11 #include "clockvector.h"
12 #include "cyclegraph.h"
14 #include "threads-model.h"
15 #include "bugmessage.h"
18 #include "newfuzzer.h"
20 #define INITIAL_THREAD_ID 0
23 * Structure for holding small ModelChecker members that should be snapshotted
25 struct model_snapshot_members {
26 model_snapshot_members() :
27 /* First thread created will have id INITIAL_THREAD_ID */
28 next_thread_id(INITIAL_THREAD_ID),
29 used_sequence_numbers(0),
34 ~model_snapshot_members() {
35 for (unsigned int i = 0;i < bugs.size();i++)
40 unsigned int next_thread_id;
41 modelclock_t used_sequence_numbers;
42 SnapVector<bug_message *> bugs;
43 /** @brief Incorrectly-ordered synchronization was made */
49 /** @brief Constructor */
50 ModelExecution::ModelExecution(ModelChecker *m, Scheduler *scheduler) :
55 thread_map(2), /* We'll always need at least 2 threads */
59 condvar_waiters_map(),
63 thrd_last_fence_release(),
64 priv(new struct model_snapshot_members ()),
65 mo_graph(new CycleGraph()),
66 fuzzer(new NewFuzzer()),
68 thrd_func_act_lists(),
71 /* Initialize a model-checker thread, for special ModelActions */
72 model_thread = new Thread(get_next_id());
73 add_thread(model_thread);
74 fuzzer->register_engine(m->get_history(), this);
75 scheduler->register_engine(this);
78 /** @brief Destructor */
79 ModelExecution::~ModelExecution()
81 for (unsigned int i = 0;i < get_num_threads();i++)
82 delete get_thread(int_to_id(i));
88 int ModelExecution::get_execution_number() const
90 return model->get_execution_number();
93 static action_list_t * get_safe_ptr_action(HashTable<const void *, action_list_t *, uintptr_t, 2> * hash, void * ptr)
95 action_list_t *tmp = hash->get(ptr);
97 tmp = new action_list_t();
103 static SnapVector<action_list_t> * get_safe_ptr_vect_action(HashTable<const void *, SnapVector<action_list_t> *, uintptr_t, 2> * hash, void * ptr)
105 SnapVector<action_list_t> *tmp = hash->get(ptr);
107 tmp = new SnapVector<action_list_t>();
113 /** @return a thread ID for a new Thread */
114 thread_id_t ModelExecution::get_next_id()
116 return priv->next_thread_id++;
119 /** @return the number of user threads created during this execution */
120 unsigned int ModelExecution::get_num_threads() const
122 return priv->next_thread_id;
125 /** @return a sequence number for a new ModelAction */
126 modelclock_t ModelExecution::get_next_seq_num()
128 return ++priv->used_sequence_numbers;
131 /** Restore the last used sequence number when actions of a thread are postponed by Fuzzer */
132 void ModelExecution::restore_last_seq_num()
134 priv->used_sequence_numbers--;
138 * @brief Should the current action wake up a given thread?
140 * @param curr The current action
141 * @param thread The thread that we might wake up
142 * @return True, if we should wake up the sleeping thread; false otherwise
144 bool ModelExecution::should_wake_up(const ModelAction *curr, const Thread *thread) const
146 const ModelAction *asleep = thread->get_pending();
147 /* Don't allow partial RMW to wake anyone up */
150 /* Synchronizing actions may have been backtracked */
151 if (asleep->could_synchronize_with(curr))
153 /* All acquire/release fences and fence-acquire/store-release */
154 if (asleep->is_fence() && asleep->is_acquire() && curr->is_release())
156 /* Fence-release + store can awake load-acquire on the same location */
157 if (asleep->is_read() && asleep->is_acquire() && curr->same_var(asleep) && curr->is_write()) {
158 ModelAction *fence_release = get_last_fence_release(curr->get_tid());
159 if (fence_release && *(get_last_action(thread->get_id())) < *fence_release)
162 /* The sleep is literally sleeping */
163 if (asleep->is_sleep()) {
164 if (fuzzer->shouldWake(asleep))
171 void ModelExecution::wake_up_sleeping_actions(ModelAction *curr)
173 for (unsigned int i = 0;i < get_num_threads();i++) {
174 Thread *thr = get_thread(int_to_id(i));
175 if (scheduler->is_sleep_set(thr)) {
176 if (should_wake_up(curr, thr)) {
177 /* Remove this thread from sleep set */
178 scheduler->remove_sleep(thr);
179 if (thr->get_pending()->is_sleep())
180 thr->set_wakeup_state(true);
186 bool ModelExecution::assert_bug(const char *msg)
188 priv->bugs.push_back(new bug_message(msg));
193 /** @return True, if any bugs have been reported for this execution */
194 bool ModelExecution::have_bug_reports() const
196 return priv->bugs.size() != 0;
199 SnapVector<bug_message *> * ModelExecution::get_bugs() const
205 * Check whether the current trace has triggered an assertion which should halt
208 * @return True, if the execution should be aborted; false otherwise
210 bool ModelExecution::has_asserted() const
212 return priv->asserted;
216 * Trigger a trace assertion which should cause this execution to be halted.
217 * This can be due to a detected bug or due to an infeasibility that should
220 void ModelExecution::set_assert()
222 priv->asserted = true;
226 * Check if we are in a deadlock. Should only be called at the end of an
227 * execution, although it should not give false positives in the middle of an
228 * execution (there should be some ENABLED thread).
230 * @return True if program is in a deadlock; false otherwise
232 bool ModelExecution::is_deadlocked() const
234 bool blocking_threads = false;
235 for (unsigned int i = 0;i < get_num_threads();i++) {
236 thread_id_t tid = int_to_id(i);
239 Thread *t = get_thread(tid);
240 if (!t->is_model_thread() && t->get_pending())
241 blocking_threads = true;
243 return blocking_threads;
247 * Check if this is a complete execution. That is, have all thread completed
248 * execution (rather than exiting because sleep sets have forced a redundant
251 * @return True if the execution is complete.
253 bool ModelExecution::is_complete_execution() const
255 for (unsigned int i = 0;i < get_num_threads();i++)
256 if (is_enabled(int_to_id(i)))
261 ModelAction * ModelExecution::convertNonAtomicStore(void * location) {
262 uint64_t value = *((const uint64_t *) location);
263 modelclock_t storeclock;
264 thread_id_t storethread;
265 getStoreThreadAndClock(location, &storethread, &storeclock);
266 setAtomicStoreFlag(location);
267 ModelAction * act = new ModelAction(NONATOMIC_WRITE, memory_order_relaxed, location, value, get_thread(storethread));
268 act->set_seq_number(storeclock);
269 add_normal_write_to_lists(act);
270 add_write_to_lists(act);
271 w_modification_order(act);
272 // model->get_history()->process_action(act, act->get_tid());
277 * Processes a read model action.
