1 /* ---------------------------------------------------------------------------
2 * $Id: Schedule.c,v 1.198 2004/05/27 15:21:37 simonmar Exp $
4 * (c) The GHC Team, 1998-2003
8 * Different GHC ways use this scheduler quite differently (see comments below)
9 * Here is the global picture:
11 * WAY Name CPP flag What's it for
12 * --------------------------------------
13 * mp GUM PAR Parallel execution on a distrib. memory machine
14 * s SMP SMP Parallel execution on a shared memory machine
15 * mg GranSim GRAN Simulation of parallel execution
16 * md GUM/GdH DIST Distributed execution (based on GUM)
18 * --------------------------------------------------------------------------*/
21 * Version with support for distributed memory parallelism aka GUM (WAY=mp):
23 The main scheduling loop in GUM iterates until a finish message is received.
24 In that case a global flag @receivedFinish@ is set and this instance of
25 the RTS shuts down. See ghc/rts/parallel/HLComms.c:processMessages()
26 for the handling of incoming messages, such as PP_FINISH.
27 Note that in the parallel case we have a system manager that coordinates
28 different PEs, each of which are running one instance of the RTS.
29 See ghc/rts/parallel/SysMan.c for the main routine of the parallel program.
30 From this routine processes executing ghc/rts/Main.c are spawned. -- HWL
32 * Version with support for simulating parallel execution aka GranSim (WAY=mg):
34 The main scheduling code in GranSim is quite different from that in std
35 (concurrent) Haskell: while concurrent Haskell just iterates over the
36 threads in the runnable queue, GranSim is event driven, i.e. it iterates
37 over the events in the global event queue. -- HWL
40 #include "PosixSource.h"
47 #include "StgStartup.h"
49 #define COMPILING_SCHEDULER
51 #include "StgMiscClosures.h"
53 #include "Interpreter.h"
54 #include "Exception.h"
61 #include "ThreadLabels.h"
63 #include "Proftimer.h"
66 #if defined(GRAN) || defined(PAR)
67 # include "GranSimRts.h"
69 # include "ParallelRts.h"
70 # include "Parallel.h"
71 # include "ParallelDebug.h"
76 #include "Capability.h"
77 #include "OSThreads.h"
80 #ifdef HAVE_SYS_TYPES_H
81 #include <sys/types.h>
96 #define USED_IN_THREADED_RTS
98 #define USED_IN_THREADED_RTS STG_UNUSED
101 #ifdef RTS_SUPPORTS_THREADS
102 #define USED_WHEN_RTS_SUPPORTS_THREADS
104 #define USED_WHEN_RTS_SUPPORTS_THREADS STG_UNUSED
107 /* Main thread queue.
108 * Locks required: sched_mutex.
110 StgMainThread *main_threads = NULL;
113 * Locks required: sched_mutex.
117 StgTSO* ActiveTSO = NULL; /* for assigning system costs; GranSim-Light only */
118 /* rtsTime TimeOfNextEvent, EndOfTimeSlice; now in GranSim.c */
121 In GranSim we have a runnable and a blocked queue for each processor.
122 In order to minimise code changes new arrays run_queue_hds/tls
123 are created. run_queue_hd is then a short cut (macro) for
124 run_queue_hds[CurrentProc] (see GranSim.h).
127 StgTSO *run_queue_hds[MAX_PROC], *run_queue_tls[MAX_PROC];
128 StgTSO *blocked_queue_hds[MAX_PROC], *blocked_queue_tls[MAX_PROC];
129 StgTSO *ccalling_threadss[MAX_PROC];
130 /* We use the same global list of threads (all_threads) in GranSim as in
131 the std RTS (i.e. we are cheating). However, we don't use this list in
132 the GranSim specific code at the moment (so we are only potentially
137 StgTSO *run_queue_hd = NULL;
138 StgTSO *run_queue_tl = NULL;
139 StgTSO *blocked_queue_hd = NULL;
140 StgTSO *blocked_queue_tl = NULL;
141 StgTSO *sleeping_queue = NULL; /* perhaps replace with a hash table? */
145 /* Linked list of all threads.
146 * Used for detecting garbage collected threads.
148 StgTSO *all_threads = NULL;
150 /* When a thread performs a safe C call (_ccall_GC, using old
151 * terminology), it gets put on the suspended_ccalling_threads
152 * list. Used by the garbage collector.
154 static StgTSO *suspended_ccalling_threads;
156 static StgTSO *threadStackOverflow(StgTSO *tso);
158 /* KH: The following two flags are shared memory locations. There is no need
159 to lock them, since they are only unset at the end of a scheduler
163 /* flag set by signal handler to precipitate a context switch */
164 nat context_switch = 0;
166 /* if this flag is set as well, give up execution */
167 rtsBool interrupted = rtsFalse;
169 /* Next thread ID to allocate.
170 * Locks required: thread_id_mutex
172 static StgThreadID next_thread_id = 1;
175 * Pointers to the state of the current thread.
176 * Rule of thumb: if CurrentTSO != NULL, then we're running a Haskell
177 * thread. If CurrentTSO == NULL, then we're at the scheduler level.
180 /* The smallest stack size that makes any sense is:
181 * RESERVED_STACK_WORDS (so we can get back from the stack overflow)
182 * + sizeofW(StgStopFrame) (the stg_stop_thread_info frame)
183 * + 1 (the closure to enter)
185 * + 1 (spare slot req'd by stg_ap_v_ret)
187 * A thread with this stack will bomb immediately with a stack
188 * overflow, which will increase its stack size.
191 #define MIN_STACK_WORDS (RESERVED_STACK_WORDS + sizeofW(StgStopFrame) + 3)
198 /* This is used in `TSO.h' and gcc 2.96 insists that this variable actually
199 * exists - earlier gccs apparently didn't.
204 static rtsBool ready_to_gc;
207 * Set to TRUE when entering a shutdown state (via shutdownHaskellAndExit()) --
208 * in an MT setting, needed to signal that a worker thread shouldn't hang around
209 * in the scheduler when it is out of work.
211 static rtsBool shutting_down_scheduler = rtsFalse;
213 void addToBlockedQueue ( StgTSO *tso );
215 static void schedule ( StgMainThread *mainThread, Capability *initialCapability );
216 void interruptStgRts ( void );
218 static void detectBlackHoles ( void );
220 #if defined(RTS_SUPPORTS_THREADS)
221 /* ToDo: carefully document the invariants that go together
222 * with these synchronisation objects.
224 Mutex sched_mutex = INIT_MUTEX_VAR;
225 Mutex term_mutex = INIT_MUTEX_VAR;
227 #endif /* RTS_SUPPORTS_THREADS */
231 rtsTime TimeOfLastYield;
232 rtsBool emitSchedule = rtsTrue;
236 static char *whatNext_strs[] = {
246 StgTSO * createSparkThread(rtsSpark spark);
247 StgTSO * activateSpark (rtsSpark spark);
250 /* ----------------------------------------------------------------------------
252 * ------------------------------------------------------------------------- */
254 #if defined(RTS_SUPPORTS_THREADS)
255 static rtsBool startingWorkerThread = rtsFalse;
257 static void taskStart(void);
261 ACQUIRE_LOCK(&sched_mutex);
262 startingWorkerThread = rtsFalse;
264 RELEASE_LOCK(&sched_mutex);
268 startSchedulerTaskIfNecessary(void)
270 if(run_queue_hd != END_TSO_QUEUE
271 || blocked_queue_hd != END_TSO_QUEUE
272 || sleeping_queue != END_TSO_QUEUE)
274 if(!startingWorkerThread)
275 { // we don't want to start another worker thread
276 // just because the last one hasn't yet reached the
277 // "waiting for capability" state
278 startingWorkerThread = rtsTrue;
279 if(!startTask(taskStart))
281 startingWorkerThread = rtsFalse;
288 /* ---------------------------------------------------------------------------
289 Main scheduling loop.
291 We use round-robin scheduling, each thread returning to the
292 scheduler loop when one of these conditions is detected:
295 * timer expires (thread yields)
300 Locking notes: we acquire the scheduler lock once at the beginning
301 of the scheduler loop, and release it when
303 * running a thread, or
304 * waiting for work, or
305 * waiting for a GC to complete.
308 In a GranSim setup this loop iterates over the global event queue.
309 This revolves around the global event queue, which determines what
310 to do next. Therefore, it's more complicated than either the
311 concurrent or the parallel (GUM) setup.
314 GUM iterates over incoming messages.
315 It starts with nothing to do (thus CurrentTSO == END_TSO_QUEUE),
316 and sends out a fish whenever it has nothing to do; in-between
317 doing the actual reductions (shared code below) it processes the
318 incoming messages and deals with delayed operations
319 (see PendingFetches).
320 This is not the ugliest code you could imagine, but it's bloody close.
322 ------------------------------------------------------------------------ */
324 schedule( StgMainThread *mainThread USED_WHEN_RTS_SUPPORTS_THREADS,
325 Capability *initialCapability )
329 StgThreadReturnCode ret;
337 rtsBool receivedFinish = rtsFalse;
339 nat tp_size, sp_size; // stats only
342 rtsBool was_interrupted = rtsFalse;
343 StgTSOWhatNext prev_what_next;
345 // Pre-condition: sched_mutex is held.
346 // We might have a capability, passed in as initialCapability.
347 cap = initialCapability;
349 #if defined(RTS_SUPPORTS_THREADS)
351 // in the threaded case, the capability is either passed in via the
352 // initialCapability parameter, or initialized inside the scheduler
356 sched_belch("### NEW SCHEDULER LOOP (main thr: %p, cap: %p)",
357 mainThread, initialCapability);
360 // simply initialise it in the non-threaded case
361 grabCapability(&cap);
365 /* set up first event to get things going */
366 /* ToDo: assign costs for system setup and init MainTSO ! */
367 new_event(CurrentProc, CurrentProc, CurrentTime[CurrentProc],
369 CurrentTSO, (StgClosure*)NULL, (rtsSpark*)NULL);
372 fprintf(stderr, "GRAN: Init CurrentTSO (in schedule) = %p\n", CurrentTSO);
373 G_TSO(CurrentTSO, 5));
375 if (RtsFlags.GranFlags.Light) {
376 /* Save current time; GranSim Light only */
377 CurrentTSO->gran.clock = CurrentTime[CurrentProc];
380 event = get_next_event();
382 while (event!=(rtsEvent*)NULL) {
383 /* Choose the processor with the next event */
384 CurrentProc = event->proc;
385 CurrentTSO = event->tso;
389 while (!receivedFinish) { /* set by processMessages */
390 /* when receiving PP_FINISH message */
392 #else // everything except GRAN and PAR
398 IF_DEBUG(scheduler, printAllThreads());
400 #if defined(RTS_SUPPORTS_THREADS)
401 // Yield the capability to higher-priority tasks if necessary.
404 yieldCapability(&cap);
407 // If we do not currently hold a capability, we wait for one
410 waitForCapability(&sched_mutex, &cap,
411 mainThread ? &mainThread->bound_thread_cond : NULL);
414 // We now have a capability...
418 // If we're interrupted (the user pressed ^C, or some other
419 // termination condition occurred), kill all the currently running
423 IF_DEBUG(scheduler, sched_belch("interrupted"));
424 interrupted = rtsFalse;
425 was_interrupted = rtsTrue;
426 #if defined(RTS_SUPPORTS_THREADS)
427 // In the threaded RTS, deadlock detection doesn't work,
428 // so just exit right away.
429 prog_belch("interrupted");
430 releaseCapability(cap);
431 RELEASE_LOCK(&sched_mutex);
432 shutdownHaskellAndExit(EXIT_SUCCESS);
438 #if defined(RTS_USER_SIGNALS)
439 // check for signals each time around the scheduler
440 if (signals_pending()) {
441 RELEASE_LOCK(&sched_mutex); /* ToDo: kill */
442 startSignalHandlers();
443 ACQUIRE_LOCK(&sched_mutex);
448 // Check whether any waiting threads need to be woken up. If the
449 // run queue is empty, and there are no other tasks running, we
450 // can wait indefinitely for something to happen.
452 if ( !EMPTY_QUEUE(blocked_queue_hd) || !EMPTY_QUEUE(sleeping_queue)
453 #if defined(RTS_SUPPORTS_THREADS)
458 awaitEvent( EMPTY_RUN_QUEUE() );
460 // we can be interrupted while waiting for I/O...
461 if (interrupted) continue;
464 * Detect deadlock: when we have no threads to run, there are no
465 * threads waiting on I/O or sleeping, and all the other tasks are
466 * waiting for work, we must have a deadlock of some description.
468 * We first try to find threads blocked on themselves (ie. black
469 * holes), and generate NonTermination exceptions where necessary.
471 * If no threads are black holed, we have a deadlock situation, so
472 * inform all the main threads.
474 #if !defined(PAR) && !defined(RTS_SUPPORTS_THREADS)
475 if ( EMPTY_THREAD_QUEUES() )
477 IF_DEBUG(scheduler, sched_belch("deadlocked, forcing major GC..."));
478 // Garbage collection can release some new threads due to
479 // either (a) finalizers or (b) threads resurrected because
480 // they are about to be send BlockedOnDeadMVar. Any threads
481 // thus released will be immediately runnable.
482 GarbageCollect(GetRoots,rtsTrue);
484 if ( !EMPTY_RUN_QUEUE() ) { goto not_deadlocked; }
487 sched_belch("still deadlocked, checking for black holes..."));
490 if ( !EMPTY_RUN_QUEUE() ) { goto not_deadlocked; }
492 #if defined(RTS_USER_SIGNALS)
493 /* If we have user-installed signal handlers, then wait
494 * for signals to arrive rather then bombing out with a
497 if ( anyUserHandlers() ) {
499 sched_belch("still deadlocked, waiting for signals..."));
503 // we might be interrupted...
