1 /* ---------------------------------------------------------------------------
2 * $Id: Schedule.c,v 1.117 2002/02/05 10:06:24 simonmar Exp $
4 * (c) The GHC Team, 1998-2000
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 distributed 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 * --------------------------------------------------------------------------*/
20 //@node Main scheduling code, , ,
21 //@section Main scheduling code
24 * Version with scheduler monitor support for SMPs (WAY=s):
26 This design provides a high-level API to create and schedule threads etc.
27 as documented in the SMP design document.
29 It uses a monitor design controlled by a single mutex to exercise control
30 over accesses to shared data structures, and builds on the Posix threads
33 The majority of state is shared. In order to keep essential per-task state,
34 there is a Capability structure, which contains all the information
35 needed to run a thread: its STG registers, a pointer to its TSO, a
36 nursery etc. During STG execution, a pointer to the capability is
37 kept in a register (BaseReg).
39 In a non-SMP build, there is one global capability, namely MainRegTable.
43 * Version with support for distributed memory parallelism aka GUM (WAY=mp):
45 The main scheduling loop in GUM iterates until a finish message is received.
46 In that case a global flag @receivedFinish@ is set and this instance of
47 the RTS shuts down. See ghc/rts/parallel/HLComms.c:processMessages()
48 for the handling of incoming messages, such as PP_FINISH.
49 Note that in the parallel case we have a system manager that coordinates
50 different PEs, each of which are running one instance of the RTS.
51 See ghc/rts/parallel/SysMan.c for the main routine of the parallel program.
52 From this routine processes executing ghc/rts/Main.c are spawned. -- HWL
54 * Version with support for simulating parallel execution aka GranSim (WAY=mg):
56 The main scheduling code in GranSim is quite different from that in std
57 (concurrent) Haskell: while concurrent Haskell just iterates over the
58 threads in the runnable queue, GranSim is event driven, i.e. it iterates
59 over the events in the global event queue. -- HWL
64 //* Variables and Data structures::
65 //* Main scheduling loop::
66 //* Suspend and Resume::
68 //* Garbage Collextion Routines::
69 //* Blocking Queue Routines::
70 //* Exception Handling Routines::
71 //* Debugging Routines::
75 //@node Includes, Variables and Data structures, Main scheduling code, Main scheduling code
76 //@subsection Includes
78 #include "PosixSource.h"
85 #include "StgStartup.h"
88 #include "StgMiscClosures.h"
90 #include "Interpreter.h"
91 #include "Exception.h"
100 #include "Proftimer.h"
101 #include "ProfHeap.h"
103 #if defined(GRAN) || defined(PAR)
104 # include "GranSimRts.h"
105 # include "GranSim.h"
106 # include "ParallelRts.h"
107 # include "Parallel.h"
108 # include "ParallelDebug.h"
109 # include "FetchMe.h"
113 #include "Capability.h"
114 #include "OSThreads.h"
119 //@node Variables and Data structures, Prototypes, Includes, Main scheduling code
120 //@subsection Variables and Data structures
124 * These are the threads which clients have requested that we run.
126 * In a 'threaded' build, we might have several concurrent clients all
127 * waiting for results, and each one will wait on a condition variable
128 * until the result is available.
130 * In non-SMP, clients are strictly nested: the first client calls
131 * into the RTS, which might call out again to C with a _ccall_GC, and
132 * eventually re-enter the RTS.
134 * Main threads information is kept in a linked list:
136 //@cindex StgMainThread
137 typedef struct StgMainThread_ {
139 SchedulerStatus stat;
141 #if defined(RTS_SUPPORTS_THREADS)
144 struct StgMainThread_ *link;
147 /* Main thread queue.
148 * Locks required: sched_mutex.
150 static StgMainThread *main_threads;
153 * Locks required: sched_mutex.
157 StgTSO* ActiveTSO = NULL; /* for assigning system costs; GranSim-Light only */
158 /* rtsTime TimeOfNextEvent, EndOfTimeSlice; now in GranSim.c */
161 In GranSim we have a runable and a blocked queue for each processor.
162 In order to minimise code changes new arrays run_queue_hds/tls
163 are created. run_queue_hd is then a short cut (macro) for
164 run_queue_hds[CurrentProc] (see GranSim.h).