278 * @param curr is the read model action to process.
279 * @param rf_set is the set of model actions we can possibly read from
280 * @return True if processing this read updates the mo_graph.
282 bool ModelExecution::process_read(ModelAction *curr, SnapVector<ModelAction *> * rf_set)
284 SnapVector<const ModelAction *> * priorset = new SnapVector<const ModelAction *>();
285 bool hasnonatomicstore = hasNonAtomicStore(curr->get_location());
286 if (hasnonatomicstore) {
287 ModelAction * nonatomicstore = convertNonAtomicStore(curr->get_location());
288 rf_set->push_back(nonatomicstore);
291 // Remove writes that violate read modification order
293 for (uint i = 0; i < rf_set->size(); i++) {
294 ModelAction * rf = (*rf_set)[i];
295 if (!r_modification_order(curr, rf, NULL, NULL, true)) {
296 (*rf_set)[i] = rf_set->back();
302 int index = fuzzer->selectWrite(curr, rf_set);
303 if (index == -1)// no feasible write exists
306 ModelAction *rf = (*rf_set)[index];
309 bool canprune = false;
310 if (r_modification_order(curr, rf, priorset, &canprune)) {
311 for(unsigned int i=0;i<priorset->size();i++) {
312 mo_graph->addEdge((*priorset)[i], rf);
315 get_thread(curr)->set_return_value(curr->get_return_value());
317 if (canprune && curr->get_type() == ATOMIC_READ) {
318 int tid = id_to_int(curr->get_tid());
319 (*obj_thrd_map.get(curr->get_location()))[tid].pop_back();
324 (*rf_set)[index] = rf_set->back();
330 * Processes a lock, trylock, or unlock model action. @param curr is
331 * the read model action to process.
333 * The try lock operation checks whether the lock is taken. If not,
334 * it falls to the normal lock operation case. If so, it returns
337 * The lock operation has already been checked that it is enabled, so
338 * it just grabs the lock and synchronizes with the previous unlock.
340 * The unlock operation has to re-enable all of the threads that are
341 * waiting on the lock.
343 * @return True if synchronization was updated; false otherwise
345 bool ModelExecution::process_mutex(ModelAction *curr)
347 cdsc::mutex *mutex = curr->get_mutex();
348 struct cdsc::mutex_state *state = NULL;
351 state = mutex->get_state();
353 switch (curr->get_type()) {
354 case ATOMIC_TRYLOCK: {
355 bool success = !state->locked;
356 curr->set_try_lock(success);
358 get_thread(curr)->set_return_value(0);
361 get_thread(curr)->set_return_value(1);
363 //otherwise fall into the lock case
365 //TODO: FIND SOME BETTER WAY TO CHECK LOCK INITIALIZED OR NOT
366 //if (curr->get_cv()->getClock(state->alloc_tid) <= state->alloc_clock)
367 // assert_bug("Lock access before initialization");
368 state->locked = get_thread(curr);
369 ModelAction *unlock = get_last_unlock(curr);
370 //synchronize with the previous unlock statement
371 if (unlock != NULL) {
372 synchronize(unlock, curr);
378 /* wake up the other threads */
379 for (unsigned int i = 0;i < get_num_threads();i++) {
380 Thread *t = get_thread(int_to_id(i));
381 Thread *curr_thrd = get_thread(curr);
382 if (t->waiting_on() == curr_thrd && t->get_pending()->is_lock())
386 /* unlock the lock - after checking who was waiting on it */
387 state->locked = NULL;
389 if (fuzzer->shouldWait(curr)) {
390 /* disable this thread */
391 get_safe_ptr_action(&condvar_waiters_map, curr->get_location())->push_back(curr);
392 scheduler->sleep(get_thread(curr));
397 case ATOMIC_TIMEDWAIT:
398 case ATOMIC_UNLOCK: {
399 //TODO: FIX WAIT SITUATION...WAITS CAN SPURIOUSLY FAIL...TIMED WAITS SHOULD PROBABLY JUST BE THE SAME AS NORMAL WAITS...THINK ABOUT PROBABILITIES THOUGH....AS IN TIMED WAIT MUST FAIL TO GUARANTEE PROGRESS...NORMAL WAIT MAY FAIL...SO NEED NORMAL WAIT TO WORK CORRECTLY IN THE CASE IT SPURIOUSLY FAILS AND IN THE CASE IT DOESN'T... TIMED WAITS MUST EVENMTUALLY RELEASE...
401 /* wake up the other threads */
402 for (unsigned int i = 0;i < get_num_threads();i++) {
403 Thread *t = get_thread(int_to_id(i));
404 Thread *curr_thrd = get_thread(curr);
405 if (t->waiting_on() == curr_thrd && t->get_pending()->is_lock())
409 /* unlock the lock - after checking who was waiting on it */
410 state->locked = NULL;
413 case ATOMIC_NOTIFY_ALL: {
414 action_list_t *waiters = get_safe_ptr_action(&condvar_waiters_map, curr->get_location());
415 //activate all the waiting threads
416 for (sllnode<ModelAction *> * rit = waiters->begin();rit != NULL;rit=rit->getNext()) {
417 scheduler->wake(get_thread(rit->getVal()));
422 case ATOMIC_NOTIFY_ONE: {
423 action_list_t *waiters = get_safe_ptr_action(&condvar_waiters_map, curr->get_location());
424 if (waiters->size() != 0) {
425 Thread * thread = fuzzer->selectNotify(waiters);
426 scheduler->wake(thread);
438 * Process a write ModelAction
439 * @param curr The ModelAction to process
440 * @return True if the mo_graph was updated or promises were resolved
442 void ModelExecution::process_write(ModelAction *curr)
444 w_modification_order(curr);
445 get_thread(curr)->set_return_value(VALUE_NONE);
449 * Process a fence ModelAction
450 * @param curr The ModelAction to process
451 * @return True if synchronization was updated
453 bool ModelExecution::process_fence(ModelAction *curr)
456 * fence-relaxed: no-op
457 * fence-release: only log the occurence (not in this function), for
458 * use in later synchronization
459 * fence-acquire (this function): search for hypothetical release
461 * fence-seq-cst: MO constraints formed in {r,w}_modification_order
463 bool updated = false;
464 if (curr->is_acquire()) {
465 action_list_t *list = &action_trace;
466 sllnode<ModelAction *> * rit;
467 /* Find X : is_read(X) && X --sb-> curr */
468 for (rit = list->end();rit != NULL;rit=rit->getPrev()) {
469 ModelAction *act = rit->getVal();
472 if (act->get_tid() != curr->get_tid())
474 /* Stop at the beginning of the thread */
475 if (act->is_thread_start())
477 /* Stop once we reach a prior fence-acquire */
478 if (act->is_fence() && act->is_acquire())
482 /* read-acquire will find its own release sequences */
483 if (act->is_acquire())
486 /* Establish hypothetical release sequences */
487 ClockVector *cv = get_hb_from_write(act->get_reads_from());
488 if (cv != NULL && curr->get_cv()->merge(cv))
496 * @brief Process the current action for thread-related activity
498 * Performs current-action processing for a THREAD_* ModelAction. Proccesses
499 * may include setting Thread status, completing THREAD_FINISH/THREAD_JOIN
500 * synchronization, etc. This function is a no-op for non-THREAD actions
501 * (e.g., ATOMIC_{READ,WRITE,RMW,LOCK}, etc.)