504 if (interrupted) { continue; }
506 if (signals_pending()) {
507 RELEASE_LOCK(&sched_mutex);
508 startSignalHandlers();
509 ACQUIRE_LOCK(&sched_mutex);
511 ASSERT(!EMPTY_RUN_QUEUE());
516 /* Probably a real deadlock. Send the current main thread the
517 * Deadlock exception (or in the SMP build, send *all* main
518 * threads the deadlock exception, since none of them can make
524 switch (m->tso->why_blocked) {
525 case BlockedOnBlackHole:
526 case BlockedOnException:
528 raiseAsync(m->tso, (StgClosure *)NonTermination_closure);
531 barf("deadlock: main thread blocked in a strange way");
537 #elif defined(RTS_SUPPORTS_THREADS)
538 // ToDo: add deadlock detection in threaded RTS
540 // ToDo: add deadlock detection in GUM (similar to SMP) -- HWL
543 #if defined(RTS_SUPPORTS_THREADS)
544 if ( EMPTY_RUN_QUEUE() ) {
545 continue; // nothing to do
550 if (RtsFlags.GranFlags.Light)
551 GranSimLight_enter_system(event, &ActiveTSO); // adjust ActiveTSO etc
553 /* adjust time based on time-stamp */
554 if (event->time > CurrentTime[CurrentProc] &&
555 event->evttype != ContinueThread)
556 CurrentTime[CurrentProc] = event->time;
558 /* Deal with the idle PEs (may issue FindWork or MoveSpark events) */
559 if (!RtsFlags.GranFlags.Light)
562 IF_DEBUG(gran, fprintf(stderr, "GRAN: switch by event-type\n"));
564 /* main event dispatcher in GranSim */
565 switch (event->evttype) {
566 /* Should just be continuing execution */
568 IF_DEBUG(gran, fprintf(stderr, "GRAN: doing ContinueThread\n"));
569 /* ToDo: check assertion
570 ASSERT(run_queue_hd != (StgTSO*)NULL &&
571 run_queue_hd != END_TSO_QUEUE);
573 /* Ignore ContinueThreads for fetching threads (if synchr comm) */
574 if (!RtsFlags.GranFlags.DoAsyncFetch &&
575 procStatus[CurrentProc]==Fetching) {
576 belch("ghuH: Spurious ContinueThread while Fetching ignored; TSO %d (%p) [PE %d]",
577 CurrentTSO->id, CurrentTSO, CurrentProc);
580 /* Ignore ContinueThreads for completed threads */
581 if (CurrentTSO->what_next == ThreadComplete) {
582 belch("ghuH: found a ContinueThread event for completed thread %d (%p) [PE %d] (ignoring ContinueThread)",
583 CurrentTSO->id, CurrentTSO, CurrentProc);
586 /* Ignore ContinueThreads for threads that are being migrated */
587 if (PROCS(CurrentTSO)==Nowhere) {
588 belch("ghuH: trying to run the migrating TSO %d (%p) [PE %d] (ignoring ContinueThread)",
589 CurrentTSO->id, CurrentTSO, CurrentProc);
592 /* The thread should be at the beginning of the run queue */
593 if (CurrentTSO!=run_queue_hds[CurrentProc]) {
594 belch("ghuH: TSO %d (%p) [PE %d] is not at the start of the run_queue when doing a ContinueThread",
595 CurrentTSO->id, CurrentTSO, CurrentProc);
596 break; // run the thread anyway
599 new_event(proc, proc, CurrentTime[proc],
601 (StgTSO*)NULL, (StgClosure*)NULL, (rtsSpark*)NULL);
603 */ /* Catches superfluous CONTINUEs -- should be unnecessary */
604 break; // now actually run the thread; DaH Qu'vam yImuHbej
607 do_the_fetchnode(event);
608 goto next_thread; /* handle next event in event queue */
611 do_the_globalblock(event);
612 goto next_thread; /* handle next event in event queue */
615 do_the_fetchreply(event);
616 goto next_thread; /* handle next event in event queue */
618 case UnblockThread: /* Move from the blocked queue to the tail of */
619 do_the_unblock(event);
620 goto next_thread; /* handle next event in event queue */
622 case ResumeThread: /* Move from the blocked queue to the tail of */
623 /* the runnable queue ( i.e. Qu' SImqa'lu') */
624 event->tso->gran.blocktime +=
625 CurrentTime[CurrentProc] - event->tso->gran.blockedat;
626 do_the_startthread(event);
627 goto next_thread; /* handle next event in event queue */
630 do_the_startthread(event);
631 goto next_thread; /* handle next event in event queue */
634 do_the_movethread(event);
635 goto next_thread; /* handle next event in event queue */
638 do_the_movespark(event);
639 goto next_thread; /* handle next event in event queue */
642 do_the_findwork(event);
643 goto next_thread; /* handle next event in event queue */
646 barf("Illegal event type %u\n", event->evttype);
649 /* This point was scheduler_loop in the old RTS */
651 IF_DEBUG(gran, belch("GRAN: after main switch"));
653 TimeOfLastEvent = CurrentTime[CurrentProc];
654 TimeOfNextEvent = get_time_of_next_event();
655 IgnoreEvents=(TimeOfNextEvent==0); // HWL HACK
656 // CurrentTSO = ThreadQueueHd;
658 IF_DEBUG(gran, belch("GRAN: time of next event is: %ld",
661 if (RtsFlags.GranFlags.Light)
662 GranSimLight_leave_system(event, &ActiveTSO);
664 EndOfTimeSlice = CurrentTime[CurrentProc]+RtsFlags.GranFlags.time_slice;
667 belch("GRAN: end of time-slice is %#lx", EndOfTimeSlice));
669 /* in a GranSim setup the TSO stays on the run queue */
671 /* Take a thread from the run queue. */
672 POP_RUN_QUEUE(t); // take_off_run_queue(t);
675 fprintf(stderr, "GRAN: About to run current thread, which is\n");
678 context_switch = 0; // turned on via GranYield, checking events and time slice
681 DumpGranEvent(GR_SCHEDULE, t));
683 procStatus[CurrentProc] = Busy;
686 if (PendingFetches != END_BF_QUEUE) {
690 /* ToDo: phps merge with spark activation above */
691 /* check whether we have local work and send requests if we have none */
692 if (EMPTY_RUN_QUEUE()) { /* no runnable threads */
693 /* :-[ no local threads => look out for local sparks */
694 /* the spark pool for the current PE */
695 pool = &(MainRegTable.rSparks); // generalise to cap = &MainRegTable
696 if (advisory_thread_count < RtsFlags.ParFlags.maxThreads &&
697 pool->hd < pool->tl) {
699 * ToDo: add GC code check that we really have enough heap afterwards!!
701 * If we're here (no runnable threads) and we have pending
702 * sparks, we must have a space problem. Get enough space
703 * to turn one of those pending sparks into a
707 spark = findSpark(rtsFalse); /* get a spark */
708 if (spark != (rtsSpark) NULL) {
709 tso = activateSpark(spark); /* turn the spark into a thread */
710 IF_PAR_DEBUG(schedule,
711 belch("==== schedule: Created TSO %d (%p); %d threads active",
712 tso->id, tso, advisory_thread_count));
714 if (tso==END_TSO_QUEUE) { /* failed to activate spark->back to loop */
715 belch("==^^ failed to activate spark");
717 } /* otherwise fall through & pick-up new tso */
719 IF_PAR_DEBUG(verbose,
720 belch("==^^ no local sparks (spark pool contains only NFs: %d)",
721 spark_queue_len(pool)));
726 /* If we still have no work we need to send a FISH to get a spark
729 if (EMPTY_RUN_QUEUE()) {
730 /* =8-[ no local sparks => look for work on other PEs */
732 * We really have absolutely no work. Send out a fish
733 * (there may be some out there already), and wait for
734 * something to arrive. We clearly can't run any threads
735 * until a SCHEDULE or RESUME arrives, and so that's what
736 * we're hoping to see. (Of course, we still have to
737 * respond to other types of messages.)
739 TIME now = msTime() /*CURRENT_TIME*/;
740 IF_PAR_DEBUG(verbose,
741 belch("-- now=%ld", now));
742 IF_PAR_DEBUG(verbose,
743 if (outstandingFishes < RtsFlags.ParFlags.maxFishes &&
744 (last_fish_arrived_at!=0 &&
745 last_fish_arrived_at+RtsFlags.ParFlags.fishDelay > now)) {
746 belch("--$$ delaying FISH until %ld (last fish %ld, delay %ld, now %ld)",
747 last_fish_arrived_at+RtsFlags.ParFlags.fishDelay,
748 last_fish_arrived_at,
749 RtsFlags.ParFlags.fishDelay, now);
752 if (outstandingFishes < RtsFlags.ParFlags.maxFishes &&
753 (last_fish_arrived_at==0 ||
754 (last_fish_arrived_at+RtsFlags.ParFlags.fishDelay <= now))) {
755 /* outstandingFishes is set in sendFish, processFish;
756 avoid flooding system with fishes via delay */
758 sendFish(pe, mytid, NEW_FISH_AGE, NEW_FISH_HISTORY,
761 // Global statistics: count no. of fishes
762 if (RtsFlags.ParFlags.ParStats.Global &&
763 RtsFlags.GcFlags.giveStats > NO_GC_STATS) {
764 globalParStats.tot_fish_mess++;
768 receivedFinish = processMessages();
771 } else if (PacketsWaiting()) { /* Look for incoming messages */
772 receivedFinish = processMessages();
775 /* Now we are sure that we have some work available */
776 ASSERT(run_queue_hd != END_TSO_QUEUE);
778 /* Take a thread from the run queue, if we have work */
779 POP_RUN_QUEUE(t); // take_off_run_queue(END_TSO_QUEUE);
780 IF_DEBUG(sanity,checkTSO(t));
782 /* ToDo: write something to the log-file
783 if (RTSflags.ParFlags.granSimStats && !sameThread)
784 DumpGranEvent(GR_SCHEDULE, RunnableThreadsHd);
788 /* the spark pool for the current PE */
789 pool = &(MainRegTable.rSparks); // generalise to cap = &MainRegTable
792 belch("--=^ %d threads, %d sparks on [%#x]",
793 run_queue_len(), spark_queue_len(pool), CURRENT_PROC));
796 if (0 && RtsFlags.ParFlags.ParStats.Full &&
797 t && LastTSO && t->id != LastTSO->id &&
798 LastTSO->why_blocked == NotBlocked &&
799 LastTSO->what_next != ThreadComplete) {
800 // if previously scheduled TSO not blocked we have to record the context switch
801 DumpVeryRawGranEvent(TimeOfLastYield, CURRENT_PROC, CURRENT_PROC,
802 GR_DESCHEDULE, LastTSO, (StgClosure *)NULL, 0, 0);
805 if (RtsFlags.ParFlags.ParStats.Full &&
806 (emitSchedule /* forced emit */ ||
807 (t && LastTSO && t->id != LastTSO->id))) {
809 we are running a different TSO, so write a schedule event to log file
810 NB: If we use fair scheduling we also have to write a deschedule
811 event for LastTSO; with unfair scheduling we know that the
812 previous tso has blocked whenever we switch to another tso, so
813 we don't need it in GUM for now
815 DumpRawGranEvent(CURRENT_PROC, CURRENT_PROC,
816 GR_SCHEDULE, t, (StgClosure *)NULL, 0, 0);
817 emitSchedule = rtsFalse;
821 #else /* !GRAN && !PAR */
823 // grab a thread from the run queue
824 ASSERT(run_queue_hd != END_TSO_QUEUE);
827 // Sanity check the thread we're about to run. This can be
828 // expensive if there is lots of thread switching going on...
829 IF_DEBUG(sanity,checkTSO(t));
834 StgMainThread *m = t->main;
841 sched_belch("### Running thread %d in bound thread", t->id));
842 // yes, the Haskell thread is bound to the current native thread
847 sched_belch("### thread %d bound to another OS thread", t->id));
848 // no, bound to a different Haskell thread: pass to that thread
849 PUSH_ON_RUN_QUEUE(t);
850 passCapability(&m->bound_thread_cond);
856 if(mainThread != NULL)
857 // The thread we want to run is bound.
860 sched_belch("### this OS thread cannot run thread %d", t->id));
861 // no, the current native thread is bound to a different
862 // Haskell thread, so pass it to any worker thread
863 PUSH_ON_RUN_QUEUE(t);
864 passCapabilityToWorker();
871 cap->r.rCurrentTSO = t;
873 /* context switches are now initiated by the timer signal, unless
874 * the user specified "context switch as often as possible", with
877 if ((RtsFlags.ConcFlags.ctxtSwitchTicks == 0
878 && (run_queue_hd != END_TSO_QUEUE
879 || blocked_queue_hd != END_TSO_QUEUE
880 || sleeping_queue != END_TSO_QUEUE)))
887 RELEASE_LOCK(&sched_mutex);
889 IF_DEBUG(scheduler, sched_belch("-->> running thread %ld %s ...",
890 t->id, whatNext_strs[t->what_next]));
893 startHeapProfTimer();
896 /* +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ */
897 /* Run the current thread
899 prev_what_next = t->what_next;
901 errno = t->saved_errno;
903 switch (prev_what_next) {
907 /* Thread already finished, return to scheduler. */
908 ret = ThreadFinished;
912 ret = StgRun((StgFunPtr) stg_returnToStackTop, &cap->r);
915 case ThreadInterpret:
916 ret = interpretBCO(cap);
920 barf("schedule: invalid what_next field");
923 // The TSO might have moved, so find the new location:
924 t = cap->r.rCurrentTSO;
926 // And save the current errno in this thread.