167 StgTSO *run_queue_hds[MAX_PROC], *run_queue_tls[MAX_PROC];
168 StgTSO *blocked_queue_hds[MAX_PROC], *blocked_queue_tls[MAX_PROC];
169 StgTSO *ccalling_threadss[MAX_PROC];
170 /* We use the same global list of threads (all_threads) in GranSim as in
171 the std RTS (i.e. we are cheating). However, we don't use this list in
172 the GranSim specific code at the moment (so we are only potentially
177 StgTSO *run_queue_hd, *run_queue_tl;
178 StgTSO *blocked_queue_hd, *blocked_queue_tl;
179 StgTSO *sleeping_queue; /* perhaps replace with a hash table? */
183 /* Linked list of all threads.
184 * Used for detecting garbage collected threads.
188 /* Threads suspended in _ccall_GC.
190 static StgTSO *suspended_ccalling_threads;
192 static StgTSO *threadStackOverflow(StgTSO *tso);
194 /* KH: The following two flags are shared memory locations. There is no need
195 to lock them, since they are only unset at the end of a scheduler
199 /* flag set by signal handler to precipitate a context switch */
200 //@cindex context_switch
203 /* if this flag is set as well, give up execution */
204 //@cindex interrupted
207 /* Next thread ID to allocate.
208 * Locks required: sched_mutex
210 //@cindex next_thread_id
211 StgThreadID next_thread_id = 1;
214 * Pointers to the state of the current thread.
215 * Rule of thumb: if CurrentTSO != NULL, then we're running a Haskell
216 * thread. If CurrentTSO == NULL, then we're at the scheduler level.
219 /* The smallest stack size that makes any sense is:
220 * RESERVED_STACK_WORDS (so we can get back from the stack overflow)
221 * + sizeofW(StgStopFrame) (the stg_stop_thread_info frame)
222 * + 1 (the realworld token for an IO thread)
223 * + 1 (the closure to enter)
225 * A thread with this stack will bomb immediately with a stack
226 * overflow, which will increase its stack size.
229 #define MIN_STACK_WORDS (RESERVED_STACK_WORDS + sizeofW(StgStopFrame) + 2)
236 /* This is used in `TSO.h' and gcc 2.96 insists that this variable actually
237 * exists - earlier gccs apparently didn't.
244 void addToBlockedQueue ( StgTSO *tso );
246 static void schedule ( void );
247 void interruptStgRts ( void );
249 static StgTSO * createThread_ ( nat size, rtsBool have_lock, StgInt pri );
251 static StgTSO * createThread_ ( nat size, rtsBool have_lock );
254 static void detectBlackHoles ( void );
257 static void sched_belch(char *s, ...);
260 #if defined(RTS_SUPPORTS_THREADS)
261 /* ToDo: carefully document the invariants that go together
262 * with these synchronisation objects.
264 Mutex sched_mutex = INIT_MUTEX_VAR;
265 Mutex term_mutex = INIT_MUTEX_VAR;
266 #if defined(THREADED_RTS)
268 * The rts_mutex is the 'big lock' that the active native
269 * thread within the RTS holds while executing code
270 * within the RTS. It is given up when the thread makes a
271 * transition out of the RTS (e.g., to perform an external
272 * C call), hopefully for another thread to enter the RTS.
275 Mutex rts_mutex = INIT_MUTEX_VAR;
277 * When a native thread has completed executing an external
278 * call, it needs to communicate the result back to the
279 * (Haskell) thread that made the call. Do this as follows:
281 * - in resumeThread(), the thread increments the counter
282 * ext_threads_waiting, and then blocks on the
284 * - upon entry to the scheduler, the thread that's currently
285 * holding the RTS lock checks ext_threads_waiting. If there
286 * are native threads waiting, it gives up its RTS lock
287 * and tries to re-grab the RTS lock [perhaps after having
288 * waited for a bit..?]
289 * - care must be taken to deal with the case where more than
290 * one external thread are waiting on the lock. [ToDo: more]
294 static nat ext_threads_waiting = 0;
296 * thread_ready_aux_mutex is used to handle the scenario where the
297 * the RTS executing thread runs out of work, but there are
298 * active external threads. The RTS executing thread gives up
299 * its RTS mutex, and blocks waiting for the thread_ready_cond.