503 * @param curr The current action
504 * @return True if synchronization was updated or a thread completed
506 void ModelExecution::process_thread_action(ModelAction *curr)
508 switch (curr->get_type()) {
509 case THREAD_CREATE: {
510 thrd_t *thrd = (thrd_t *)curr->get_location();
511 struct thread_params *params = (struct thread_params *)curr->get_value();
512 Thread *th = new Thread(get_next_id(), thrd, params->func, params->arg, get_thread(curr));
513 curr->set_thread_operand(th);
515 th->set_creation(curr);
518 case PTHREAD_CREATE: {
519 (*(uint32_t *)curr->get_location()) = pthread_counter++;
521 struct pthread_params *params = (struct pthread_params *)curr->get_value();
522 Thread *th = new Thread(get_next_id(), NULL, params->func, params->arg, get_thread(curr));
523 curr->set_thread_operand(th);
525 th->set_creation(curr);
527 if ( pthread_map.size() < pthread_counter )
528 pthread_map.resize( pthread_counter );
529 pthread_map[ pthread_counter-1 ] = th;
534 Thread *blocking = curr->get_thread_operand();
535 ModelAction *act = get_last_action(blocking->get_id());
536 synchronize(act, curr);
540 Thread *blocking = curr->get_thread_operand();
541 ModelAction *act = get_last_action(blocking->get_id());
542 synchronize(act, curr);
543 break; // WL: to be add (modified)
546 case THREADONLY_FINISH:
547 case THREAD_FINISH: {
548 Thread *th = get_thread(curr);
549 if (curr->get_type() == THREAD_FINISH &&
550 th == model->getInitThread()) {
556 /* Wake up any joining threads */
557 for (unsigned int i = 0;i < get_num_threads();i++) {
558 Thread *waiting = get_thread(int_to_id(i));
559 if (waiting->waiting_on() == th &&
560 waiting->get_pending()->is_thread_join())
561 scheduler->wake(waiting);
570 Thread *th = get_thread(curr);
571 th->set_pending(curr);
572 scheduler->add_sleep(th);
581 * Initialize the current action by performing one or more of the following
582 * actions, as appropriate: merging RMWR and RMWC/RMW actions,
583 * manipulating backtracking sets, allocating and
584 * initializing clock vectors, and computing the promises to fulfill.
586 * @param curr The current action, as passed from the user context; may be
587 * freed/invalidated after the execution of this function, with a different
588 * action "returned" its place (pass-by-reference)
589 * @return True if curr is a newly-explored action; false otherwise
591 bool ModelExecution::initialize_curr_action(ModelAction **curr)
593 if ((*curr)->is_rmwc() || (*curr)->is_rmw()) {
594 ModelAction *newcurr = process_rmw(*curr);
600 ModelAction *newcurr = *curr;
602 newcurr->set_seq_number(get_next_seq_num());
603 /* Always compute new clock vector */
604 newcurr->create_cv(get_parent_action(newcurr->get_tid()));
606 /* Assign most recent release fence */
607 newcurr->set_last_fence_release(get_last_fence_release(newcurr->get_tid()));
609 return true; /* This was a new ModelAction */
614 * @brief Establish reads-from relation between two actions
616 * Perform basic operations involved with establishing a concrete rf relation,
617 * including setting the ModelAction data and checking for release sequences.
619 * @param act The action that is reading (must be a read)
620 * @param rf The action from which we are reading (must be a write)
622 * @return True if this read established synchronization
625 void ModelExecution::read_from(ModelAction *act, ModelAction *rf)
628 ASSERT(rf->is_write());
630 act->set_read_from(rf);
631 if (act->is_acquire()) {
632 ClockVector *cv = get_hb_from_write(rf);
635 act->get_cv()->merge(cv);
640 * @brief Synchronizes two actions
642 * When A synchronizes with B (or A --sw-> B), B inherits A's clock vector.
643 * This function performs the synchronization as well as providing other hooks
644 * for other checks along with synchronization.
646 * @param first The left-hand side of the synchronizes-with relation
647 * @param second The right-hand side of the synchronizes-with relation
648 * @return True if the synchronization was successful (i.e., was consistent
649 * with the execution order); false otherwise
651 bool ModelExecution::synchronize(const ModelAction *first, ModelAction *second)
653 if (*second < *first) {
654 ASSERT(0); //This should not happend
657 return second->synchronize_with(first);
661 * @brief Check whether a model action is enabled.
663 * Checks whether an operation would be successful (i.e., is a lock already
664 * locked, or is the joined thread already complete).
666 * For yield-blocking, yields are never enabled.
668 * @param curr is the ModelAction to check whether it is enabled.
669 * @return a bool that indicates whether the action is enabled.
671 bool ModelExecution::check_action_enabled(ModelAction *curr) {
672 if (curr->is_lock()) {
673 cdsc::mutex *lock = curr->get_mutex();
674 struct cdsc::mutex_state *state = lock->get_state();
677 } else if (curr->is_thread_join()) {
678 Thread *blocking = curr->get_thread_operand();
679 if (!blocking->is_complete()) {
682 } else if (curr->is_sleep()) {
683 if (!fuzzer->shouldSleep(curr))
691 * This is the heart of the model checker routine. It performs model-checking
692 * actions corresponding to a given "current action." Among other processes, it
693 * calculates reads-from relationships, updates synchronization clock vectors,
694 * forms a memory_order constraints graph, and handles replay/backtrack
695 * execution when running permutations of previously-observed executions.