927 t->saved_errno = errno;
929 /* +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ */
931 /* Costs for the scheduler are assigned to CCS_SYSTEM */
937 ACQUIRE_LOCK(&sched_mutex);
939 #ifdef RTS_SUPPORTS_THREADS
940 IF_DEBUG(scheduler,fprintf(stderr,"sched (task %p): ", osThreadId()););
941 #elif !defined(GRAN) && !defined(PAR)
942 IF_DEBUG(scheduler,fprintf(stderr,"sched: "););
946 /* HACK 675: if the last thread didn't yield, make sure to print a
947 SCHEDULE event to the log file when StgRunning the next thread, even
948 if it is the same one as before */
950 TimeOfLastYield = CURRENT_TIME;
956 IF_DEBUG(gran, DumpGranEvent(GR_DESCHEDULE, t));
957 globalGranStats.tot_heapover++;
959 globalParStats.tot_heapover++;
962 // did the task ask for a large block?
963 if (cap->r.rHpAlloc > BLOCK_SIZE_W) {
964 // if so, get one and push it on the front of the nursery.
968 blocks = (nat)BLOCK_ROUND_UP(cap->r.rHpAlloc * sizeof(W_)) / BLOCK_SIZE;
970 IF_DEBUG(scheduler,belch("--<< thread %ld (%s) stopped: requesting a large block (size %d)",
971 t->id, whatNext_strs[t->what_next], blocks));
973 // don't do this if it would push us over the
974 // alloc_blocks_lim limit; we'll GC first.
975 if (alloc_blocks + blocks < alloc_blocks_lim) {
977 alloc_blocks += blocks;
978 bd = allocGroup( blocks );
980 // link the new group into the list
981 bd->link = cap->r.rCurrentNursery;
982 bd->u.back = cap->r.rCurrentNursery->u.back;
983 if (cap->r.rCurrentNursery->u.back != NULL) {
984 cap->r.rCurrentNursery->u.back->link = bd;
986 ASSERT(g0s0->blocks == cap->r.rCurrentNursery &&
987 g0s0->blocks == cap->r.rNursery);
988 cap->r.rNursery = g0s0->blocks = bd;
990 cap->r.rCurrentNursery->u.back = bd;
992 // initialise it as a nursery block. We initialise the
993 // step, gen_no, and flags field of *every* sub-block in
994 // this large block, because this is easier than making
995 // sure that we always find the block head of a large
996 // block whenever we call Bdescr() (eg. evacuate() and
997 // isAlive() in the GC would both have to do this, at
1001 for (x = bd; x < bd + blocks; x++) {
1008 // don't forget to update the block count in g0s0.
1009 g0s0->n_blocks += blocks;
1010 // This assert can be a killer if the app is doing lots
1011 // of large block allocations.
1012 ASSERT(countBlocks(g0s0->blocks) == g0s0->n_blocks);
1014 // now update the nursery to point to the new block
1015 cap->r.rCurrentNursery = bd;
1017 // we might be unlucky and have another thread get on the
1018 // run queue before us and steal the large block, but in that
1019 // case the thread will just end up requesting another large
1021 PUSH_ON_RUN_QUEUE(t);
1026 /* make all the running tasks block on a condition variable,
1027 * maybe set context_switch and wait till they all pile in,
1028 * then have them wait on a GC condition variable.
1030 IF_DEBUG(scheduler,belch("--<< thread %ld (%s) stopped: HeapOverflow",
1031 t->id, whatNext_strs[t->what_next]));
1034 ASSERT(!is_on_queue(t,CurrentProc));
1036 /* Currently we emit a DESCHEDULE event before GC in GUM.
1037 ToDo: either add separate event to distinguish SYSTEM time from rest
1038 or just nuke this DESCHEDULE (and the following SCHEDULE) */
1039 if (0 && RtsFlags.ParFlags.ParStats.Full) {
1040 DumpRawGranEvent(CURRENT_PROC, CURRENT_PROC,
1041 GR_DESCHEDULE, t, (StgClosure *)NULL, 0, 0);
1042 emitSchedule = rtsTrue;
1046 ready_to_gc = rtsTrue;
1047 context_switch = 1; /* stop other threads ASAP */
1048 PUSH_ON_RUN_QUEUE(t);
1049 /* actual GC is done at the end of the while loop */
1055 DumpGranEvent(GR_DESCHEDULE, t));
1056 globalGranStats.tot_stackover++;
1059 // DumpGranEvent(GR_DESCHEDULE, t);
1060 globalParStats.tot_stackover++;
1062 IF_DEBUG(scheduler,belch("--<< thread %ld (%s) stopped, StackOverflow",
1063 t->id, whatNext_strs[t->what_next]));
1064 /* just adjust the stack for this thread, then pop it back
1069 /* enlarge the stack */
1070 StgTSO *new_t = threadStackOverflow(t);
1072 /* This TSO has moved, so update any pointers to it from the
1073 * main thread stack. It better not be on any other queues...
1074 * (it shouldn't be).
1076 if (t->main != NULL) {
1077 t->main->tso = new_t;
1079 PUSH_ON_RUN_QUEUE(new_t);
1083 case ThreadYielding:
1086 DumpGranEvent(GR_DESCHEDULE, t));
1087 globalGranStats.tot_yields++;
1090 // DumpGranEvent(GR_DESCHEDULE, t);
1091 globalParStats.tot_yields++;
1093 /* put the thread back on the run queue. Then, if we're ready to
1094 * GC, check whether this is the last task to stop. If so, wake
1095 * up the GC thread. getThread will block during a GC until the
1099 if (t->what_next != prev_what_next) {
1100 belch("--<< thread %ld (%s) stopped to switch evaluators",
1101 t->id, whatNext_strs[t->what_next]);
1103 belch("--<< thread %ld (%s) stopped, yielding",
1104 t->id, whatNext_strs[t->what_next]);
1109 //belch("&& Doing sanity check on yielding TSO %ld.", t->id);
1111 ASSERT(t->link == END_TSO_QUEUE);
1113 // Shortcut if we're just switching evaluators: don't bother
1114 // doing stack squeezing (which can be expensive), just run the
1116 if (t->what_next != prev_what_next) {
1123 ASSERT(!is_on_queue(t,CurrentProc));
1126 //belch("&& Doing sanity check on all ThreadQueues (and their TSOs).");
1127 checkThreadQsSanity(rtsTrue));
1131 if (RtsFlags.ParFlags.doFairScheduling) {
1132 /* this does round-robin scheduling; good for concurrency */
1133 APPEND_TO_RUN_QUEUE(t);
1135 /* this does unfair scheduling; good for parallelism */
1136 PUSH_ON_RUN_QUEUE(t);
1139 // this does round-robin scheduling; good for concurrency
1140 APPEND_TO_RUN_QUEUE(t);
1144 /* add a ContinueThread event to actually process the thread */
1145 new_event(CurrentProc, CurrentProc, CurrentTime[CurrentProc],
1147 t, (StgClosure*)NULL, (rtsSpark*)NULL);
1149 belch("GRAN: eventq and runnableq after adding yielded thread to queue again:");
1158 belch("--<< thread %ld (%p; %s) stopped, blocking on node %p [PE %d] with BQ: ",
1159 t->id, t, whatNext_strs[t->what_next], t->block_info.closure, (t->block_info.closure==(StgClosure*)NULL ? 99 : where_is(t->block_info.closure)));
1160 if (t->block_info.closure!=(StgClosure*)NULL) print_bq(t->block_info.closure));
1162 // ??? needed; should emit block before
1164 DumpGranEvent(GR_DESCHEDULE, t));
1165 prune_eventq(t, (StgClosure *)NULL); // prune ContinueThreads for t
1168 ASSERT(procStatus[CurrentProc]==Busy ||
1169 ((procStatus[CurrentProc]==Fetching) &&
1170 (t->block_info.closure!=(StgClosure*)NULL)));
1171 if (run_queue_hds[CurrentProc] == END_TSO_QUEUE &&
1172 !(!RtsFlags.GranFlags.DoAsyncFetch &&
1173 procStatus[CurrentProc]==Fetching))
1174 procStatus[CurrentProc] = Idle;
1178 belch("--<< thread %ld (%p; %s) stopped, blocking on node %p with BQ: ",
1179 t->id, t, whatNext_strs[t->what_next], t->block_info.closure));
1182 if (t->block_info.closure!=(StgClosure*)NULL)
1183 print_bq(t->block_info.closure));
1185 /* Send a fetch (if BlockedOnGA) and dump event to log file */
1188 /* whatever we schedule next, we must log that schedule */
1189 emitSchedule = rtsTrue;
1192 /* don't need to do anything. Either the thread is blocked on
1193 * I/O, in which case we'll have called addToBlockedQueue
1194 * previously, or it's blocked on an MVar or Blackhole, in which
1195 * case it'll be on the relevant queue already.
1198 fprintf(stderr, "--<< thread %d (%s) stopped: ",
1199 t->id, whatNext_strs[t->what_next]);
1200 printThreadBlockage(t);
1201 fprintf(stderr, "\n"));
1204 /* Only for dumping event to log file
1205 ToDo: do I need this in GranSim, too?
1212 case ThreadFinished:
1213 /* Need to check whether this was a main thread, and if so, signal
1214 * the task that started it with the return value. If we have no
1215 * more main threads, we probably need to stop all the tasks until
1218 /* We also end up here if the thread kills itself with an
1219 * uncaught exception, see Exception.hc.
1221 IF_DEBUG(scheduler,belch("--++ thread %d (%s) finished",
1222 t->id, whatNext_strs[t->what_next]));
1224 endThread(t, CurrentProc); // clean-up the thread
1226 /* For now all are advisory -- HWL */
1227 //if(t->priority==AdvisoryPriority) ??
1228 advisory_thread_count--;
1231 if(t->dist.priority==RevalPriority)
1235 if (RtsFlags.ParFlags.ParStats.Full &&
1236 !RtsFlags.ParFlags.ParStats.Suppressed)
1237 DumpEndEvent(CURRENT_PROC, t, rtsFalse /* not mandatory */);
1241 // Check whether the thread that just completed was a main
1242 // thread, and if so return with the result.
1244 // There is an assumption here that all thread completion goes
1245 // through this point; we need to make sure that if a thread
1246 // ends up in the ThreadKilled state, that it stays on the run
1247 // queue so it can be dealt with here.
1250 #if defined(RTS_SUPPORTS_THREADS)
1253 mainThread->tso == t
1257 // We are a bound thread: this must be our thread that just
1259 ASSERT(mainThread->tso == t);
1261 if (t->what_next == ThreadComplete) {
1262 if (mainThread->ret) {
1263 // NOTE: return val is tso->sp[1] (see StgStartup.hc)
1264 *(mainThread->ret) = (StgClosure *)mainThread->tso->sp[1];
1266 mainThread->stat = Success;
1268 if (mainThread->ret) {
1269 *(mainThread->ret) = NULL;
1271 if (was_interrupted) {
1272 mainThread->stat = Interrupted;
1274 mainThread->stat = Killed;
1278 removeThreadLabel((StgWord)mainThread->tso->id);
1280 if (mainThread->prev == NULL) {
1281 main_threads = mainThread->link;
1283 mainThread->prev->link = mainThread->link;
1285 if (mainThread->link != NULL) {
1286 mainThread->link->prev = NULL;
1288 releaseCapability(cap);
1292 #ifdef RTS_SUPPORTS_THREADS
1293 ASSERT(t->main == NULL);
1295 if (t->main != NULL) {
1296 // Must be a main thread that is not the topmost one. Leave
1297 // it on the run queue until the stack has unwound to the
1298 // point where we can deal with this. Leaving it on the run
1299 // queue also ensures that the garbage collector knows about
1300 // this thread and its return value (it gets dropped from the
1301 // all_threads list so there's no other way to find it).
1302 APPEND_TO_RUN_QUEUE(t);
1308 barf("schedule: invalid thread return code %d", (int)ret);
1312 // When we have +RTS -i0 and we're heap profiling, do a census at
1313 // every GC. This lets us get repeatable runs for debugging.
1314 if (performHeapProfile ||
1315 (RtsFlags.ProfFlags.profileInterval==0 &&
1316 RtsFlags.ProfFlags.doHeapProfile && ready_to_gc)) {
1317 GarbageCollect(GetRoots, rtsTrue);
1319 performHeapProfile = rtsFalse;
1320 ready_to_gc = rtsFalse; // we already GC'd
1325 /* everybody back, start the GC.
1326 * Could do it in this thread, or signal a condition var
1327 * to do it in another thread. Either way, we need to
1328 * broadcast on gc_pending_cond afterward.
1330 #if defined(RTS_SUPPORTS_THREADS)
1331 IF_DEBUG(scheduler,sched_belch("doing GC"));
1333 GarbageCollect(GetRoots,rtsFalse);
1334 ready_to_gc = rtsFalse;
1336 /* add a ContinueThread event to continue execution of current thread */
1337 new_event(CurrentProc, CurrentProc, CurrentTime[CurrentProc],
1339 t, (StgClosure*)NULL, (rtsSpark*)NULL);
1341 fprintf(stderr, "GRAN: eventq and runnableq after Garbage collection:\n");
1349 IF_GRAN_DEBUG(unused,
1350 print_eventq(EventHd));
1352 event = get_next_event();
1355 /* ToDo: wait for next message to arrive rather than busy wait */
1358 } /* end of while(1) */
1360 IF_PAR_DEBUG(verbose,
1361 belch("== Leaving schedule() after having received Finish"));
1364 /* ---------------------------------------------------------------------------
1365 * rtsSupportsBoundThreads(): is the RTS built to support bound threads?
