2011 for (unsigned int i = 0; i < promises->size(); i++) {
2012 Promise *promise = (*promises)[i];
2013 const ModelAction *act = promise->get_action();
2014 if ((old_cv == NULL || !old_cv->synchronized_since(act)) &&
2015 merge_cv->synchronized_since(act)) {
2016 if (promise->increment_threads(tid)) {
2017 //Promise has failed
2018 failed_promise = true;
2025 void ModelChecker::check_promises_thread_disabled() {
2026 for (unsigned int i = 0; i < promises->size(); i++) {
2027 Promise *promise = (*promises)[i];
2028 if (promise->check_promise()) {
2029 failed_promise = true;
2035 /** Checks promises in response to addition to modification order for threads.
2037 * pthread is the thread that performed the read that created the promise
2039 * pread is the read that created the promise
2041 * pwrite is either the first write to same location as pread by
2042 * pthread that is sequenced after pread or the value read by the
2043 * first read to the same lcoation as pread by pthread that is
2044 * sequenced after pread..
2046 * 1. If tid=pthread, then we check what other threads are reachable
2047 * through the mode order starting with pwrite. Those threads cannot
2048 * perform a write that will resolve the promise due to modification
2049 * order constraints.
2051 * 2. If the tid is not pthread, we check whether pwrite can reach the
2052 * action write through the modification order. If so, that thread
2053 * cannot perform a future write that will resolve the promise due to
2054 * modificatin order constraints.
2056 * @parem tid The thread that either read from the model action
2057 * write, or actually did the model action write.
2059 * @parem write The ModelAction representing the relevant write.
2062 void ModelChecker::mo_check_promises(thread_id_t tid, const ModelAction *write) {
2063 void * location = write->get_location();
2064 for (unsigned int i = 0; i < promises->size(); i++) {
2065 Promise *promise = (*promises)[i];
2066 const ModelAction *act = promise->get_action();
2068 //Is this promise on the same location?
2069 if ( act->get_location() != location )
2072 //same thread as the promise
2073 if ( act->get_tid()==tid ) {
2075 //do we have a pwrite for the promise, if not, set it
2076 if (promise->get_write() == NULL ) {
2077 promise->set_write(write);
2078 //The pwrite cannot happen before the promise
2079 if (write->happens_before(act) && (write != act)) {
2080 failed_promise = true;
2084 if (mo_graph->checkPromise(write, promise)) {
2085 failed_promise = true;
2090 //Don't do any lookups twice for the same thread
2091 if (promise->has_sync_thread(tid))
2094 if (promise->get_write()&&mo_graph->checkReachable(promise->get_write(), write)) {
2095 if (promise->increment_threads(tid)) {
2096 failed_promise = true;
2104 * Compute the set of writes that may break the current pending release
2105 * sequence. This information is extracted from previou release sequence
2108 * @param curr The current ModelAction. Must be a release sequence fixup
2111 void ModelChecker::compute_relseq_breakwrites(ModelAction *curr)
2113 if (pending_rel_seqs->empty())
2116 struct release_seq *pending = pending_rel_seqs->back();
2117 for (unsigned int i = 0; i < pending->writes.size(); i++) {
2118 const ModelAction *write = pending->writes[i];
2119 curr->get_node()->add_relseq_break(write);
2122 /* NULL means don't break the sequence; just synchronize */
2123 curr->get_node()->add_relseq_break(NULL);
2127 * Build up an initial set of all past writes that this 'read' action may read
2128 * from. This set is determined by the clock vector's "happens before"
2130 * @param curr is the current ModelAction that we are exploring; it must be a
2133 void ModelChecker::build_reads_from_past(ModelAction *curr)
2135 std::vector<action_list_t> *thrd_lists = get_safe_ptr_vect_action(obj_thrd_map, curr->get_location());
2137 ASSERT(curr->is_read());
2139 ModelAction *last_seq_cst = NULL;
2141 /* Track whether this object has been initialized */
2142 bool initialized = false;
2144 if (curr->is_seqcst()) {
2145 last_seq_cst = get_last_seq_cst(curr);
2146 /* We have to at least see the last sequentially consistent write,