697 * @param curr The current action to process
698 * @return The ModelAction that is actually executed; may be different than
701 ModelAction * ModelExecution::check_current_action(ModelAction *curr)
704 bool second_part_of_rmw = curr->is_rmwc() || curr->is_rmw();
705 bool newly_explored = initialize_curr_action(&curr);
709 wake_up_sleeping_actions(curr);
711 /* Add uninitialized actions to lists */
712 if (!second_part_of_rmw)
713 add_uninit_action_to_lists(curr);
715 SnapVector<ModelAction *> * rf_set = NULL;
716 /* Build may_read_from set for newly-created actions */
717 if (newly_explored && curr->is_read())
718 rf_set = build_may_read_from(curr);
720 if (curr->is_read() && !second_part_of_rmw) {
721 process_read(curr, rf_set);
724 /* bool success = process_read(curr, rf_set);
727 return curr; // Do not add action to lists
730 ASSERT(rf_set == NULL);
732 /* Add the action to lists */
733 if (!second_part_of_rmw)
734 add_action_to_lists(curr);
736 if (curr->is_write())
737 add_write_to_lists(curr);
739 process_thread_action(curr);
741 if (curr->is_write())
744 if (curr->is_fence())
747 if (curr->is_mutex_op())
753 /** Close out a RMWR by converting previous RMWR into a RMW or READ. */
754 ModelAction * ModelExecution::process_rmw(ModelAction *act) {
755 ModelAction *lastread = get_last_action(act->get_tid());
756 lastread->process_rmw(act);
758 mo_graph->addRMWEdge(lastread->get_reads_from(), lastread);
764 * @brief Updates the mo_graph with the constraints imposed from the current
767 * Basic idea is the following: Go through each other thread and find
768 * the last action that happened before our read. Two cases:
770 * -# The action is a write: that write must either occur before
771 * the write we read from or be the write we read from.
772 * -# The action is a read: the write that that action read from
773 * must occur before the write we read from or be the same write.
775 * @param curr The current action. Must be a read.
776 * @param rf The ModelAction or Promise that curr reads from. Must be a write.
777 * @param check_only If true, then only check whether the current action satisfies
778 * read modification order or not, without modifiying priorset and canprune.
780 * @return True if modification order edges were added; false otherwise
783 bool ModelExecution::r_modification_order(ModelAction *curr, const ModelAction *rf,
784 SnapVector<const ModelAction *> * priorset, bool * canprune, bool check_only)
786 SnapVector<action_list_t> *thrd_lists = obj_thrd_map.get(curr->get_location());
788 ASSERT(curr->is_read());
790 /* Last SC fence in the current thread */
791 ModelAction *last_sc_fence_local = get_last_seq_cst_fence(curr->get_tid(), NULL);
793 int tid = curr->get_tid();
794 ModelAction *prev_same_thread = NULL;
795 /* Iterate over all threads */
796 for (i = 0;i < thrd_lists->size();i++, tid = (((unsigned int)(tid+1)) == thrd_lists->size()) ? 0 : tid + 1) {
797 /* Last SC fence in thread tid */
798 ModelAction *last_sc_fence_thread_local = NULL;
800 last_sc_fence_thread_local = get_last_seq_cst_fence(int_to_id(tid), NULL);
802 /* Last SC fence in thread tid, before last SC fence in current thread */
803 ModelAction *last_sc_fence_thread_before = NULL;
804 if (last_sc_fence_local)
805 last_sc_fence_thread_before = get_last_seq_cst_fence(int_to_id(tid), last_sc_fence_local);
807 //Only need to iterate if either hb has changed for thread in question or SC fence after last operation...
808 if (prev_same_thread != NULL &&
809 (prev_same_thread->get_cv()->getClock(tid) == curr->get_cv()->getClock(tid)) &&
810 (last_sc_fence_thread_local == NULL || *last_sc_fence_thread_local < *prev_same_thread)) {
814 /* Iterate over actions in thread, starting from most recent */
815 action_list_t *list = &(*thrd_lists)[tid];
816 sllnode<ModelAction *> * rit;
817 for (rit = list->end();rit != NULL;rit=rit->getPrev()) {
818 ModelAction *act = rit->getVal();
823 /* Don't want to add reflexive edges on 'rf' */
824 if (act->equals(rf)) {
825 if (act->happens_before(curr))
831 if (act->is_write()) {
832 /* C++, Section 29.3 statement 5 */
833 if (curr->is_seqcst() && last_sc_fence_thread_local &&
834 *act < *last_sc_fence_thread_local) {
835 if (mo_graph->checkReachable(rf, act))
838 priorset->push_back(act);
841 /* C++, Section 29.3 statement 4 */
842 else if (act->is_seqcst() && last_sc_fence_local &&
843 *act < *last_sc_fence_local) {
844 if (mo_graph->checkReachable(rf, act))
847 priorset->push_back(act);
850 /* C++, Section 29.3 statement 6 */
851 else if (last_sc_fence_thread_before &&
852 *act < *last_sc_fence_thread_before) {
853 if (mo_graph->checkReachable(rf, act))
856 priorset->push_back(act);
862 * Include at most one act per-thread that "happens
865 if (act->happens_before(curr)) {
867 if (last_sc_fence_local == NULL ||
868 (*last_sc_fence_local < *act)) {
869 prev_same_thread = act;
872 if (act->is_write()) {
873 if (mo_graph->checkReachable(rf, act))
876 priorset->push_back(act);
878 const ModelAction *prevrf = act->get_reads_from();
879 if (!prevrf->equals(rf)) {
880 if (mo_graph->checkReachable(rf, prevrf))
883 priorset->push_back(prevrf);
885 if (act->get_tid() == curr->get_tid()) {
886 //Can prune curr from obj list
900 * Updates the mo_graph with the constraints imposed from the current write.
902 * Basic idea is the following: Go through each other thread and find
903 * the lastest action that happened before our write. Two cases:
905 * (1) The action is a write => that write must occur before
908 * (2) The action is a read => the write that that action read from
909 * must occur before the current write.
911 * This method also handles two other issues:
913 * (I) Sequential Consistency: Making sure that if the current write is
914 * seq_cst, that it occurs after the previous seq_cst write.
916 * (II) Sending the write back to non-synchronizing reads.
918 * @param curr The current action. Must be a write.