1366 * used by Control.Concurrent for error checking.
1367 * ------------------------------------------------------------------------- */
1370 rtsSupportsBoundThreads(void)
1379 /* ---------------------------------------------------------------------------
1380 * isThreadBound(tso): check whether tso is bound to an OS thread.
1381 * ------------------------------------------------------------------------- */
1384 isThreadBound(StgTSO* tso USED_IN_THREADED_RTS)
1387 return (tso->main != NULL);
1392 /* ---------------------------------------------------------------------------
1393 * Singleton fork(). Do not copy any running threads.
1394 * ------------------------------------------------------------------------- */
1396 #ifndef mingw32_TARGET_OS
1397 #define FORKPROCESS_PRIMOP_SUPPORTED
1400 #ifdef FORKPROCESS_PRIMOP_SUPPORTED
1402 deleteThreadImmediately(StgTSO *tso);
1405 forkProcess(HsStablePtr *entry
1406 #ifndef FORKPROCESS_PRIMOP_SUPPORTED
1411 #ifdef FORKPROCESS_PRIMOP_SUPPORTED
1417 IF_DEBUG(scheduler,sched_belch("forking!"));
1418 rts_lock(); // This not only acquires sched_mutex, it also
1419 // makes sure that no other threads are running
1423 if (pid) { /* parent */
1425 /* just return the pid */
1429 } else { /* child */
1432 // delete all threads
1433 run_queue_hd = run_queue_tl = END_TSO_QUEUE;
1435 for (t = all_threads; t != END_TSO_QUEUE; t = next) {
1438 // don't allow threads to catch the ThreadKilled exception
1439 deleteThreadImmediately(t);
1442 // wipe the main thread list
1443 while((m = main_threads) != NULL) {
1444 main_threads = m->link;
1445 # ifdef THREADED_RTS
1446 closeCondition(&m->bound_thread_cond);
1451 # ifdef RTS_SUPPORTS_THREADS
1452 resetTaskManagerAfterFork(); // tell startTask() and friends that
1453 startingWorkerThread = rtsFalse; // we have no worker threads any more
1454 resetWorkerWakeupPipeAfterFork();
1457 rc = rts_evalStableIO(entry, NULL); // run the action
1458 rts_checkSchedStatus("forkProcess",rc);
1462 hs_exit(); // clean up and exit
1465 #else /* !FORKPROCESS_PRIMOP_SUPPORTED */
1466 barf("forkProcess#: primop not supported, sorry!\n");
1471 /* ---------------------------------------------------------------------------
1472 * deleteAllThreads(): kill all the live threads.
1474 * This is used when we catch a user interrupt (^C), before performing
1475 * any necessary cleanups and running finalizers.
1477 * Locks: sched_mutex held.
1478 * ------------------------------------------------------------------------- */
1481 deleteAllThreads ( void )
1484 IF_DEBUG(scheduler,sched_belch("deleting all threads"));
1485 for (t = all_threads; t != END_TSO_QUEUE; t = next) {
1486 next = t->global_link;
1490 // The run queue now contains a bunch of ThreadKilled threads. We
1491 // must not throw these away: the main thread(s) will be in there
1492 // somewhere, and the main scheduler loop has to deal with it.
1493 // Also, the run queue is the only thing keeping these threads from
1494 // being GC'd, and we don't want the "main thread has been GC'd" panic.
1496 ASSERT(blocked_queue_hd == END_TSO_QUEUE);
1497 ASSERT(sleeping_queue == END_TSO_QUEUE);
1500 /* startThread and insertThread are now in GranSim.c -- HWL */
1503 /* ---------------------------------------------------------------------------
1504 * Suspending & resuming Haskell threads.
1506 * When making a "safe" call to C (aka _ccall_GC), the task gives back
1507 * its capability before calling the C function. This allows another
1508 * task to pick up the capability and carry on running Haskell
1509 * threads. It also means that if the C call blocks, it won't lock
1512 * The Haskell thread making the C call is put to sleep for the
1513 * duration of the call, on the susepended_ccalling_threads queue. We
1514 * give out a token to the task, which it can use to resume the thread
1515 * on return from the C function.
1516 * ------------------------------------------------------------------------- */
1519 suspendThread( StgRegTable *reg,
1528 int saved_errno = errno;
1530 /* assume that *reg is a pointer to the StgRegTable part
1533 cap = (Capability *)((void *)((unsigned char*)reg - sizeof(StgFunTable)));
1535 ACQUIRE_LOCK(&sched_mutex);
1538 sched_belch("thread %d did a _ccall_gc (is_concurrent: %d)", cap->r.rCurrentTSO->id,concCall));
1540 // XXX this might not be necessary --SDM
1541 cap->r.rCurrentTSO->what_next = ThreadRunGHC;
1543 threadPaused(cap->r.rCurrentTSO);
1544 cap->r.rCurrentTSO->link = suspended_ccalling_threads;
1545 suspended_ccalling_threads = cap->r.rCurrentTSO;
1547 if(cap->r.rCurrentTSO->blocked_exceptions == NULL) {
1548 cap->r.rCurrentTSO->why_blocked = BlockedOnCCall;
1549 cap->r.rCurrentTSO->blocked_exceptions = END_TSO_QUEUE;
1551 cap->r.rCurrentTSO->why_blocked = BlockedOnCCall_NoUnblockExc;
1554 /* Use the thread ID as the token; it should be unique */
1555 tok = cap->r.rCurrentTSO->id;
1557 /* Hand back capability */
1558 releaseCapability(cap);
1560 #if defined(RTS_SUPPORTS_THREADS)
1561 /* Preparing to leave the RTS, so ensure there's a native thread/task
1562 waiting to take over.
1564 IF_DEBUG(scheduler, sched_belch("worker (token %d): leaving RTS", tok));
1567 /* Other threads _might_ be available for execution; signal this */
1569 RELEASE_LOCK(&sched_mutex);
1571 errno = saved_errno;
1576 resumeThread( StgInt tok,
1577 rtsBool concCall STG_UNUSED )
1579 StgTSO *tso, **prev;
1581 int saved_errno = errno;
1583 #if defined(RTS_SUPPORTS_THREADS)
1584 /* Wait for permission to re-enter the RTS with the result. */
1585 ACQUIRE_LOCK(&sched_mutex);
1586 waitForReturnCapability(&sched_mutex, &cap);
1588 IF_DEBUG(scheduler, sched_belch("worker (token %d): re-entering RTS", tok));
1590 grabCapability(&cap);
1593 /* Remove the thread off of the suspended list */
1594 prev = &suspended_ccalling_threads;
1595 for (tso = suspended_ccalling_threads;
1596 tso != END_TSO_QUEUE;
1597 prev = &tso->link, tso = tso->link) {
1598 if (tso->id == (StgThreadID)tok) {
1603 if (tso == END_TSO_QUEUE) {
1604 barf("resumeThread: thread not found");
1606 tso->link = END_TSO_QUEUE;
1608 if(tso->why_blocked == BlockedOnCCall) {
1609 awakenBlockedQueueNoLock(tso->blocked_exceptions);
1610 tso->blocked_exceptions = NULL;
1613 /* Reset blocking status */
1614 tso->why_blocked = NotBlocked;
1616 cap->r.rCurrentTSO = tso;
1617 RELEASE_LOCK(&sched_mutex);
1618 errno = saved_errno;
1623 /* ---------------------------------------------------------------------------
1625 * ------------------------------------------------------------------------ */
1626 static void unblockThread(StgTSO *tso);
1628 /* ---------------------------------------------------------------------------
1629 * Comparing Thread ids.
1631 * This is used from STG land in the implementation of the
1632 * instances of Eq/Ord for ThreadIds.
1633 * ------------------------------------------------------------------------ */
1636 cmp_thread(StgPtr tso1, StgPtr tso2)
1638 StgThreadID id1 = ((StgTSO *)tso1)->id;
1639 StgThreadID id2 = ((StgTSO *)tso2)->id;
1641 if (id1 < id2) return (-1);
1642 if (id1 > id2) return 1;
1646 /* ---------------------------------------------------------------------------
1647 * Fetching the ThreadID from an StgTSO.
1649 * This is used in the implementation of Show for ThreadIds.
1650 * ------------------------------------------------------------------------ */
1652 rts_getThreadId(StgPtr tso)
1654 return ((StgTSO *)tso)->id;
1659 labelThread(StgPtr tso, char *label)
1664 /* Caveat: Once set, you can only set the thread name to "" */
1665 len = strlen(label)+1;
1666 buf = stgMallocBytes(len * sizeof(char), "Schedule.c:labelThread()");
1667 strncpy(buf,label,len);
1668 /* Update will free the old memory for us */
1669 updateThreadLabel(((StgTSO *)tso)->id,buf);
1673 /* ---------------------------------------------------------------------------
1674 Create a new thread.
1676 The new thread starts with the given stack size. Before the
1677 scheduler can run, however, this thread needs to have a closure
1678 (and possibly some arguments) pushed on its stack. See
1679 pushClosure() in Schedule.h.
1681 createGenThread() and createIOThread() (in SchedAPI.h) are
1682 convenient packaged versions of this function.
1684 currently pri (priority) is only used in a GRAN setup -- HWL
1685 ------------------------------------------------------------------------ */
1687 /* currently pri (priority) is only used in a GRAN setup -- HWL */
1689 createThread(nat size, StgInt pri)
1692 createThread(nat size)
1699 /* First check whether we should create a thread at all */
1701 /* check that no more than RtsFlags.ParFlags.maxThreads threads are created */
1702 if (advisory_thread_count >= RtsFlags.ParFlags.maxThreads) {
1704 belch("{createThread}Daq ghuH: refusing to create another thread; no more than %d threads allowed (currently %d)",
1705 RtsFlags.ParFlags.maxThreads, advisory_thread_count);
1706 return END_TSO_QUEUE;
1712 ASSERT(!RtsFlags.GranFlags.Light || CurrentProc==0);
1715 // ToDo: check whether size = stack_size - TSO_STRUCT_SIZEW
1717 /* catch ridiculously small stack sizes */
1718 if (size < MIN_STACK_WORDS + TSO_STRUCT_SIZEW) {
1719 size = MIN_STACK_WORDS + TSO_STRUCT_SIZEW;
1722 stack_size = size - TSO_STRUCT_SIZEW;
1724 tso = (StgTSO *)allocate(size);
1725 TICK_ALLOC_TSO(stack_size, 0);
1727 SET_HDR(tso, &stg_TSO_info, CCS_SYSTEM);
1729 SET_GRAN_HDR(tso, ThisPE);
1732 // Always start with the compiled code evaluator
1733 tso->what_next = ThreadRunGHC;
1735 tso->id = next_thread_id++;
1736 tso->why_blocked = NotBlocked;
1737 tso->blocked_exceptions = NULL;
1739 tso->saved_errno = 0;
1742 tso->stack_size = stack_size;
1743 tso->max_stack_size = round_to_mblocks(RtsFlags.GcFlags.maxStkSize)
1745 tso->sp = (P_)&(tso->stack) + stack_size;
1748 tso->prof.CCCS = CCS_MAIN;
1751 /* put a stop frame on the stack */
1752 tso->sp -= sizeofW(StgStopFrame);
1753 SET_HDR((StgClosure*)tso->sp,(StgInfoTable *)&stg_stop_thread_info,CCS_SYSTEM);
1756 tso->link = END_TSO_QUEUE;
1757 /* uses more flexible routine in GranSim */
1758 insertThread(tso, CurrentProc);
1760 /* In a non-GranSim setup the pushing of a TSO onto the runq is separated
1766 if (RtsFlags.GranFlags.GranSimStats.Full)
1767 DumpGranEvent(GR_START,tso);
1769 if (RtsFlags.ParFlags.ParStats.Full)
1770 DumpGranEvent(GR_STARTQ,tso);
1771 /* HACk to avoid SCHEDULE
1775 /* Link the new thread on the global thread list.
1777 tso->global_link = all_threads;
1781 tso->dist.priority = MandatoryPriority; //by default that is...
1785 tso->gran.pri = pri;
1787 tso->gran.magic = TSO_MAGIC; // debugging only
1789 tso->gran.sparkname = 0;
1790 tso->gran.startedat = CURRENT_TIME;
1791 tso->gran.exported = 0;
1792 tso->gran.basicblocks = 0;
1793 tso->gran.allocs = 0;
1794 tso->gran.exectime = 0;
1795 tso->gran.fetchtime = 0;
1796 tso->gran.fetchcount = 0;
1797 tso->gran.blocktime = 0;
1798 tso->gran.blockcount = 0;
1799 tso->gran.blockedat = 0;
1800 tso->gran.globalsparks = 0;
1801 tso->gran.localsparks = 0;
1802 if (RtsFlags.GranFlags.Light)
1803 tso->gran.clock = Now; /* local clock */
1805 tso->gran.clock = 0;
1807 IF_DEBUG(gran,printTSO(tso));
1810 tso->par.magic = TSO_MAGIC; // debugging only
1812 tso->par.sparkname = 0;
1813 tso->par.startedat = CURRENT_TIME;
1814 tso->par.exported = 0;
1815 tso->par.basicblocks = 0;
1816 tso->par.allocs = 0;
1817 tso->par.exectime = 0;
1818 tso->par.fetchtime = 0;
1819 tso->par.fetchcount = 0;
1820 tso->par.blocktime = 0;
1821 tso->par.blockcount = 0;
1822 tso->par.blockedat = 0;
1823 tso->par.globalsparks = 0;
1824 tso->par.localsparks = 0;
1828 globalGranStats.tot_threads_created++;
1829 globalGranStats.threads_created_on_PE[CurrentProc]++;
1830 globalGranStats.tot_sq_len += spark_queue_len(CurrentProc);
1831 globalGranStats.tot_sq_probes++;
1833 // collect parallel global statistics (currently done together with GC stats)
1834 if (RtsFlags.ParFlags.ParStats.Global &&
1835 RtsFlags.GcFlags.giveStats > NO_GC_STATS) {
1836 //fprintf(stderr, "Creating thread %d @ %11.2f\n", tso->id, usertime());
1837 globalParStats.tot_threads_created++;
1843 belch("==__ schedule: Created TSO %d (%p);",
1844 CurrentProc, tso, tso->id));
1846 IF_PAR_DEBUG(verbose,
1847 belch("==__ schedule: Created TSO %d (%p); %d threads active",
1848 tso->id, tso, advisory_thread_count));
1850 IF_DEBUG(scheduler,sched_belch("created thread %ld, stack size = %lx words",
1851 tso->id, tso->stack_size));
1858 all parallel thread creation calls should fall through the following routine.