300 * Unfortunately, a condition variable needs to be associated
301 * with a mutex in pthreads, so rts_thread_waiting_mutex is
302 * used for just this purpose.
305 Mutex thread_ready_aux_mutex = INIT_MUTEX_VAR;
309 /* thread_ready_cond: when signalled, a thread has
310 * become runnable. When used?
312 Condition thread_ready_cond = INIT_COND_VAR;
313 Condition gc_pending_cond = INIT_COND_VAR;
320 rtsTime TimeOfLastYield;
321 rtsBool emitSchedule = rtsTrue;
325 char *whatNext_strs[] = {
333 char *threadReturnCode_strs[] = {
334 "HeapOverflow", /* might also be StackOverflow */
343 StgTSO * createSparkThread(rtsSpark spark);
344 StgTSO * activateSpark (rtsSpark spark);
348 * The thread state for the main thread.
349 // ToDo: check whether not needed any more
353 #if defined(PAR) || defined(RTS_SUPPORTS_THREADS)
354 static void taskStart(void);
358 /* threads start up using 'taskStart', so make them
359 them grab the RTS lock. */
360 #if defined(THREADED_RTS)
361 ACQUIRE_LOCK(&rts_mutex);
370 //@node Main scheduling loop, Suspend and Resume, Prototypes, Main scheduling code
371 //@subsection Main scheduling loop
373 /* ---------------------------------------------------------------------------
374 Main scheduling loop.
376 We use round-robin scheduling, each thread returning to the
377 scheduler loop when one of these conditions is detected:
380 * timer expires (thread yields)
385 Locking notes: we acquire the scheduler lock once at the beginning
386 of the scheduler loop, and release it when
388 * running a thread, or
389 * waiting for work, or
390 * waiting for a GC to complete.
393 In a GranSim setup this loop iterates over the global event queue.
394 This revolves around the global event queue, which determines what
395 to do next. Therefore, it's more complicated than either the
396 concurrent or the parallel (GUM) setup.
399 GUM iterates over incoming messages.
400 It starts with nothing to do (thus CurrentTSO == END_TSO_QUEUE),
401 and sends out a fish whenever it has nothing to do; in-between
402 doing the actual reductions (shared code below) it processes the
403 incoming messages and deals with delayed operations
404 (see PendingFetches).
405 This is not the ugliest code you could imagine, but it's bloody close.
407 ------------------------------------------------------------------------ */
414 StgThreadReturnCode ret;
422 rtsBool receivedFinish = rtsFalse;
424 nat tp_size, sp_size; // stats only
427 rtsBool was_interrupted = rtsFalse;
429 ACQUIRE_LOCK(&sched_mutex);
431 #if defined(THREADED_RTS)
432 /* ToDo: consider SMP support */
433 if (ext_threads_waiting > 0) {
434 /* (At least) one external thread is waiting to
435 * to deposit the result of an external call.
436 * Give way to one of them by giving up the RTS
439 RELEASE_LOCK(&sched_mutex);
440 RELEASE_LOCK(&rts_mutex);
441 /* ToDo: come up with mechanism that guarantees that
442 * the main thread doesn't loop here.
445 /* ToDo: longjmp() */
452 /* set up first event to get things going */
453 /* ToDo: assign costs for system setup and init MainTSO ! */
454 new_event(CurrentProc, CurrentProc, CurrentTime[CurrentProc],
456 CurrentTSO, (StgClosure*)NULL, (rtsSpark*)NULL);
459 fprintf(stderr, "GRAN: Init CurrentTSO (in schedule) = %p\n", CurrentTSO);
460 G_TSO(CurrentTSO, 5));
462 if (RtsFlags.GranFlags.Light) {
463 /* Save current time; GranSim Light only */
464 CurrentTSO->gran.clock = CurrentTime[CurrentProc];
467 event = get_next_event();
469 while (event!=(rtsEvent*)NULL) {
470 /* Choose the processor with the next event */
471 CurrentProc = event->proc;
472 CurrentTSO = event->tso;
476 while (!receivedFinish) { /* set by processMessages */
477 /* when receiving PP_FINISH message */
484 IF_DEBUG(scheduler, printAllThreads());
486 /* If we're interrupted (the user pressed ^C, or some other
487 * termination condition occurred), kill all the currently running
491 IF_DEBUG(scheduler, sched_belch("interrupted"));
493 interrupted = rtsFalse;
494 was_interrupted = rtsTrue;
497 /* Go through the list of main threads and wake up any
498 * clients whose computations have finished. ToDo: this
499 * should be done more efficiently without a linear scan
500 * of the main threads list, somehow...