2147 so we are initialized. */
2148 if (last_seq_cst != NULL)
2152 /* Iterate over all threads */
2153 for (i = 0; i < thrd_lists->size(); i++) {
2154 /* Iterate over actions in thread, starting from most recent */
2155 action_list_t *list = &(*thrd_lists)[i];
2156 action_list_t::reverse_iterator rit;
2157 for (rit = list->rbegin(); rit != list->rend(); rit++) {
2158 ModelAction *act = *rit;
2160 /* Only consider 'write' actions */
2161 if (!act->is_write() || act == curr)
2164 /* Don't consider more than one seq_cst write if we are a seq_cst read. */
2165 if (!curr->is_seqcst() || (!act->is_seqcst() && (last_seq_cst == NULL || !act->happens_before(last_seq_cst))) || act == last_seq_cst) {
2166 if (!curr->get_sleep_flag() || curr->is_seqcst() || sleep_can_read_from(curr, act)) {
2167 DEBUG("Adding action to may_read_from:\n");
2168 if (DBG_ENABLED()) {
2172 curr->get_node()->add_read_from(act);
2176 /* Include at most one act per-thread that "happens before" curr */
2177 if (act->happens_before(curr)) {
2185 /** @todo Need a more informative way of reporting errors. */
2186 printf("ERROR: may read from uninitialized atomic\n");
2190 if (DBG_ENABLED() || !initialized) {
2191 printf("Reached read action:\n");
2193 printf("Printing may_read_from\n");
2194 curr->get_node()->print_may_read_from();
2195 printf("End printing may_read_from\n");
2199 bool ModelChecker::sleep_can_read_from(ModelAction * curr, const ModelAction *write) {
2201 Node *prevnode=write->get_node()->get_parent();
2203 bool thread_sleep=prevnode->enabled_status(curr->get_tid())==THREAD_SLEEP_SET;
2204 if (write->is_release()&&thread_sleep)
2206 if (!write->is_rmw()) {
2209 if (write->get_reads_from()==NULL)
2211 write=write->get_reads_from();
2215 static void print_list(action_list_t *list)
2217 action_list_t::iterator it;
2219 printf("---------------------------------------------------------------------\n");
2221 unsigned int hash=0;
2223 for (it = list->begin(); it != list->end(); it++) {
2225 hash=hash^(hash<<3)^((*it)->hash());
2227 printf("HASH %u\n", hash);
2228 printf("---------------------------------------------------------------------\n");
2231 #if SUPPORT_MOD_ORDER_DUMP
2232 void ModelChecker::dumpGraph(char *filename) {
2234 sprintf(buffer, "%s.dot",filename);
2235 FILE *file=fopen(buffer, "w");
2236 fprintf(file, "digraph %s {\n",filename);
2237 mo_graph->dumpNodes(file);
2238 ModelAction ** thread_array=(ModelAction **)model_calloc(1, sizeof(ModelAction *)*get_num_threads());
2240 for (action_list_t::iterator it = action_trace->begin(); it != action_trace->end(); it++) {
2241 ModelAction *action=*it;
2242 if (action->is_read()) {
2243 fprintf(file, "N%u [label=\"%u, T%u\"];\n", action->get_seq_number(),action->get_seq_number(), action->get_tid());
2244 if (action->get_reads_from()!=NULL)
2245 fprintf(file, "N%u -> N%u[label=\"rf\", color=red];\n", action->get_seq_number(), action->get_reads_from()->get_seq_number());
2247 if (thread_array[action->get_tid()] != NULL) {
2248 fprintf(file, "N%u -> N%u[label=\"sb\", color=blue];\n", thread_array[action->get_tid()]->get_seq_number(), action->get_seq_number());
2251 thread_array[action->get_tid()]=action;
2253 fprintf(file,"}\n");
2254 model_free(thread_array);
2259 void ModelChecker::print_summary()
2262 printf("Number of executions: %d\n", num_executions);
2263 printf("Number of feasible executions: %d\n", num_feasible_executions);
2264 printf("Total nodes created: %d\n", node_stack->get_total_nodes());
2266 #if SUPPORT_MOD_ORDER_DUMP
2268 char buffername[100];
2269 sprintf(buffername, "exec%04u", num_executions);
2270 mo_graph->dumpGraphToFile(buffername);
2271 sprintf(buffername, "graph%04u", num_executions);
2272 dumpGraph(buffername);
2275 if (!isfinalfeasible())
2276 printf("INFEASIBLE EXECUTION!\n");
2277 print_list(action_trace);
2282 * Add a Thread to the system for the first time. Should only be called once
2284 * @param t The Thread to add
2286 void ModelChecker::add_thread(Thread *t)
2288 thread_map->put(id_to_int(t->get_id()), t);
2289 scheduler->add_thread(t);
2293 * Removes a thread from the scheduler.