1861 createSparkThread(rtsSpark spark)
1863 ASSERT(spark != (rtsSpark)NULL);
1864 if (advisory_thread_count >= RtsFlags.ParFlags.maxThreads)
1866 barf("{createSparkThread}Daq ghuH: refusing to create another thread; no more than %d threads allowed (currently %d)",
1867 RtsFlags.ParFlags.maxThreads, advisory_thread_count);
1868 return END_TSO_QUEUE;
1872 tso = createThread(RtsFlags.GcFlags.initialStkSize);
1873 if (tso==END_TSO_QUEUE)
1874 barf("createSparkThread: Cannot create TSO");
1876 tso->priority = AdvisoryPriority;
1878 pushClosure(tso,spark);
1879 PUSH_ON_RUN_QUEUE(tso);
1880 advisory_thread_count++;
1887 Turn a spark into a thread.
1888 ToDo: fix for SMP (needs to acquire SCHED_MUTEX!)
1892 activateSpark (rtsSpark spark)
1896 tso = createSparkThread(spark);
1897 if (RtsFlags.ParFlags.ParStats.Full) {
1898 //ASSERT(run_queue_hd == END_TSO_QUEUE); // I think ...
1899 IF_PAR_DEBUG(verbose,
1900 belch("==^^ activateSpark: turning spark of closure %p (%s) into a thread",
1901 (StgClosure *)spark, info_type((StgClosure *)spark)));
1903 // ToDo: fwd info on local/global spark to thread -- HWL
1904 // tso->gran.exported = spark->exported;
1905 // tso->gran.locked = !spark->global;
1906 // tso->gran.sparkname = spark->name;
1912 static SchedulerStatus waitThread_(/*out*/StgMainThread* m,
1913 Capability *initialCapability
1917 /* ---------------------------------------------------------------------------
1920 * scheduleThread puts a thread on the head of the runnable queue.
1921 * This will usually be done immediately after a thread is created.
1922 * The caller of scheduleThread must create the thread using e.g.
1923 * createThread and push an appropriate closure
1924 * on this thread's stack before the scheduler is invoked.
1925 * ------------------------------------------------------------------------ */
1927 static void scheduleThread_ (StgTSO* tso);
1930 scheduleThread_(StgTSO *tso)
1932 // Precondition: sched_mutex must be held.
1933 PUSH_ON_RUN_QUEUE(tso);
1938 scheduleThread(StgTSO* tso)
1940 ACQUIRE_LOCK(&sched_mutex);
1941 scheduleThread_(tso);
1942 RELEASE_LOCK(&sched_mutex);
1945 #if defined(RTS_SUPPORTS_THREADS)
1946 static Condition bound_cond_cache;
1947 static int bound_cond_cache_full = 0;
1952 scheduleWaitThread(StgTSO* tso, /*[out]*/HaskellObj* ret,
1953 Capability *initialCapability)
1955 // Precondition: sched_mutex must be held
1958 m = stgMallocBytes(sizeof(StgMainThread), "waitThread");
1963 m->link = main_threads;
1965 if (main_threads != NULL) {
1966 main_threads->prev = m;
1970 #if defined(RTS_SUPPORTS_THREADS)
1971 // Allocating a new condition for each thread is expensive, so we
1972 // cache one. This is a pretty feeble hack, but it helps speed up
1973 // consecutive call-ins quite a bit.
1974 if (bound_cond_cache_full) {
1975 m->bound_thread_cond = bound_cond_cache;
1976 bound_cond_cache_full = 0;
1978 initCondition(&m->bound_thread_cond);
1982 /* Put the thread on the main-threads list prior to scheduling the TSO.
1983 Failure to do so introduces a race condition in the MT case (as
1984 identified by Wolfgang Thaller), whereby the new task/OS thread
1985 created by scheduleThread_() would complete prior to the thread
1986 that spawned it managed to put 'itself' on the main-threads list.
1987 The upshot of it all being that the worker thread wouldn't get to
1988 signal the completion of the its work item for the main thread to
1989 see (==> it got stuck waiting.) -- sof 6/02.
1991 IF_DEBUG(scheduler, sched_belch("waiting for thread (%d)", tso->id));
1993 PUSH_ON_RUN_QUEUE(tso);
1994 // NB. Don't call THREAD_RUNNABLE() here, because the thread is
1995 // bound and only runnable by *this* OS thread, so waking up other
1996 // workers will just slow things down.
1998 return waitThread_(m, initialCapability);
2001 /* ---------------------------------------------------------------------------
2004 * Initialise the scheduler. This resets all the queues - if the
2005 * queues contained any threads, they'll be garbage collected at the
2008 * ------------------------------------------------------------------------ */
2016 for (i=0; i<=MAX_PROC; i++) {
2017 run_queue_hds[i] = END_TSO_QUEUE;
2018 run_queue_tls[i] = END_TSO_QUEUE;
2019 blocked_queue_hds[i] = END_TSO_QUEUE;
2020 blocked_queue_tls[i] = END_TSO_QUEUE;
2021 ccalling_threadss[i] = END_TSO_QUEUE;
2022 sleeping_queue = END_TSO_QUEUE;
2025 run_queue_hd = END_TSO_QUEUE;
2026 run_queue_tl = END_TSO_QUEUE;
2027 blocked_queue_hd = END_TSO_QUEUE;
2028 blocked_queue_tl = END_TSO_QUEUE;
2029 sleeping_queue = END_TSO_QUEUE;
2032 suspended_ccalling_threads = END_TSO_QUEUE;
2034 main_threads = NULL;
2035 all_threads = END_TSO_QUEUE;
2040 RtsFlags.ConcFlags.ctxtSwitchTicks =
2041 RtsFlags.ConcFlags.ctxtSwitchTime / TICK_MILLISECS;
2043 #if defined(RTS_SUPPORTS_THREADS)
2044 /* Initialise the mutex and condition variables used by
2046 initMutex(&sched_mutex);
2047 initMutex(&term_mutex);
2050 ACQUIRE_LOCK(&sched_mutex);
2052 /* A capability holds the state a native thread needs in
2053 * order to execute STG code. At least one capability is
2054 * floating around (only SMP builds have more than one).
2058 #if defined(RTS_SUPPORTS_THREADS)
2059 /* start our haskell execution tasks */
2060 startTaskManager(0,taskStart);
2063 #if /* defined(SMP) ||*/ defined(PAR)
2067 RELEASE_LOCK(&sched_mutex);
2071 exitScheduler( void )
2073 #if defined(RTS_SUPPORTS_THREADS)
2076 shutting_down_scheduler = rtsTrue;
2079 /* ----------------------------------------------------------------------------
2080 Managing the per-task allocation areas.
2082 Each capability comes with an allocation area. These are
2083 fixed-length block lists into which allocation can be done.
2085 ToDo: no support for two-space collection at the moment???
2086 ------------------------------------------------------------------------- */
2090 waitThread_(StgMainThread* m, Capability *initialCapability)
2092 SchedulerStatus stat;
2094 // Precondition: sched_mutex must be held.
2095 IF_DEBUG(scheduler, sched_belch("new main thread (%d)", m->tso->id));
2098 /* GranSim specific init */
2099 CurrentTSO = m->tso; // the TSO to run
2100 procStatus[MainProc] = Busy; // status of main PE
2101 CurrentProc = MainProc; // PE to run it on
2102 schedule(m,initialCapability);
2104 schedule(m,initialCapability);
2105 ASSERT(m->stat != NoStatus);
2110 #if defined(RTS_SUPPORTS_THREADS)
2111 // Free the condition variable, returning it to the cache if possible.
2112 if (!bound_cond_cache_full) {
2113 bound_cond_cache = m->bound_thread_cond;
2114 bound_cond_cache_full = 1;
2116 closeCondition(&m->bound_thread_cond);
2120 IF_DEBUG(scheduler, sched_belch("main thread (%d) finished", m->tso->id));
2123 // Postcondition: sched_mutex still held
2127 /* ---------------------------------------------------------------------------
2128 Where are the roots that we know about?
2130 - all the threads on the runnable queue
2131 - all the threads on the blocked queue
2132 - all the threads on the sleeping queue
2133 - all the thread currently executing a _ccall_GC
2134 - all the "main threads"
2136 ------------------------------------------------------------------------ */
2138 /* This has to be protected either by the scheduler monitor, or by the
2139 garbage collection monitor (probably the latter).
2144 GetRoots( evac_fn evac )
2149 for (i=0; i<=RtsFlags.GranFlags.proc; i++) {
2150 if ((run_queue_hds[i] != END_TSO_QUEUE) && ((run_queue_hds[i] != NULL)))
2151 evac((StgClosure **)&run_queue_hds[i]);
2152 if ((run_queue_tls[i] != END_TSO_QUEUE) && ((run_queue_tls[i] != NULL)))
2153 evac((StgClosure **)&run_queue_tls[i]);
2155 if ((blocked_queue_hds[i] != END_TSO_QUEUE) && ((blocked_queue_hds[i] != NULL)))
2156 evac((StgClosure **)&blocked_queue_hds[i]);
2157 if ((blocked_queue_tls[i] != END_TSO_QUEUE) && ((blocked_queue_tls[i] != NULL)))
2158 evac((StgClosure **)&blocked_queue_tls[i]);
2159 if ((ccalling_threadss[i] != END_TSO_QUEUE) && ((ccalling_threadss[i] != NULL)))
2160 evac((StgClosure **)&ccalling_threads[i]);
2167 if (run_queue_hd != END_TSO_QUEUE) {
2168 ASSERT(run_queue_tl != END_TSO_QUEUE);
2169 evac((StgClosure **)&run_queue_hd);
2170 evac((StgClosure **)&run_queue_tl);
2173 if (blocked_queue_hd != END_TSO_QUEUE) {
2174 ASSERT(blocked_queue_tl != END_TSO_QUEUE);
2175 evac((StgClosure **)&blocked_queue_hd);
2176 evac((StgClosure **)&blocked_queue_tl);
2179 if (sleeping_queue != END_TSO_QUEUE) {
2180 evac((StgClosure **)&sleeping_queue);
2184 if (suspended_ccalling_threads != END_TSO_QUEUE) {
2185 evac((StgClosure **)&suspended_ccalling_threads);
2188 #if defined(PAR) || defined(GRAN)
2189 markSparkQueue(evac);
2192 #if defined(RTS_USER_SIGNALS)
2193 // mark the signal handlers (signals should be already blocked)
2194 markSignalHandlers(evac);
2198 /* -----------------------------------------------------------------------------
2201 This is the interface to the garbage collector from Haskell land.
2202 We provide this so that external C code can allocate and garbage
2203 collect when called from Haskell via _ccall_GC.
2205 It might be useful to provide an interface whereby the programmer
2206 can specify more roots (ToDo).
2208 This needs to be protected by the GC condition variable above. KH.
2209 -------------------------------------------------------------------------- */
2211 static void (*extra_roots)(evac_fn);
2216 /* Obligated to hold this lock upon entry */
2217 ACQUIRE_LOCK(&sched_mutex);
2218 GarbageCollect(GetRoots,rtsFalse);
2219 RELEASE_LOCK(&sched_mutex);
2223 performMajorGC(void)
2225 ACQUIRE_LOCK(&sched_mutex);
2226 GarbageCollect(GetRoots,rtsTrue);
2227 RELEASE_LOCK(&sched_mutex);
2231 AllRoots(evac_fn evac)
2233 GetRoots(evac); // the scheduler's roots
2234 extra_roots(evac); // the user's roots
2238 performGCWithRoots(void (*get_roots)(evac_fn))
2240 ACQUIRE_LOCK(&sched_mutex);
2241 extra_roots = get_roots;
2242 GarbageCollect(AllRoots,rtsFalse);
2243 RELEASE_LOCK(&sched_mutex);
2246 /* -----------------------------------------------------------------------------
2249 If the thread has reached its maximum stack size, then raise the
2250 StackOverflow exception in the offending thread. Otherwise
2251 relocate the TSO into a larger chunk of memory and adjust its stack
2253 -------------------------------------------------------------------------- */
2256 threadStackOverflow(StgTSO *tso)
2258 nat new_stack_size, new_tso_size, stack_words;
2262 IF_DEBUG(sanity,checkTSO(tso));
2263 if (tso->stack_size >= tso->max_stack_size) {
2266 belch("@@ threadStackOverflow of TSO %d (%p): stack too large (now %ld; max is %ld)",
2267 tso->id, tso, tso->stack_size, tso->max_stack_size);
2268 /* If we're debugging, just print out the top of the stack */
2269 printStackChunk(tso->sp, stg_min(tso->stack+tso->stack_size,
2272 /* Send this thread the StackOverflow exception */
2273 raiseAsync(tso, (StgClosure *)stackOverflow_closure);
2277 /* Try to double the current stack size. If that takes us over the
2278 * maximum stack size for this thread, then use the maximum instead.