502 #if defined(RTS_SUPPORTS_THREADS)
504 StgMainThread *m, **prev;
505 prev = &main_threads;
506 for (m = main_threads; m != NULL; m = m->link) {
507 switch (m->tso->what_next) {
510 *(m->ret) = (StgClosure *)m->tso->sp[0];
514 broadcastCondition(&m->wakeup);
517 if (m->ret) *(m->ret) = NULL;
519 if (was_interrupted) {
520 m->stat = Interrupted;
524 broadcastCondition(&m->wakeup);
532 #else /* not threaded */
535 /* in GUM do this only on the Main PE */
538 /* If our main thread has finished or been killed, return.
541 StgMainThread *m = main_threads;
542 if (m->tso->what_next == ThreadComplete
543 || m->tso->what_next == ThreadKilled) {
544 main_threads = main_threads->link;
545 if (m->tso->what_next == ThreadComplete) {
546 /* we finished successfully, fill in the return value */
547 if (m->ret) { *(m->ret) = (StgClosure *)m->tso->sp[0]; };
551 if (m->ret) { *(m->ret) = NULL; };
552 if (was_interrupted) {
553 m->stat = Interrupted;
563 /* Top up the run queue from our spark pool. We try to make the
564 * number of threads in the run queue equal to the number of
567 * Disable spark support in SMP for now, non-essential & requires
568 * a little bit of work to make it compile cleanly. -- sof 1/02.
570 #if 0 /* defined(SMP) */
572 nat n = getFreeCapabilities();
573 StgTSO *tso = run_queue_hd;
575 /* Count the run queue */
576 while (n > 0 && tso != END_TSO_QUEUE) {
583 spark = findSpark(rtsFalse);
585 break; /* no more sparks in the pool */
587 /* I'd prefer this to be done in activateSpark -- HWL */
588 /* tricky - it needs to hold the scheduler lock and
589 * not try to re-acquire it -- SDM */
590 createSparkThread(spark);
592 sched_belch("==^^ turning spark of closure %p into a thread",
593 (StgClosure *)spark));
596 /* We need to wake up the other tasks if we just created some
599 if (getFreeCapabilities() - n > 1) {
600 signalCondition( &thread_ready_cond );
605 /* check for signals each time around the scheduler */
606 #ifndef mingw32_TARGET_OS
607 if (signals_pending()) {
608 startSignalHandlers();
612 /* Check whether any waiting threads need to be woken up. If the
613 * run queue is empty, and there are no other tasks running, we
614 * can wait indefinitely for something to happen.
615 * ToDo: what if another client comes along & requests another
618 if (blocked_queue_hd != END_TSO_QUEUE || sleeping_queue != END_TSO_QUEUE) {
620 (run_queue_hd == END_TSO_QUEUE)
622 && allFreeCapabilities()
626 /* we can be interrupted while waiting for I/O... */
627 if (interrupted) continue;
630 * Detect deadlock: when we have no threads to run, there are no
631 * threads waiting on I/O or sleeping, and all the other tasks are
632 * waiting for work, we must have a deadlock of some description.
634 * We first try to find threads blocked on themselves (ie. black
635 * holes), and generate NonTermination exceptions where necessary.
637 * If no threads are black holed, we have a deadlock situation, so
638 * inform all the main threads.
641 if (blocked_queue_hd == END_TSO_QUEUE
642 && run_queue_hd == END_TSO_QUEUE
643 && sleeping_queue == END_TSO_QUEUE
645 && allFreeCapabilities()
646 #elif defined(THREADED_RTS)
647 && suspended_ccalling_threads == END_TSO_QUEUE
651 IF_DEBUG(scheduler, sched_belch("deadlocked, forcing major GC..."));
652 RELEASE_LOCK(&sched_mutex);
653 GarbageCollect(GetRoots,rtsTrue);
654 ACQUIRE_LOCK(&sched_mutex);
655 IF_DEBUG(scheduler, sched_belch("GC done."));
656 if (blocked_queue_hd == END_TSO_QUEUE
657 && run_queue_hd == END_TSO_QUEUE
658 && sleeping_queue == END_TSO_QUEUE) {
660 IF_DEBUG(scheduler, sched_belch("still deadlocked, checking for black holes..."));
663 /* No black holes, so probably a real deadlock. Send the
664 * current main thread the Deadlock exception (or in the SMP
665 * build, send *all* main threads the deadlock exception,
666 * since none of them can make progress).