919 * @param send_fv A vector for stashing reads to which we may pass our future
920 * value. If NULL, then don't record any future values.
921 * @return True if modification order edges were added; false otherwise
923 void ModelExecution::w_modification_order(ModelAction *curr)
925 SnapVector<action_list_t> *thrd_lists = obj_thrd_map.get(curr->get_location());
927 ASSERT(curr->is_write());
929 SnapList<ModelAction *> edgeset;
931 if (curr->is_seqcst()) {
932 /* We have to at least see the last sequentially consistent write,
933 so we are initialized. */
934 ModelAction *last_seq_cst = get_last_seq_cst_write(curr);
935 if (last_seq_cst != NULL) {
936 edgeset.push_back(last_seq_cst);
938 //update map for next query
939 obj_last_sc_map.put(curr->get_location(), curr);
942 /* Last SC fence in the current thread */
943 ModelAction *last_sc_fence_local = get_last_seq_cst_fence(curr->get_tid(), NULL);
945 /* Iterate over all threads */
946 for (i = 0;i < thrd_lists->size();i++) {
947 /* Last SC fence in thread i, before last SC fence in current thread */
948 ModelAction *last_sc_fence_thread_before = NULL;
949 if (last_sc_fence_local && int_to_id((int)i) != curr->get_tid())
950 last_sc_fence_thread_before = get_last_seq_cst_fence(int_to_id(i), last_sc_fence_local);
952 /* Iterate over actions in thread, starting from most recent */
953 action_list_t *list = &(*thrd_lists)[i];
954 sllnode<ModelAction*>* rit;
955 for (rit = list->end();rit != NULL;rit=rit->getPrev()) {
956 ModelAction *act = rit->getVal();
959 * 1) If RMW and it actually read from something, then we
960 * already have all relevant edges, so just skip to next
963 * 2) If RMW and it didn't read from anything, we should
964 * whatever edge we can get to speed up convergence.
966 * 3) If normal write, we need to look at earlier actions, so
967 * continue processing list.
969 if (curr->is_rmw()) {
970 if (curr->get_reads_from() != NULL)
978 /* C++, Section 29.3 statement 7 */
979 if (last_sc_fence_thread_before && act->is_write() &&
980 *act < *last_sc_fence_thread_before) {
981 edgeset.push_back(act);
986 * Include at most one act per-thread that "happens
989 if (act->happens_before(curr)) {
991 * Note: if act is RMW, just add edge:
993 * The following edge should be handled elsewhere:
994 * readfrom(act) --mo--> act
997 edgeset.push_back(act);
998 else if (act->is_read()) {
999 //if previous read accessed a null, just keep going
1000 edgeset.push_back(act->get_reads_from());
1006 mo_graph->addEdges(&edgeset, curr);
1011 * Arbitrary reads from the future are not allowed. Section 29.3 part 9 places
1012 * some constraints. This method checks one the following constraint (others
1013 * require compiler support):
1015 * If X --hb-> Y --mo-> Z, then X should not read from Z.
1016 * If X --hb-> Y, A --rf-> Y, and A --mo-> Z, then X should not read from Z.
1018 bool ModelExecution::mo_may_allow(const ModelAction *writer, const ModelAction *reader)
1020 SnapVector<action_list_t> *thrd_lists = obj_thrd_map.get(reader->get_location());
1022 /* Iterate over all threads */
1023 for (i = 0;i < thrd_lists->size();i++) {
1024 const ModelAction *write_after_read = NULL;
1026 /* Iterate over actions in thread, starting from most recent */
1027 action_list_t *list = &(*thrd_lists)[i];
1028 sllnode<ModelAction *>* rit;
1029 for (rit = list->end();rit != NULL;rit=rit->getPrev()) {
1030 ModelAction *act = rit->getVal();
1032 /* Don't disallow due to act == reader */
1033 if (!reader->happens_before(act) || reader == act)
1035 else if (act->is_write())
1036 write_after_read = act;
1037 else if (act->is_read() && act->get_reads_from() != NULL)
1038 write_after_read = act->get_reads_from();
1041 if (write_after_read && write_after_read != writer && mo_graph->checkReachable(write_after_read, writer))
1048 * Computes the clock vector that happens before propagates from this write.
1050 * @param rf The action that might be part of a release sequence. Must be a
1052 * @return ClockVector of happens before relation.
1055 ClockVector * ModelExecution::get_hb_from_write(ModelAction *rf) const {
1056 SnapVector<ModelAction *> * processset = NULL;
1057 for ( ;rf != NULL;rf = rf->get_reads_from()) {
1058 ASSERT(rf->is_write());
1059 if (!rf->is_rmw() || (rf->is_acquire() && rf->is_release()) || rf->get_rfcv() != NULL)
1061 if (processset == NULL)
1062 processset = new SnapVector<ModelAction *>();
1063 processset->push_back(rf);
1066 int i = (processset == NULL) ? 0 : processset->size();
1068 ClockVector * vec = NULL;
1070 if (rf->get_rfcv() != NULL) {
1071 vec = rf->get_rfcv();
1072 } else if (rf->is_acquire() && rf->is_release()) {
1074 } else if (rf->is_release() && !rf->is_rmw()) {
1076 } else if (rf->is_release()) {
1077 //have rmw that is release and doesn't have a rfcv
1078 (vec = new ClockVector(vec, NULL))->merge(rf->get_cv());
1081 //operation that isn't release
1082 if (rf->get_last_fence_release()) {
1084 vec = rf->get_last_fence_release()->get_cv();
1086 (vec=new ClockVector(vec, NULL))->merge(rf->get_last_fence_release()->get_cv());
1092 rf = (*processset)[i];
1096 if (processset != NULL)
1102 * Performs various bookkeeping operations for the current ModelAction when it is
1103 * the first atomic action occurred at its memory location.
1105 * For instance, adds uninitialized action to the per-object, per-thread action vector
1106 * and to the action trace list of all thread actions.
1108 * @param act is the ModelAction to process.