2279 * Finally round up so the TSO ends up as a whole number of blocks.
2281 new_stack_size = stg_min(tso->stack_size * 2, tso->max_stack_size);
2282 new_tso_size = (nat)BLOCK_ROUND_UP(new_stack_size * sizeof(W_) +
2283 TSO_STRUCT_SIZE)/sizeof(W_);
2284 new_tso_size = round_to_mblocks(new_tso_size); /* Be MBLOCK-friendly */
2285 new_stack_size = new_tso_size - TSO_STRUCT_SIZEW;
2287 IF_DEBUG(scheduler, fprintf(stderr,"== sched: increasing stack size from %d words to %d.\n", tso->stack_size, new_stack_size));
2289 dest = (StgTSO *)allocate(new_tso_size);
2290 TICK_ALLOC_TSO(new_stack_size,0);
2292 /* copy the TSO block and the old stack into the new area */
2293 memcpy(dest,tso,TSO_STRUCT_SIZE);
2294 stack_words = tso->stack + tso->stack_size - tso->sp;
2295 new_sp = (P_)dest + new_tso_size - stack_words;
2296 memcpy(new_sp, tso->sp, stack_words * sizeof(W_));
2298 /* relocate the stack pointers... */
2300 dest->stack_size = new_stack_size;
2302 /* Mark the old TSO as relocated. We have to check for relocated
2303 * TSOs in the garbage collector and any primops that deal with TSOs.
2305 * It's important to set the sp value to just beyond the end
2306 * of the stack, so we don't attempt to scavenge any part of the
2309 tso->what_next = ThreadRelocated;
2311 tso->sp = (P_)&(tso->stack[tso->stack_size]);
2312 tso->why_blocked = NotBlocked;
2313 dest->mut_link = NULL;
2315 IF_PAR_DEBUG(verbose,
2316 belch("@@ threadStackOverflow of TSO %d (now at %p): stack size increased to %ld",
2317 tso->id, tso, tso->stack_size);
2318 /* If we're debugging, just print out the top of the stack */
2319 printStackChunk(tso->sp, stg_min(tso->stack+tso->stack_size,
2322 IF_DEBUG(sanity,checkTSO(tso));
2324 IF_DEBUG(scheduler,printTSO(dest));
2330 /* ---------------------------------------------------------------------------
2331 Wake up a queue that was blocked on some resource.
2332 ------------------------------------------------------------------------ */
2336 unblockCount ( StgBlockingQueueElement *bqe, StgClosure *node )
2341 unblockCount ( StgBlockingQueueElement *bqe, StgClosure *node )
2343 /* write RESUME events to log file and
2344 update blocked and fetch time (depending on type of the orig closure) */
2345 if (RtsFlags.ParFlags.ParStats.Full) {
2346 DumpRawGranEvent(CURRENT_PROC, CURRENT_PROC,
2347 GR_RESUMEQ, ((StgTSO *)bqe), ((StgTSO *)bqe)->block_info.closure,
2348 0, 0 /* spark_queue_len(ADVISORY_POOL) */);
2349 if (EMPTY_RUN_QUEUE())
2350 emitSchedule = rtsTrue;
2352 switch (get_itbl(node)->type) {
2354 ((StgTSO *)bqe)->par.fetchtime += CURRENT_TIME-((StgTSO *)bqe)->par.blockedat;
2359 ((StgTSO *)bqe)->par.blocktime += CURRENT_TIME-((StgTSO *)bqe)->par.blockedat;
2366 barf("{unblockOneLocked}Daq Qagh: unexpected closure in blocking queue");
2373 static StgBlockingQueueElement *
2374 unblockOneLocked(StgBlockingQueueElement *bqe, StgClosure *node)
2377 PEs node_loc, tso_loc;
2379 node_loc = where_is(node); // should be lifted out of loop
2380 tso = (StgTSO *)bqe; // wastes an assignment to get the type right
2381 tso_loc = where_is((StgClosure *)tso);
2382 if (IS_LOCAL_TO(PROCS(node),tso_loc)) { // TSO is local
2383 /* !fake_fetch => TSO is on CurrentProc is same as IS_LOCAL_TO */
2384 ASSERT(CurrentProc!=node_loc || tso_loc==CurrentProc);
2385 CurrentTime[CurrentProc] += RtsFlags.GranFlags.Costs.lunblocktime;
2386 // insertThread(tso, node_loc);
2387 new_event(tso_loc, tso_loc, CurrentTime[CurrentProc],
2389 tso, node, (rtsSpark*)NULL);
2390 tso->link = END_TSO_QUEUE; // overwrite link just to be sure
2393 } else { // TSO is remote (actually should be FMBQ)
2394 CurrentTime[CurrentProc] += RtsFlags.GranFlags.Costs.mpacktime +
2395 RtsFlags.GranFlags.Costs.gunblocktime +
2396 RtsFlags.GranFlags.Costs.latency;
2397 new_event(tso_loc, CurrentProc, CurrentTime[CurrentProc],
2399 tso, node, (rtsSpark*)NULL);
2400 tso->link = END_TSO_QUEUE; // overwrite link just to be sure
2403 /* the thread-queue-overhead is accounted for in either Resume or UnblockThread */
2405 fprintf(stderr," %s TSO %d (%p) [PE %d] (block_info.closure=%p) (next=%p) ,",
2406 (node_loc==tso_loc ? "Local" : "Global"),
2407 tso->id, tso, CurrentProc, tso->block_info.closure, tso->link));
2408 tso->block_info.closure = NULL;
2409 IF_DEBUG(scheduler,belch("-- Waking up thread %ld (%p)",
2413 static StgBlockingQueueElement *
2414 unblockOneLocked(StgBlockingQueueElement *bqe, StgClosure *node)
2416 StgBlockingQueueElement *next;
2418 switch (get_itbl(bqe)->type) {
2420 ASSERT(((StgTSO *)bqe)->why_blocked != NotBlocked);
2421 /* if it's a TSO just push it onto the run_queue */
2423 // ((StgTSO *)bqe)->link = END_TSO_QUEUE; // debugging?
2424 PUSH_ON_RUN_QUEUE((StgTSO *)bqe);
2426 unblockCount(bqe, node);
2427 /* reset blocking status after dumping event */
2428 ((StgTSO *)bqe)->why_blocked = NotBlocked;
2432 /* if it's a BLOCKED_FETCH put it on the PendingFetches list */
2434 bqe->link = (StgBlockingQueueElement *)PendingFetches;
2435 PendingFetches = (StgBlockedFetch *)bqe;
2439 /* can ignore this case in a non-debugging setup;
2440 see comments on RBHSave closures above */
2442 /* check that the closure is an RBHSave closure */
2443 ASSERT(get_itbl((StgClosure *)bqe) == &stg_RBH_Save_0_info ||
2444 get_itbl((StgClosure *)bqe) == &stg_RBH_Save_1_info ||
2445 get_itbl((StgClosure *)bqe) == &stg_RBH_Save_2_info);
2449 barf("{unblockOneLocked}Daq Qagh: Unexpected IP (%#lx; %s) in blocking queue at %#lx\n",
2450 get_itbl((StgClosure *)bqe), info_type((StgClosure *)bqe),
2454 IF_PAR_DEBUG(bq, fprintf(stderr, ", %p (%s)", bqe, info_type((StgClosure*)bqe)));
2458 #else /* !GRAN && !PAR */
2460 unblockOneLocked(StgTSO *tso)
2464 ASSERT(get_itbl(tso)->type == TSO);
2465 ASSERT(tso->why_blocked != NotBlocked);
2466 tso->why_blocked = NotBlocked;
2468 PUSH_ON_RUN_QUEUE(tso);
2470 IF_DEBUG(scheduler,sched_belch("waking up thread %ld", tso->id));
2475 #if defined(GRAN) || defined(PAR)
2476 INLINE_ME StgBlockingQueueElement *
2477 unblockOne(StgBlockingQueueElement *bqe, StgClosure *node)
2479 ACQUIRE_LOCK(&sched_mutex);
2480 bqe = unblockOneLocked(bqe, node);
2481 RELEASE_LOCK(&sched_mutex);
2486 unblockOne(StgTSO *tso)
2488 ACQUIRE_LOCK(&sched_mutex);
2489 tso = unblockOneLocked(tso);
2490 RELEASE_LOCK(&sched_mutex);
2497 awakenBlockedQueue(StgBlockingQueueElement *q, StgClosure *node)
2499 StgBlockingQueueElement *bqe;
2504 belch("##-_ AwBQ for node %p on PE %d @ %ld by TSO %d (%p): ", \
2505 node, CurrentProc, CurrentTime[CurrentProc],
2506 CurrentTSO->id, CurrentTSO));
2508 node_loc = where_is(node);
2510 ASSERT(q == END_BQ_QUEUE ||
2511 get_itbl(q)->type == TSO || // q is either a TSO or an RBHSave
2512 get_itbl(q)->type == CONSTR); // closure (type constructor)
2513 ASSERT(is_unique(node));
2515 /* FAKE FETCH: magically copy the node to the tso's proc;
2516 no Fetch necessary because in reality the node should not have been
2517 moved to the other PE in the first place
2519 if (CurrentProc!=node_loc) {
2521 belch("## node %p is on PE %d but CurrentProc is %d (TSO %d); assuming fake fetch and adjusting bitmask (old: %#x)",
2522 node, node_loc, CurrentProc, CurrentTSO->id,
2523 // CurrentTSO, where_is(CurrentTSO),
2524 node->header.gran.procs));
2525 node->header.gran.procs = (node->header.gran.procs) | PE_NUMBER(CurrentProc);
2527 belch("## new bitmask of node %p is %#x",
2528 node, node->header.gran.procs));
2529 if (RtsFlags.GranFlags.GranSimStats.Global) {
2530 globalGranStats.tot_fake_fetches++;
2535 // ToDo: check: ASSERT(CurrentProc==node_loc);
2536 while (get_itbl(bqe)->type==TSO) { // q != END_TSO_QUEUE) {
2539 bqe points to the current element in the queue
2540 next points to the next element in the queue
2542 //tso = (StgTSO *)bqe; // wastes an assignment to get the type right
2543 //tso_loc = where_is(tso);
2545 bqe = unblockOneLocked(bqe, node);
2548 /* if this is the BQ of an RBH, we have to put back the info ripped out of
2549 the closure to make room for the anchor of the BQ */
2550 if (bqe!=END_BQ_QUEUE) {
2551 ASSERT(get_itbl(node)->type == RBH && get_itbl(bqe)->type == CONSTR);
2553 ASSERT((info_ptr==&RBH_Save_0_info) ||
2554 (info_ptr==&RBH_Save_1_info) ||
2555 (info_ptr==&RBH_Save_2_info));
2557 /* cf. convertToRBH in RBH.c for writing the RBHSave closure */
2558 ((StgRBH *)node)->blocking_queue = (StgBlockingQueueElement *)((StgRBHSave *)bqe)->payload[0];
2559 ((StgRBH *)node)->mut_link = (StgMutClosure *)((StgRBHSave *)bqe)->payload[1];
2562 belch("## Filled in RBH_Save for %p (%s) at end of AwBQ",
2563 node, info_type(node)));
2566 /* statistics gathering */
2567 if (RtsFlags.GranFlags.GranSimStats.Global) {
2568 // globalGranStats.tot_bq_processing_time += bq_processing_time;
2569 globalGranStats.tot_bq_len += len; // total length of all bqs awakened
2570 // globalGranStats.tot_bq_len_local += len_local; // same for local TSOs only
2571 globalGranStats.tot_awbq++; // total no. of bqs awakened
2574 fprintf(stderr,"## BQ Stats of %p: [%d entries] %s\n",
2575 node, len, (bqe!=END_BQ_QUEUE) ? "RBH" : ""));
2579 awakenBlockedQueue(StgBlockingQueueElement *q, StgClosure *node)
2581 StgBlockingQueueElement *bqe;
2583 ACQUIRE_LOCK(&sched_mutex);
2585 IF_PAR_DEBUG(verbose,
2586 belch("##-_ AwBQ for node %p on [%x]: ",
2590 if(get_itbl(q)->type == CONSTR || q==END_BQ_QUEUE) {
2591 IF_PAR_DEBUG(verbose, belch("## ... nothing to unblock so lets just return. RFP (BUG?)"));
2596 ASSERT(q == END_BQ_QUEUE ||
2597 get_itbl(q)->type == TSO ||
2598 get_itbl(q)->type == BLOCKED_FETCH ||
2599 get_itbl(q)->type == CONSTR);
2602 while (get_itbl(bqe)->type==TSO ||
2603 get_itbl(bqe)->type==BLOCKED_FETCH) {
2604 bqe = unblockOneLocked(bqe, node);
2606 RELEASE_LOCK(&sched_mutex);
2609 #else /* !GRAN && !PAR */
2612 awakenBlockedQueueNoLock(StgTSO *tso)
2614 while (tso != END_TSO_QUEUE) {
2615 tso = unblockOneLocked(tso);
2620 awakenBlockedQueue(StgTSO *tso)
2622 ACQUIRE_LOCK(&sched_mutex);
2623 while (tso != END_TSO_QUEUE) {
2624 tso = unblockOneLocked(tso);
2626 RELEASE_LOCK(&sched_mutex);
2630 /* ---------------------------------------------------------------------------
2632 - usually called inside a signal handler so it mustn't do anything fancy.