668 if (run_queue_hd == END_TSO_QUEUE) {
670 #if defined(RTS_SUPPORTS_THREADS)
671 for (m = main_threads; m != NULL; m = m->link) {
672 switch (m->tso->why_blocked) {
673 case BlockedOnBlackHole:
674 raiseAsync(m->tso, (StgClosure *)NonTermination_closure);
676 case BlockedOnException:
678 raiseAsync(m->tso, (StgClosure *)Deadlock_closure);
681 barf("deadlock: main thread blocked in a strange way");
686 switch (m->tso->why_blocked) {
687 case BlockedOnBlackHole:
688 raiseAsync(m->tso, (StgClosure *)NonTermination_closure);
690 case BlockedOnException:
692 raiseAsync(m->tso, (StgClosure *)Deadlock_closure);
695 barf("deadlock: main thread blocked in a strange way");
699 #if defined(RTS_SUPPORTS_THREADS)
700 if ( run_queue_hd == END_TSO_QUEUE ) {
701 IF_DEBUG(scheduler, sched_belch("all done, it seems...shut down."));
702 shutdownHaskellAndExit(0);
706 ASSERT( run_queue_hd != END_TSO_QUEUE );
710 /* ToDo: add deadlock detection in GUM (similar to SMP) -- HWL */
714 /* If there's a GC pending, don't do anything until it has
718 IF_DEBUG(scheduler,sched_belch("waiting for GC"));
719 waitCondition( &gc_pending_cond, &sched_mutex );
724 /* block until we've got a thread on the run queue and a free
727 while ( run_queue_hd == END_TSO_QUEUE
728 || noFreeCapabilities()
730 IF_DEBUG(scheduler, sched_belch("waiting for work"));
731 waitCondition( &thread_ready_cond, &sched_mutex );
732 IF_DEBUG(scheduler, sched_belch("work now available"));
734 #elif defined(THREADED_RTS)
735 if ( run_queue_hd == END_TSO_QUEUE ) {
736 /* no work available, wait for external calls to complete. */
737 IF_DEBUG(scheduler, sched_belch("worker thread (%d): waiting for external thread to complete..", osThreadId()));
738 RELEASE_LOCK(&sched_mutex);
739 RELEASE_LOCK(&rts_mutex);
740 /* Sigh - need to have a mutex locked in order to wait on the
741 condition variable. */
742 ACQUIRE_LOCK(&thread_ready_aux_mutex);
743 waitCondition(&thread_ready_cond, &thread_ready_aux_mutex);
744 RELEASE_LOCK(&thread_ready_aux_mutex);
745 IF_DEBUG(scheduler, sched_belch("worker thread (%d): re-awakened from no-work slumber..\n", osThreadId()));
746 /* ToDo: longjmp() */
754 if (RtsFlags.GranFlags.Light)
755 GranSimLight_enter_system(event, &ActiveTSO); // adjust ActiveTSO etc
757 /* adjust time based on time-stamp */
758 if (event->time > CurrentTime[CurrentProc] &&
759 event->evttype != ContinueThread)
760 CurrentTime[CurrentProc] = event->time;
762 /* Deal with the idle PEs (may issue FindWork or MoveSpark events) */
763 if (!RtsFlags.GranFlags.Light)
766 IF_DEBUG(gran, fprintf(stderr, "GRAN: switch by event-type\n"));
768 /* main event dispatcher in GranSim */
769 switch (event->evttype) {
770 /* Should just be continuing execution */
772 IF_DEBUG(gran, fprintf(stderr, "GRAN: doing ContinueThread\n"));
773 /* ToDo: check assertion
774 ASSERT(run_queue_hd != (StgTSO*)NULL &&
775 run_queue_hd != END_TSO_QUEUE);
777 /* Ignore ContinueThreads for fetching threads (if synchr comm) */
778 if (!RtsFlags.GranFlags.DoAsyncFetch &&
779 procStatus[CurrentProc]==Fetching) {
780 belch("ghuH: Spurious ContinueThread while Fetching ignored; TSO %d (%p) [PE %d]",
781 CurrentTSO->id, CurrentTSO, CurrentProc);
784 /* Ignore ContinueThreads for completed threads */
785 if (CurrentTSO->what_next == ThreadComplete) {