1110 void ModelExecution::add_uninit_action_to_lists(ModelAction *act)
1112 int tid = id_to_int(act->get_tid());
1113 ModelAction *uninit = NULL;
1115 action_list_t *list = get_safe_ptr_action(&obj_map, act->get_location());
1116 if (list->empty() && act->is_atomic_var()) {
1117 uninit = get_uninitialized_action(act);
1118 uninit_id = id_to_int(uninit->get_tid());
1119 list->push_front(uninit);
1120 SnapVector<action_list_t> *vec = get_safe_ptr_vect_action(&obj_wr_thrd_map, act->get_location());
1121 if ((int)vec->size() <= uninit_id) {
1122 int oldsize = (int) vec->size();
1123 vec->resize(uninit_id + 1);
1124 for(int i = oldsize;i < uninit_id + 1;i++) {
1125 new (&(*vec)[i]) action_list_t();
1128 (*vec)[uninit_id].push_front(uninit);
1131 // Update action trace, a total order of all actions
1133 action_trace.push_front(uninit);
1135 // Update obj_thrd_map, a per location, per thread, order of actions
1136 SnapVector<action_list_t> *vec = get_safe_ptr_vect_action(&obj_thrd_map, act->get_location());
1137 if ((int)vec->size() <= tid) {
1138 uint oldsize = vec->size();
1139 vec->resize(priv->next_thread_id);
1140 for(uint i = oldsize;i < priv->next_thread_id;i++)
1141 new (&(*vec)[i]) action_list_t();
1144 (*vec)[uninit_id].push_front(uninit);
1146 // Update thrd_last_action, the last action taken by each thrad
1147 if ((int)thrd_last_action.size() <= tid)
1148 thrd_last_action.resize(get_num_threads());
1150 thrd_last_action[uninit_id] = uninit;
1155 * Performs various bookkeeping operations for the current ModelAction. For
1156 * instance, adds action to the per-object, per-thread action vector and to the
1157 * action trace list of all thread actions.
1159 * @param act is the ModelAction to add.
1161 void ModelExecution::add_action_to_lists(ModelAction *act)
1163 int tid = id_to_int(act->get_tid());
1164 action_list_t *list = get_safe_ptr_action(&obj_map, act->get_location());
1165 list->push_back(act);
1167 // Update action trace, a total order of all actions
1168 action_trace.push_back(act);
1170 // Update obj_thrd_map, a per location, per thread, order of actions
1171 SnapVector<action_list_t> *vec = get_safe_ptr_vect_action(&obj_thrd_map, act->get_location());
1172 if ((int)vec->size() <= tid) {
1173 uint oldsize = vec->size();
1174 vec->resize(priv->next_thread_id);
1175 for(uint i = oldsize;i < priv->next_thread_id;i++)
1176 new (&(*vec)[i]) action_list_t();
1178 (*vec)[tid].push_back(act);
1180 // Update thrd_last_action, the last action taken by each thrad
1181 if ((int)thrd_last_action.size() <= tid)
1182 thrd_last_action.resize(get_num_threads());
1183 thrd_last_action[tid] = act;
1185 // Update thrd_last_fence_release, the last release fence taken by each thread
1186 if (act->is_fence() && act->is_release()) {
1187 if ((int)thrd_last_fence_release.size() <= tid)
1188 thrd_last_fence_release.resize(get_num_threads());
1189 thrd_last_fence_release[tid] = act;
1192 if (act->is_wait()) {
1193 void *mutex_loc = (void *) act->get_value();
1194 get_safe_ptr_action(&obj_map, mutex_loc)->push_back(act);
1196 SnapVector<action_list_t> *vec = get_safe_ptr_vect_action(&obj_thrd_map, mutex_loc);
1197 if ((int)vec->size() <= tid) {
1198 uint oldsize = vec->size();
1199 vec->resize(priv->next_thread_id);
1200 for(uint i = oldsize;i < priv->next_thread_id;i++)
1201 new (&(*vec)[i]) action_list_t();
1203 (*vec)[tid].push_back(act);
1207 void insertIntoActionList(action_list_t *list, ModelAction *act) {
1208 sllnode<ModelAction*> * rit = list->end();
1209 modelclock_t next_seq = act->get_seq_number();
1210 if (rit == NULL || (rit->getVal()->get_seq_number() == next_seq))
1211 list->push_back(act);
1213 for(;rit != NULL;rit=rit->getPrev()) {
1214 if (rit->getVal()->get_seq_number() == next_seq) {
1215 list->insertAfter(rit, act);
1222 void insertIntoActionListAndSetCV(action_list_t *list, ModelAction *act) {
1223 sllnode<ModelAction*> * rit = list->end();
1224 modelclock_t next_seq = act->get_seq_number();
1226 act->create_cv(NULL);
1227 } else if (rit->getVal()->get_seq_number() == next_seq) {
1228 act->create_cv(rit->getVal());
1229 list->push_back(act);
1231 for(;rit != NULL;rit=rit->getPrev()) {
1232 if (rit->getVal()->get_seq_number() == next_seq) {
1233 act->create_cv(rit->getVal());
1234 list->insertAfter(rit, act);
1242 * Performs various bookkeeping operations for a normal write. The
1243 * complication is that we are typically inserting a normal write
1244 * lazily, so we need to insert it into the middle of lists.
1246 * @param act is the ModelAction to add.
1249 void ModelExecution::add_normal_write_to_lists(ModelAction *act)
1251 int tid = id_to_int(act->get_tid());
1252 insertIntoActionListAndSetCV(&action_trace, act);
1254 action_list_t *list = get_safe_ptr_action(&obj_map, act->get_location());
1255 insertIntoActionList(list, act);
1257 // Update obj_thrd_map, a per location, per thread, order of actions
1258 SnapVector<action_list_t> *vec = get_safe_ptr_vect_action(&obj_thrd_map, act->get_location());
1259 if (tid >= (int)vec->size()) {
1260 uint oldsize =vec->size();
1261 vec->resize(priv->next_thread_id);
1262 for(uint i=oldsize;i<priv->next_thread_id;i++)
1263 new (&(*vec)[i]) action_list_t();
1265 insertIntoActionList(&(*vec)[tid],act);
1267 // Update thrd_last_action, the last action taken by each thrad
1268 if (thrd_last_action[tid]->get_seq_number() == act->get_seq_number())
1269 thrd_last_action[tid] = act;
1273 void ModelExecution::add_write_to_lists(ModelAction *write) {
1274 SnapVector<action_list_t> *vec = get_safe_ptr_vect_action(&obj_wr_thrd_map, write->get_location());
1275 int tid = id_to_int(write->get_tid());
1276 if (tid >= (int)vec->size()) {
1277 uint oldsize =vec->size();
1278 vec->resize(priv->next_thread_id);
1279 for(uint i=oldsize;i<priv->next_thread_id;i++)
1280 new (&(*vec)[i]) action_list_t();
1282 (*vec)[tid].push_back(write);
1286 * @brief Get the last action performed by a particular Thread
1287 * @param tid The thread ID of the Thread in question
1288 * @return The last action in the thread
1290 ModelAction * ModelExecution::get_last_action(thread_id_t tid) const
1292 int threadid = id_to_int(tid);
1293 if (threadid < (int)thrd_last_action.size())
1294 return thrd_last_action[id_to_int(tid)];
1300 * @brief Get the last fence release performed by a particular Thread
1301 * @param tid The thread ID of the Thread in question
1302 * @return The last fence release in the thread, if one exists; NULL otherwise
1304 ModelAction * ModelExecution::get_last_fence_release(thread_id_t tid) const
1306 int threadid = id_to_int(tid);
1307 if (threadid < (int)thrd_last_fence_release.size())
1308 return thrd_last_fence_release[id_to_int(tid)];
1314 * Gets the last memory_order_seq_cst write (in the total global sequence)
1315 * performed on a particular object (i.e., memory location), not including the
1317 * @param curr The current ModelAction; also denotes the object location to
1319 * @return The last seq_cst write
1321 ModelAction * ModelExecution::get_last_seq_cst_write(ModelAction *curr) const
1323 void *location = curr->get_location();
1324 return obj_last_sc_map.get(location);
1328 * Gets the last memory_order_seq_cst fence (in the total global sequence)
1329 * performed in a particular thread, prior to a particular fence.