2633 ------------------------------------------------------------------------ */
2636 interruptStgRts(void)
2640 #ifdef RTS_SUPPORTS_THREADS
2641 wakeBlockedWorkerThread();
2645 /* -----------------------------------------------------------------------------
2648 This is for use when we raise an exception in another thread, which
2650 This has nothing to do with the UnblockThread event in GranSim. -- HWL
2651 -------------------------------------------------------------------------- */
2653 #if defined(GRAN) || defined(PAR)
2655 NB: only the type of the blocking queue is different in GranSim and GUM
2656 the operations on the queue-elements are the same
2657 long live polymorphism!
2659 Locks: sched_mutex is held upon entry and exit.
2663 unblockThread(StgTSO *tso)
2665 StgBlockingQueueElement *t, **last;
2667 switch (tso->why_blocked) {
2670 return; /* not blocked */
2673 ASSERT(get_itbl(tso->block_info.closure)->type == MVAR);
2675 StgBlockingQueueElement *last_tso = END_BQ_QUEUE;
2676 StgMVar *mvar = (StgMVar *)(tso->block_info.closure);
2678 last = (StgBlockingQueueElement **)&mvar->head;
2679 for (t = (StgBlockingQueueElement *)mvar->head;
2681 last = &t->link, last_tso = t, t = t->link) {
2682 if (t == (StgBlockingQueueElement *)tso) {
2683 *last = (StgBlockingQueueElement *)tso->link;
2684 if (mvar->tail == tso) {
2685 mvar->tail = (StgTSO *)last_tso;
2690 barf("unblockThread (MVAR): TSO not found");
2693 case BlockedOnBlackHole:
2694 ASSERT(get_itbl(tso->block_info.closure)->type == BLACKHOLE_BQ);
2696 StgBlockingQueue *bq = (StgBlockingQueue *)(tso->block_info.closure);
2698 last = &bq->blocking_queue;
2699 for (t = bq->blocking_queue;
2701 last = &t->link, t = t->link) {
2702 if (t == (StgBlockingQueueElement *)tso) {
2703 *last = (StgBlockingQueueElement *)tso->link;
2707 barf("unblockThread (BLACKHOLE): TSO not found");
2710 case BlockedOnException:
2712 StgTSO *target = tso->block_info.tso;
2714 ASSERT(get_itbl(target)->type == TSO);
2716 if (target->what_next == ThreadRelocated) {
2717 target = target->link;
2718 ASSERT(get_itbl(target)->type == TSO);
2721 ASSERT(target->blocked_exceptions != NULL);
2723 last = (StgBlockingQueueElement **)&target->blocked_exceptions;
2724 for (t = (StgBlockingQueueElement *)target->blocked_exceptions;
2726 last = &t->link, t = t->link) {
2727 ASSERT(get_itbl(t)->type == TSO);
2728 if (t == (StgBlockingQueueElement *)tso) {
2729 *last = (StgBlockingQueueElement *)tso->link;
2733 barf("unblockThread (Exception): TSO not found");
2737 case BlockedOnWrite:
2738 #if defined(mingw32_TARGET_OS)
2739 case BlockedOnDoProc:
2742 /* take TSO off blocked_queue */
2743 StgBlockingQueueElement *prev = NULL;
2744 for (t = (StgBlockingQueueElement *)blocked_queue_hd; t != END_BQ_QUEUE;
2745 prev = t, t = t->link) {
2746 if (t == (StgBlockingQueueElement *)tso) {
2748 blocked_queue_hd = (StgTSO *)t->link;
2749 if ((StgBlockingQueueElement *)blocked_queue_tl == t) {
2750 blocked_queue_tl = END_TSO_QUEUE;
2753 prev->link = t->link;
2754 if ((StgBlockingQueueElement *)blocked_queue_tl == t) {
2755 blocked_queue_tl = (StgTSO *)prev;
2761 barf("unblockThread (I/O): TSO not found");
2764 case BlockedOnDelay:
2766 /* take TSO off sleeping_queue */
2767 StgBlockingQueueElement *prev = NULL;
2768 for (t = (StgBlockingQueueElement *)sleeping_queue; t != END_BQ_QUEUE;
2769 prev = t, t = t->link) {
2770 if (t == (StgBlockingQueueElement *)tso) {
2772 sleeping_queue = (StgTSO *)t->link;
2774 prev->link = t->link;
2779 barf("unblockThread (delay): TSO not found");
2783 barf("unblockThread");
2787 tso->link = END_TSO_QUEUE;
2788 tso->why_blocked = NotBlocked;
2789 tso->block_info.closure = NULL;
2790 PUSH_ON_RUN_QUEUE(tso);
2794 unblockThread(StgTSO *tso)
2798 /* To avoid locking unnecessarily. */
2799 if (tso->why_blocked == NotBlocked) {
2803 switch (tso->why_blocked) {
2806 ASSERT(get_itbl(tso->block_info.closure)->type == MVAR);
2808 StgTSO *last_tso = END_TSO_QUEUE;
2809 StgMVar *mvar = (StgMVar *)(tso->block_info.closure);
2812 for (t = mvar->head; t != END_TSO_QUEUE;
2813 last = &t->link, last_tso = t, t = t->link) {
2816 if (mvar->tail == tso) {
2817 mvar->tail = last_tso;
2822 barf("unblockThread (MVAR): TSO not found");
2825 case BlockedOnBlackHole:
2826 ASSERT(get_itbl(tso->block_info.closure)->type == BLACKHOLE_BQ);
2828 StgBlockingQueue *bq = (StgBlockingQueue *)(tso->block_info.closure);
2830 last = &bq->blocking_queue;
2831 for (t = bq->blocking_queue; t != END_TSO_QUEUE;
2832 last = &t->link, t = t->link) {
2838 barf("unblockThread (BLACKHOLE): TSO not found");
2841 case BlockedOnException:
2843 StgTSO *target = tso->block_info.tso;
2845 ASSERT(get_itbl(target)->type == TSO);
2847 while (target->what_next == ThreadRelocated) {
2848 target = target->link;
2849 ASSERT(get_itbl(target)->type == TSO);
2852 ASSERT(target->blocked_exceptions != NULL);
2854 last = &target->blocked_exceptions;
2855 for (t = target->blocked_exceptions; t != END_TSO_QUEUE;
2856 last = &t->link, t = t->link) {
2857 ASSERT(get_itbl(t)->type == TSO);
2863 barf("unblockThread (Exception): TSO not found");
2867 case BlockedOnWrite:
2868 #if defined(mingw32_TARGET_OS)
2869 case BlockedOnDoProc:
2872 StgTSO *prev = NULL;
2873 for (t = blocked_queue_hd; t != END_TSO_QUEUE;
2874 prev = t, t = t->link) {
2877 blocked_queue_hd = t->link;
2878 if (blocked_queue_tl == t) {
2879 blocked_queue_tl = END_TSO_QUEUE;
2882 prev->link = t->link;
2883 if (blocked_queue_tl == t) {
2884 blocked_queue_tl = prev;
2890 barf("unblockThread (I/O): TSO not found");
2893 case BlockedOnDelay:
2895 StgTSO *prev = NULL;
2896 for (t = sleeping_queue; t != END_TSO_QUEUE;
2897 prev = t, t = t->link) {
2900 sleeping_queue = t->link;
2902 prev->link = t->link;
2907 barf("unblockThread (delay): TSO not found");
2911 barf("unblockThread");
2915 tso->link = END_TSO_QUEUE;
2916 tso->why_blocked = NotBlocked;
2917 tso->block_info.closure = NULL;
2918 PUSH_ON_RUN_QUEUE(tso);
2922 /* -----------------------------------------------------------------------------
2925 * The following function implements the magic for raising an
2926 * asynchronous exception in an existing thread.
2928 * We first remove the thread from any queue on which it might be
2929 * blocked. The possible blockages are MVARs and BLACKHOLE_BQs.
2931 * We strip the stack down to the innermost CATCH_FRAME, building
2932 * thunks in the heap for all the active computations, so they can
2933 * be restarted if necessary. When we reach a CATCH_FRAME, we build
2934 * an application of the handler to the exception, and push it on
2935 * the top of the stack.
2937 * How exactly do we save all the active computations? We create an
2938 * AP_STACK for every UpdateFrame on the stack. Entering one of these
2939 * AP_STACKs pushes everything from the corresponding update frame
2940 * upwards onto the stack. (Actually, it pushes everything up to the
2941 * next update frame plus a pointer to the next AP_STACK object.
2942 * Entering the next AP_STACK object pushes more onto the stack until we
2943 * reach the last AP_STACK object - at which point the stack should look
2944 * exactly as it did when we killed the TSO and we can continue
2945 * execution by entering the closure on top of the stack.
2947 * We can also kill a thread entirely - this happens if either (a) the
2948 * exception passed to raiseAsync is NULL, or (b) there's no
2949 * CATCH_FRAME on the stack. In either case, we strip the entire
2950 * stack and replace the thread with a zombie.
2952 * Locks: sched_mutex held upon entry nor exit.
2954 * -------------------------------------------------------------------------- */
2957 deleteThread(StgTSO *tso)
2959 raiseAsync(tso,NULL);
2962 #ifdef FORKPROCESS_PRIMOP_SUPPORTED
2964 deleteThreadImmediately(StgTSO *tso)
2965 { // for forkProcess only:
2966 // delete thread without giving it a chance to catch the KillThread exception
2968 if (tso->what_next == ThreadComplete || tso->what_next == ThreadKilled) {
2972 if (tso->why_blocked != BlockedOnCCall &&
2973 tso->why_blocked != BlockedOnCCall_NoUnblockExc) {
2977 tso->what_next = ThreadKilled;
2982 raiseAsyncWithLock(StgTSO *tso, StgClosure *exception)
2984 /* When raising async exs from contexts where sched_mutex isn't held;
2985 use raiseAsyncWithLock(). */
2986 ACQUIRE_LOCK(&sched_mutex);
2987 raiseAsync(tso,exception);
2988 RELEASE_LOCK(&sched_mutex);
2992 raiseAsync(StgTSO *tso, StgClosure *exception)
2994 StgRetInfoTable *info;
2997 // Thread already dead?
2998 if (tso->what_next == ThreadComplete || tso->what_next == ThreadKilled) {
3003 sched_belch("raising exception in thread %ld.", tso->id));
3005 // Remove it from any blocking queues
3010 // The stack freezing code assumes there's a closure pointer on
3011 // the top of the stack, so we have to arrange that this is the case...
3013 if (sp[0] == (W_)&stg_enter_info) {
3017 sp[0] = (W_)&stg_dummy_ret_closure;
3023 // 1. Let the top of the stack be the "current closure"
3025 // 2. Walk up the stack until we find either an UPDATE_FRAME or a
3028 // 3. If it's an UPDATE_FRAME, then make an AP_STACK containing the
3029 // current closure applied to the chunk of stack up to (but not
3030 // including) the update frame. This closure becomes the "current
3031 // closure". Go back to step 2.
3033 // 4. If it's a CATCH_FRAME, then leave the exception handler on
3034 // top of the stack applied to the exception.
3036 // 5. If it's a STOP_FRAME, then kill the thread.
3041 info = get_ret_itbl((StgClosure *)frame);
3043 while (info->i.type != UPDATE_FRAME
3044 && (info->i.type != CATCH_FRAME || exception == NULL)
3045 && info->i.type != STOP_FRAME) {
3046 frame += stack_frame_sizeW((StgClosure *)frame);
3047 info = get_ret_itbl((StgClosure *)frame);
3050 switch (info->i.type) {
3053 // If we find a CATCH_FRAME, and we've got an exception to raise,
3054 // then build the THUNK raise(exception), and leave it on
3055 // top of the CATCH_FRAME ready to enter.
3059 StgCatchFrame *cf = (StgCatchFrame *)frame;
3063 // we've got an exception to raise, so let's pass it to the
3064 // handler in this frame.
3066 raise = (StgClosure *)allocate(sizeofW(StgClosure)+1);
3067 TICK_ALLOC_SE_THK(1,0);
3068 SET_HDR(raise,&stg_raise_info,cf->header.prof.ccs);
3069 raise->payload[0] = exception;
3071 // throw away the stack from Sp up to the CATCH_FRAME.
3075 /* Ensure that async excpetions are blocked now, so we don't get
3076 * a surprise exception before we get around to executing the
3079 if (tso->blocked_exceptions == NULL) {
3080 tso->blocked_exceptions = END_TSO_QUEUE;
3083 /* Put the newly-built THUNK on top of the stack, ready to execute
3084 * when the thread restarts.
3087 sp[-1] = (W_)&stg_enter_info;
3089 tso->what_next = ThreadRunGHC;
3090 IF_DEBUG(sanity, checkTSO(tso));
3099 // First build an AP_STACK consisting of the stack chunk above the
3100 // current update frame, with the top word on the stack as the
3103 words = frame - sp - 1;
3104 ap = (StgAP_STACK *)allocate(PAP_sizeW(words));
3107 ap->fun = (StgClosure *)sp[0];
3109 for(i=0; i < (nat)words; ++i) {
3110 ap->payload[i] = (StgClosure *)*sp++;
3113 SET_HDR(ap,&stg_AP_STACK_info,
3114 ((StgClosure *)frame)->header.prof.ccs /* ToDo */);
3115 TICK_ALLOC_UP_THK(words+1,0);
3118 fprintf(stderr, "sched: Updating ");
3119 printPtr((P_)((StgUpdateFrame *)frame)->updatee);
3120 fprintf(stderr, " with ");
3121 printObj((StgClosure *)ap);
3124 // Replace the updatee with an indirection - happily
3125 // this will also wake up any threads currently
3126 // waiting on the result.