2294 * @param the thread to remove.
2296 void ModelChecker::remove_thread(Thread *t)
2298 scheduler->remove_thread(t);
2302 * @brief Get a Thread reference by its ID
2303 * @param tid The Thread's ID
2304 * @return A Thread reference
2306 Thread * ModelChecker::get_thread(thread_id_t tid) const
2308 return thread_map->get(id_to_int(tid));
2312 * @brief Get a reference to the Thread in which a ModelAction was executed
2313 * @param act The ModelAction
2314 * @return A Thread reference
2316 Thread * ModelChecker::get_thread(ModelAction *act) const
2318 return get_thread(act->get_tid());
2322 * @brief Check if a Thread is currently enabled
2323 * @param t The Thread to check
2324 * @return True if the Thread is currently enabled
2326 bool ModelChecker::is_enabled(Thread *t) const
2328 return scheduler->is_enabled(t);
2332 * @brief Check if a Thread is currently enabled
2333 * @param tid The ID of the Thread to check
2334 * @return True if the Thread is currently enabled
2336 bool ModelChecker::is_enabled(thread_id_t tid) const
2338 return scheduler->is_enabled(tid);
2342 * Switch from a user-context to the "master thread" context (a.k.a. system
2343 * context). This switch is made with the intention of exploring a particular
2344 * model-checking action (described by a ModelAction object). Must be called
2345 * from a user-thread context.
2347 * @param act The current action that will be explored. May be NULL only if
2348 * trace is exiting via an assertion (see ModelChecker::set_assert and
2349 * ModelChecker::has_asserted).
2350 * @return Return status from the 'swap' call (i.e., success/fail, 0/-1)
2352 int ModelChecker::switch_to_master(ModelAction *act)
2355 Thread *old = thread_current();
2356 set_current_action(act);
2357 old->set_state(THREAD_READY);
2358 return Thread::swap(old, &system_context);
2362 * Takes the next step in the execution, if possible.
2363 * @return Returns true (success) if a step was taken and false otherwise.
2365 bool ModelChecker::take_step() {
2369 Thread *curr = priv->current_action ? get_thread(priv->current_action) : NULL;
2371 if (curr->get_state() == THREAD_READY) {
2372 ASSERT(priv->current_action);
2374 priv->nextThread = check_current_action(priv->current_action);
2375 priv->current_action = NULL;
2377 if (curr->is_blocked() || curr->is_complete())
2378 scheduler->remove_thread(curr);
2383 Thread *next = scheduler->next_thread(priv->nextThread);
2385 /* Infeasible -> don't take any more steps */
2388 else if (isfeasibleprefix() && have_bug_reports()) {
2393 if (params.bound != 0) {
2394 if (priv->used_sequence_numbers > params.bound) {
2399 DEBUG("(%d, %d)\n", curr ? id_to_int(curr->get_id()) : -1,
2400 next ? id_to_int(next->get_id()) : -1);
2403 * Launch end-of-execution release sequence fixups only when there are:
2405 * (1) no more user threads to run (or when execution replay chooses
2406 * the 'model_thread')
2407 * (2) pending release sequences
2408 * (3) pending assertions (i.e., data races)
2409 * (4) no pending promises
2411 if (!pending_rel_seqs->empty() && (!next || next->is_model_thread()) &&
2412 isfinalfeasible() && !unrealizedraces.empty()) {
2413 printf("*** WARNING: release sequence fixup action (%zu pending release seuqences) ***\n",
2414 pending_rel_seqs->size());
2415 ModelAction *fixup = new ModelAction(MODEL_FIXUP_RELSEQ,
2416 std::memory_order_seq_cst, NULL, VALUE_NONE,
2418 set_current_action(fixup);
2422 /* next == NULL -> don't take any more steps */
2426 next->set_state(THREAD_RUNNING);
2428 if (next->get_pending() != NULL) {
2429 /* restart a pending action */
2430 set_current_action(next->get_pending());
2431 next->set_pending(NULL);
2432 next->set_state(THREAD_READY);
2436 /* Return false only if swap fails with an error */
2437 return (Thread::swap(&system_context, next) == 0);
2440 /** Runs the current execution until threre are no more steps to take. */
2441 void ModelChecker::finish_execution() {
2444 while (take_step());