1330 * @param tid The ID of the thread to check
1331 * @param before_fence The fence from which to begin the search; if NULL, then
1332 * search for the most recent fence in the thread.
1333 * @return The last prior seq_cst fence in the thread, if exists; otherwise, NULL
1335 ModelAction * ModelExecution::get_last_seq_cst_fence(thread_id_t tid, const ModelAction *before_fence) const
1337 /* All fences should have location FENCE_LOCATION */
1338 action_list_t *list = obj_map.get(FENCE_LOCATION);
1343 sllnode<ModelAction*>* rit = list->end();
1346 for (;rit != NULL;rit=rit->getPrev())
1347 if (rit->getVal() == before_fence)
1350 ASSERT(rit->getVal() == before_fence);
1354 for (;rit != NULL;rit=rit->getPrev()) {
1355 ModelAction *act = rit->getVal();
1356 if (act->is_fence() && (tid == act->get_tid()) && act->is_seqcst())
1363 * Gets the last unlock operation performed on a particular mutex (i.e., memory
1364 * location). This function identifies the mutex according to the current
1365 * action, which is presumed to perform on the same mutex.
1366 * @param curr The current ModelAction; also denotes the object location to
1368 * @return The last unlock operation
1370 ModelAction * ModelExecution::get_last_unlock(ModelAction *curr) const
1372 void *location = curr->get_location();
1374 action_list_t *list = obj_map.get(location);
1375 /* Find: max({i in dom(S) | isUnlock(t_i) && samevar(t_i, t)}) */
1376 sllnode<ModelAction*>* rit;
1377 for (rit = list->end();rit != NULL;rit=rit->getPrev())
1378 if (rit->getVal()->is_unlock() || rit->getVal()->is_wait())
1379 return rit->getVal();
1383 ModelAction * ModelExecution::get_parent_action(thread_id_t tid) const
1385 ModelAction *parent = get_last_action(tid);
1387 parent = get_thread(tid)->get_creation();
1392 * Returns the clock vector for a given thread.
1393 * @param tid The thread whose clock vector we want
1394 * @return Desired clock vector
1396 ClockVector * ModelExecution::get_cv(thread_id_t tid) const
1398 ModelAction *firstaction=get_parent_action(tid);
1399 return firstaction != NULL ? firstaction->get_cv() : NULL;
1402 bool valequals(uint64_t val1, uint64_t val2, int size) {
1405 return ((uint8_t)val1) == ((uint8_t)val2);
1407 return ((uint16_t)val1) == ((uint16_t)val2);
1409 return ((uint32_t)val1) == ((uint32_t)val2);
1419 * Build up an initial set of all past writes that this 'read' action may read
1420 * from, as well as any previously-observed future values that must still be valid.
1422 * @param curr is the current ModelAction that we are exploring; it must be a
1425 SnapVector<ModelAction *> * ModelExecution::build_may_read_from(ModelAction *curr)
1427 SnapVector<action_list_t> *thrd_lists = obj_wr_thrd_map.get(curr->get_location());
1429 ASSERT(curr->is_read());
1431 ModelAction *last_sc_write = NULL;
1433 if (curr->is_seqcst())
1434 last_sc_write = get_last_seq_cst_write(curr);
1436 SnapVector<ModelAction *> * rf_set = new SnapVector<ModelAction *>();
1438 /* Iterate over all threads */
1439 for (i = 0;i < thrd_lists->size();i++) {
1440 /* Iterate over actions in thread, starting from most recent */
1441 action_list_t *list = &(*thrd_lists)[i];
1442 sllnode<ModelAction *> * rit;
1443 for (rit = list->end();rit != NULL;rit=rit->getPrev()) {
1444 ModelAction *act = rit->getVal();
1449 /* Don't consider more than one seq_cst write if we are a seq_cst read. */
1450 bool allow_read = true;
1452 if (curr->is_seqcst() && (act->is_seqcst() || (last_sc_write != NULL && act->happens_before(last_sc_write))) && act != last_sc_write)
1455 /* Need to check whether we will have two RMW reading from the same value */
1456 if (curr->is_rmwr()) {
1457 /* It is okay if we have a failing CAS */
1458 if (!curr->is_rmwrcas() ||
1459 valequals(curr->get_value(), act->get_value(), curr->getSize())) {
1460 //Need to make sure we aren't the second RMW
1461 CycleNode * node = mo_graph->getNode_noCreate(act);
1462 if (node != NULL && node->getRMW() != NULL) {
1463 //we are the second RMW
1470 /* Only add feasible reads */
1471 rf_set->push_back(act);
1474 /* Include at most one act per-thread that "happens before" curr */
1475 if (act->happens_before(curr))
1480 if (DBG_ENABLED()) {
1481 model_print("Reached read action:\n");
1483 model_print("End printing read_from_past\n");
1489 * @brief Get an action representing an uninitialized atomic
1491 * This function may create a new one.