3128 // Warning: if we're in a loop, more than one update frame on
3129 // the stack may point to the same object. Be careful not to
3130 // overwrite an IND_OLDGEN in this case, because we'll screw
3131 // up the mutable lists. To be on the safe side, don't
3132 // overwrite any kind of indirection at all. See also
3133 // threadSqueezeStack in GC.c, where we have to make a similar
3136 if (!closure_IND(((StgUpdateFrame *)frame)->updatee)) {
3137 // revert the black hole
3138 UPD_IND_NOLOCK(((StgUpdateFrame *)frame)->updatee,ap);
3140 sp += sizeofW(StgUpdateFrame) - 1;
3141 sp[0] = (W_)ap; // push onto stack
3146 // We've stripped the entire stack, the thread is now dead.
3147 sp += sizeofW(StgStopFrame);
3148 tso->what_next = ThreadKilled;
3159 /* -----------------------------------------------------------------------------
3160 resurrectThreads is called after garbage collection on the list of
3161 threads found to be garbage. Each of these threads will be woken
3162 up and sent a signal: BlockedOnDeadMVar if the thread was blocked
3163 on an MVar, or NonTermination if the thread was blocked on a Black
3166 Locks: sched_mutex isn't held upon entry nor exit.
3167 -------------------------------------------------------------------------- */
3170 resurrectThreads( StgTSO *threads )
3174 for (tso = threads; tso != END_TSO_QUEUE; tso = next) {
3175 next = tso->global_link;
3176 tso->global_link = all_threads;
3178 IF_DEBUG(scheduler, sched_belch("resurrecting thread %d", tso->id));
3180 switch (tso->why_blocked) {
3182 case BlockedOnException:
3183 /* Called by GC - sched_mutex lock is currently held. */
3184 raiseAsync(tso,(StgClosure *)BlockedOnDeadMVar_closure);
3186 case BlockedOnBlackHole:
3187 raiseAsync(tso,(StgClosure *)NonTermination_closure);
3190 /* This might happen if the thread was blocked on a black hole
3191 * belonging to a thread that we've just woken up (raiseAsync
3192 * can wake up threads, remember...).
3196 barf("resurrectThreads: thread blocked in a strange way");
3201 /* -----------------------------------------------------------------------------
3202 * Blackhole detection: if we reach a deadlock, test whether any
3203 * threads are blocked on themselves. Any threads which are found to
3204 * be self-blocked get sent a NonTermination exception.
3206 * This is only done in a deadlock situation in order to avoid
3207 * performance overhead in the normal case.
3209 * Locks: sched_mutex is held upon entry and exit.
3210 * -------------------------------------------------------------------------- */
3213 detectBlackHoles( void )
3215 StgTSO *tso = all_threads;
3217 StgClosure *blocked_on;
3218 StgRetInfoTable *info;
3220 for (tso = all_threads; tso != END_TSO_QUEUE; tso = tso->global_link) {
3222 while (tso->what_next == ThreadRelocated) {
3224 ASSERT(get_itbl(tso)->type == TSO);
3227 if (tso->why_blocked != BlockedOnBlackHole) {
3230 blocked_on = tso->block_info.closure;
3232 frame = (StgClosure *)tso->sp;
3235 info = get_ret_itbl(frame);
3236 switch (info->i.type) {
3238 if (((StgUpdateFrame *)frame)->updatee == blocked_on) {
3239 /* We are blocking on one of our own computations, so
3240 * send this thread the NonTermination exception.
3243 sched_belch("thread %d is blocked on itself", tso->id));
3244 raiseAsync(tso, (StgClosure *)NonTermination_closure);
3248 frame = (StgClosure *) ((StgUpdateFrame *)frame + 1);
3254 // normal stack frames; do nothing except advance the pointer
3256 (StgPtr)frame += stack_frame_sizeW(frame);
3263 /* ----------------------------------------------------------------------------
3264 * Debugging: why is a thread blocked
3265 * [Also provides useful information when debugging threaded programs
3266 * at the Haskell source code level, so enable outside of DEBUG. --sof 7/02]
3267 ------------------------------------------------------------------------- */
3271 printThreadBlockage(StgTSO *tso)
3273 switch (tso->why_blocked) {
3275 fprintf(stderr,"is blocked on read from fd %d", tso->block_info.fd);
3277 case BlockedOnWrite:
3278 fprintf(stderr,"is blocked on write to fd %d", tso->block_info.fd);
3280 #if defined(mingw32_TARGET_OS)
3281 case BlockedOnDoProc:
3282 fprintf(stderr,"is blocked on proc (request: %d)", tso->block_info.async_result->reqID);
3285 case BlockedOnDelay:
3286 fprintf(stderr,"is blocked until %d", tso->block_info.target);
3289 fprintf(stderr,"is blocked on an MVar");
3291 case BlockedOnException:
3292 fprintf(stderr,"is blocked on delivering an exception to thread %d",
3293 tso->block_info.tso->id);
3295 case BlockedOnBlackHole:
3296 fprintf(stderr,"is blocked on a black hole");
3299 fprintf(stderr,"is not blocked");
3303 fprintf(stderr,"is blocked on global address; local FM_BQ is %p (%s)",
3304 tso->block_info.closure, info_type(tso->block_info.closure));
3306 case BlockedOnGA_NoSend:
3307 fprintf(stderr,"is blocked on global address (no send); local FM_BQ is %p (%s)",
3308 tso->block_info.closure, info_type(tso->block_info.closure));
3311 case BlockedOnCCall:
3312 fprintf(stderr,"is blocked on an external call");
3314 case BlockedOnCCall_NoUnblockExc:
3315 fprintf(stderr,"is blocked on an external call (exceptions were already blocked)");
3318 barf("printThreadBlockage: strange tso->why_blocked: %d for TSO %d (%d)",
3319 tso->why_blocked, tso->id, tso);
3325 printThreadStatus(StgTSO *tso)
3327 switch (tso->what_next) {
3329 fprintf(stderr,"has been killed");
3331 case ThreadComplete:
3332 fprintf(stderr,"has completed");
3335 printThreadBlockage(tso);
3340 printAllThreads(void)
3346 char time_string[TIME_STR_LEN], node_str[NODE_STR_LEN];
3347 ullong_format_string(TIME_ON_PROC(CurrentProc),
3348 time_string, rtsFalse/*no commas!*/);
3350 fprintf(stderr, "all threads at [%s]:\n", time_string);
3352 char time_string[TIME_STR_LEN], node_str[NODE_STR_LEN];
3353 ullong_format_string(CURRENT_TIME,
3354 time_string, rtsFalse/*no commas!*/);
3356 fprintf(stderr,"all threads at [%s]:\n", time_string);
3358 fprintf(stderr,"all threads:\n");
3361 for (t = all_threads; t != END_TSO_QUEUE; t = t->global_link) {
3362 fprintf(stderr, "\tthread %d @ %p ", t->id, (void *)t);
3363 label = lookupThreadLabel(t->id);
3364 if (label) fprintf(stderr,"[\"%s\"] ",(char *)label);
3365 printThreadStatus(t);
3366 fprintf(stderr,"\n");
3373 Print a whole blocking queue attached to node (debugging only).
3377 print_bq (StgClosure *node)
3379 StgBlockingQueueElement *bqe;
3383 fprintf(stderr,"## BQ of closure %p (%s): ",
3384 node, info_type(node));
3386 /* should cover all closures that may have a blocking queue */
3387 ASSERT(get_itbl(node)->type == BLACKHOLE_BQ ||
3388 get_itbl(node)->type == FETCH_ME_BQ ||
3389 get_itbl(node)->type == RBH ||
3390 get_itbl(node)->type == MVAR);
3392 ASSERT(node!=(StgClosure*)NULL); // sanity check
3394 print_bqe(((StgBlockingQueue*)node)->blocking_queue);
3398 Print a whole blocking queue starting with the element bqe.
3401 print_bqe (StgBlockingQueueElement *bqe)
3406 NB: In a parallel setup a BQ of an RBH must end with an RBH_Save closure;
3408 for (end = (bqe==END_BQ_QUEUE);
3409 !end; // iterate until bqe points to a CONSTR
3410 end = (get_itbl(bqe)->type == CONSTR) || (bqe->link==END_BQ_QUEUE),
3411 bqe = end ? END_BQ_QUEUE : bqe->link) {
3412 ASSERT(bqe != END_BQ_QUEUE); // sanity check
3413 ASSERT(bqe != (StgBlockingQueueElement *)NULL); // sanity check
3414 /* types of closures that may appear in a blocking queue */
3415 ASSERT(get_itbl(bqe)->type == TSO ||
3416 get_itbl(bqe)->type == BLOCKED_FETCH ||
3417 get_itbl(bqe)->type == CONSTR);
3418 /* only BQs of an RBH end with an RBH_Save closure */
3419 //ASSERT(get_itbl(bqe)->type != CONSTR || get_itbl(node)->type == RBH);
3421 switch (get_itbl(bqe)->type) {
3423 fprintf(stderr," TSO %u (%x),",
3424 ((StgTSO *)bqe)->id, ((StgTSO *)bqe));
3427 fprintf(stderr," BF (node=%p, ga=((%x, %d, %x)),",
3428 ((StgBlockedFetch *)bqe)->node,
3429 ((StgBlockedFetch *)bqe)->ga.payload.gc.gtid,
3430 ((StgBlockedFetch *)bqe)->ga.payload.gc.slot,
3431 ((StgBlockedFetch *)bqe)->ga.weight);
3434 fprintf(stderr," %s (IP %p),",
3435 (get_itbl(bqe) == &stg_RBH_Save_0_info ? "RBH_Save_0" :
3436 get_itbl(bqe) == &stg_RBH_Save_1_info ? "RBH_Save_1" :
3437 get_itbl(bqe) == &stg_RBH_Save_2_info ? "RBH_Save_2" :
3438 "RBH_Save_?"), get_itbl(bqe));
3441 barf("Unexpected closure type %s in blocking queue", // of %p (%s)",
3442 info_type((StgClosure *)bqe)); // , node, info_type(node));
3446 fputc('\n', stderr);
3448 # elif defined(GRAN)
3450 print_bq (StgClosure *node)
3452 StgBlockingQueueElement *bqe;
3453 PEs node_loc, tso_loc;
3456 /* should cover all closures that may have a blocking queue */
3457 ASSERT(get_itbl(node)->type == BLACKHOLE_BQ ||
3458 get_itbl(node)->type == FETCH_ME_BQ ||
3459 get_itbl(node)->type == RBH);
3461 ASSERT(node!=(StgClosure*)NULL); // sanity check
3462 node_loc = where_is(node);
3464 fprintf(stderr,"## BQ of closure %p (%s) on [PE %d]: ",
3465 node, info_type(node), node_loc);
3468 NB: In a parallel setup a BQ of an RBH must end with an RBH_Save closure;
3470 for (bqe = ((StgBlockingQueue*)node)->blocking_queue, end = (bqe==END_BQ_QUEUE);
3471 !end; // iterate until bqe points to a CONSTR
3472 end = (get_itbl(bqe)->type == CONSTR) || (bqe->link==END_BQ_QUEUE), bqe = end ? END_BQ_QUEUE : bqe->link) {
3473 ASSERT(bqe != END_BQ_QUEUE); // sanity check
3474 ASSERT(bqe != (StgBlockingQueueElement *)NULL); // sanity check
3475 /* types of closures that may appear in a blocking queue */
3476 ASSERT(get_itbl(bqe)->type == TSO ||
3477 get_itbl(bqe)->type == CONSTR);
3478 /* only BQs of an RBH end with an RBH_Save closure */
3479 ASSERT(get_itbl(bqe)->type != CONSTR || get_itbl(node)->type == RBH);
3481 tso_loc = where_is((StgClosure *)bqe);
3482 switch (get_itbl(bqe)->type) {
3484 fprintf(stderr," TSO %d (%p) on [PE %d],",
3485 ((StgTSO *)bqe)->id, (StgTSO *)bqe, tso_loc);
3488 fprintf(stderr," %s (IP %p),",
3489 (get_itbl(bqe) == &stg_RBH_Save_0_info ? "RBH_Save_0" :
3490 get_itbl(bqe) == &stg_RBH_Save_1_info ? "RBH_Save_1" :
3491 get_itbl(bqe) == &stg_RBH_Save_2_info ? "RBH_Save_2" :
3492 "RBH_Save_?"), get_itbl(bqe));
3495 barf("Unexpected closure type %s in blocking queue of %p (%s)",
3496 info_type((StgClosure *)bqe), node, info_type(node));
3500 fputc('\n', stderr);
3504 Nice and easy: only TSOs on the blocking queue
3507 print_bq (StgClosure *node)
3511 ASSERT(node!=(StgClosure*)NULL); // sanity check
3512 for (tso = ((StgBlockingQueue*)node)->blocking_queue;
3513 tso != END_TSO_QUEUE;
3515 ASSERT(tso!=NULL && tso!=END_TSO_QUEUE); // sanity check
3516 ASSERT(get_itbl(tso)->type == TSO); // guess what, sanity check
3517 fprintf(stderr," TSO %d (%p),", tso->id, tso);
3519 fputc('\n', stderr);
3530 for (i=0, tso=run_queue_hd;
3531 tso != END_TSO_QUEUE;
3540 sched_belch(char *s, ...)
3544 #ifdef RTS_SUPPORTS_THREADS
3545 fprintf(stderr, "sched (task %p): ", osThreadId());
3547 fprintf(stderr, "== ");
3549 fprintf(stderr, "sched: ");
3551 vfprintf(stderr, s, ap);
3552 fprintf(stderr, "\n");