1493 * @param curr The current action, which prompts the creation of an UNINIT action
1494 * @return A pointer to the UNINIT ModelAction
1496 ModelAction * ModelExecution::get_uninitialized_action(ModelAction *curr) const
1498 ModelAction *act = curr->get_uninit_action();
1500 act = new ModelAction(ATOMIC_UNINIT, std::memory_order_relaxed, curr->get_location(), params->uninitvalue, model_thread);
1501 curr->set_uninit_action(act);
1503 act->create_cv(NULL);
1507 static void print_list(action_list_t *list)
1509 sllnode<ModelAction*> *it;
1511 model_print("------------------------------------------------------------------------------------\n");
1512 model_print("# t Action type MO Location Value Rf CV\n");
1513 model_print("------------------------------------------------------------------------------------\n");
1515 unsigned int hash = 0;
1517 for (it = list->begin();it != NULL;it=it->getNext()) {
1518 const ModelAction *act = it->getVal();
1519 if (act->get_seq_number() > 0)
1521 hash = hash^(hash<<3)^(it->getVal()->hash());
1523 model_print("HASH %u\n", hash);
1524 model_print("------------------------------------------------------------------------------------\n");
1527 #if SUPPORT_MOD_ORDER_DUMP
1528 void ModelExecution::dumpGraph(char *filename)
1531 sprintf(buffer, "%s.dot", filename);
1532 FILE *file = fopen(buffer, "w");
1533 fprintf(file, "digraph %s {\n", filename);
1534 mo_graph->dumpNodes(file);
1535 ModelAction **thread_array = (ModelAction **)model_calloc(1, sizeof(ModelAction *) * get_num_threads());
1537 for (sllnode<ModelAction*>* it = action_trace.begin();it != NULL;it=it->getNext()) {
1538 ModelAction *act = it->getVal();
1539 if (act->is_read()) {
1540 mo_graph->dot_print_node(file, act);
1541 mo_graph->dot_print_edge(file,
1542 act->get_reads_from(),
1544 "label=\"rf\", color=red, weight=2");
1546 if (thread_array[act->get_tid()]) {
1547 mo_graph->dot_print_edge(file,
1548 thread_array[id_to_int(act->get_tid())],
1550 "label=\"sb\", color=blue, weight=400");
1553 thread_array[act->get_tid()] = act;
1555 fprintf(file, "}\n");
1556 model_free(thread_array);
1561 /** @brief Prints an execution trace summary. */
1562 void ModelExecution::print_summary()
1564 #if SUPPORT_MOD_ORDER_DUMP
1565 char buffername[100];
1566 sprintf(buffername, "exec%04u", get_execution_number());
1567 mo_graph->dumpGraphToFile(buffername);
1568 sprintf(buffername, "graph%04u", get_execution_number());
1569 dumpGraph(buffername);
1572 model_print("Execution trace %d:", get_execution_number());
1573 if (scheduler->all_threads_sleeping())
1574 model_print(" SLEEP-SET REDUNDANT");
1575 if (have_bug_reports())
1576 model_print(" DETECTED BUG(S)");
1580 print_list(&action_trace);
1586 * Add a Thread to the system for the first time. Should only be called once
1588 * @param t The Thread to add
1590 void ModelExecution::add_thread(Thread *t)
1592 unsigned int i = id_to_int(t->get_id());
1593 if (i >= thread_map.size())
1594 thread_map.resize(i + 1);
1596 if (!t->is_model_thread())
1597 scheduler->add_thread(t);
1601 * @brief Get a Thread reference by its ID
1602 * @param tid The Thread's ID
1603 * @return A Thread reference
1605 Thread * ModelExecution::get_thread(thread_id_t tid) const
1607 unsigned int i = id_to_int(tid);
1608 if (i < thread_map.size())
1609 return thread_map[i];
1614 * @brief Get a reference to the Thread in which a ModelAction was executed
1615 * @param act The ModelAction
1616 * @return A Thread reference
1618 Thread * ModelExecution::get_thread(const ModelAction *act) const
1620 return get_thread(act->get_tid());
1624 * @brief Get a Thread reference by its pthread ID
1625 * @param index The pthread's ID
1626 * @return A Thread reference
1628 Thread * ModelExecution::get_pthread(pthread_t pid) {
1634 uint32_t thread_id = x.v;
1635 if (thread_id < pthread_counter + 1) return pthread_map[thread_id];
1640 * @brief Check if a Thread is currently enabled
1641 * @param t The Thread to check
1642 * @return True if the Thread is currently enabled
1644 bool ModelExecution::is_enabled(Thread *t) const
1646 return scheduler->is_enabled(t);
1650 * @brief Check if a Thread is currently enabled
1651 * @param tid The ID of the Thread to check
1652 * @return True if the Thread is currently enabled
1654 bool ModelExecution::is_enabled(thread_id_t tid) const
1656 return scheduler->is_enabled(tid);
1660 * @brief Select the next thread to execute based on the curren action
1662 * RMW actions occur in two parts, and we cannot split them. And THREAD_CREATE
1663 * actions should be followed by the execution of their child thread. In either
1664 * case, the current action should determine the next thread schedule.
1666 * @param curr The current action
1667 * @return The next thread to run, if the current action will determine this
1668 * selection; otherwise NULL
1670 Thread * ModelExecution::action_select_next_thread(const ModelAction *curr) const
1672 /* Do not split atomic RMW */
1673 if (curr->is_rmwr() && !paused_by_fuzzer(curr))
1674 return get_thread(curr);
1675 /* Follow CREATE with the created thread */
1676 /* which is not needed, because model.cc takes care of this */
1677 if (curr->get_type() == THREAD_CREATE)
1678 return curr->get_thread_operand();
1679 if (curr->get_type() == PTHREAD_CREATE) {
1680 return curr->get_thread_operand();
1685 /** @param act A read atomic action */
1686 bool ModelExecution::paused_by_fuzzer(const ModelAction * act) const
1688 ASSERT(act->is_read());
1690 // Actions paused by fuzzer have their sequence number reset to 0
1691 return act->get_seq_number() == 0;
1695 * Takes the next step in the execution, if possible.
1696 * @param curr The current step to take
1697 * @return Returns the next Thread to run, if any; NULL if this execution
1700 Thread * ModelExecution::take_step(ModelAction *curr)
1702 Thread *curr_thrd = get_thread(curr);
1703 ASSERT(curr_thrd->get_state() == THREAD_READY);
1705 ASSERT(check_action_enabled(curr)); /* May have side effects? */
1706 curr = check_current_action(curr);
1709 /* Process this action in ModelHistory for records */
1710 // model->get_history()->process_action( curr, curr->get_tid() );
1712 if (curr_thrd->is_blocked() || curr_thrd->is_complete())
1713 scheduler->remove_thread(curr_thrd);
1715 return action_select_next_thread(curr);
1718 Fuzzer * ModelExecution::getFuzzer() {