786 belch("ghuH: found a ContinueThread event for completed thread %d (%p) [PE %d] (ignoring ContinueThread)",
787 CurrentTSO->id, CurrentTSO, CurrentProc);
790 /* Ignore ContinueThreads for threads that are being migrated */
791 if (PROCS(CurrentTSO)==Nowhere) {
792 belch("ghuH: trying to run the migrating TSO %d (%p) [PE %d] (ignoring ContinueThread)",
793 CurrentTSO->id, CurrentTSO, CurrentProc);
796 /* The thread should be at the beginning of the run queue */
797 if (CurrentTSO!=run_queue_hds[CurrentProc]) {
798 belch("ghuH: TSO %d (%p) [PE %d] is not at the start of the run_queue when doing a ContinueThread",
799 CurrentTSO->id, CurrentTSO, CurrentProc);
800 break; // run the thread anyway
803 new_event(proc, proc, CurrentTime[proc],
805 (StgTSO*)NULL, (StgClosure*)NULL, (rtsSpark*)NULL);
807 */ /* Catches superfluous CONTINUEs -- should be unnecessary */
808 break; // now actually run the thread; DaH Qu'vam yImuHbej
811 do_the_fetchnode(event);
812 goto next_thread; /* handle next event in event queue */
815 do_the_globalblock(event);
816 goto next_thread; /* handle next event in event queue */
819 do_the_fetchreply(event);
820 goto next_thread; /* handle next event in event queue */
822 case UnblockThread: /* Move from the blocked queue to the tail of */
823 do_the_unblock(event);
824 goto next_thread; /* handle next event in event queue */
826 case ResumeThread: /* Move from the blocked queue to the tail of */
827 /* the runnable queue ( i.e. Qu' SImqa'lu') */
828 event->tso->gran.blocktime +=
829 CurrentTime[CurrentProc] - event->tso->gran.blockedat;
830 do_the_startthread(event);
831 goto next_thread; /* handle next event in event queue */
834 do_the_startthread(event);
835 goto next_thread; /* handle next event in event queue */
838 do_the_movethread(event);
839 goto next_thread; /* handle next event in event queue */
842 do_the_movespark(event);
843 goto next_thread; /* handle next event in event queue */
846 do_the_findwork(event);
847 goto next_thread; /* handle next event in event queue */
850 barf("Illegal event type %u\n", event->evttype);
853 /* This point was scheduler_loop in the old RTS */
855 IF_DEBUG(gran, belch("GRAN: after main switch"));
857 TimeOfLastEvent = CurrentTime[CurrentProc];
858 TimeOfNextEvent = get_time_of_next_event();
859 IgnoreEvents=(TimeOfNextEvent==0); // HWL HACK
860 // CurrentTSO = ThreadQueueHd;
862 IF_DEBUG(gran, belch("GRAN: time of next event is: %ld",
865 if (RtsFlags.GranFlags.Light)
866 GranSimLight_leave_system(event, &ActiveTSO);
868 EndOfTimeSlice = CurrentTime[CurrentProc]+RtsFlags.GranFlags.time_slice;
871 belch("GRAN: end of time-slice is %#lx", EndOfTimeSlice));
873 /* in a GranSim setup the TSO stays on the run queue */
875 /* Take a thread from the run queue. */
876 t = POP_RUN_QUEUE(); // take_off_run_queue(t);
879 fprintf(stderr, "GRAN: About to run current thread, which is\n");
882 context_switch = 0; // turned on via GranYield, checking events and time slice
885 DumpGranEvent(GR_SCHEDULE, t));
887 procStatus[CurrentProc] = Busy;
890 if (PendingFetches != END_BF_QUEUE) {
894 /* ToDo: phps merge with spark activation above */
895 /* check whether we have local work and send requests if we have none */
896 if (EMPTY_RUN_QUEUE()) { /* no runnable threads */
897 /* :-[ no local threads => look out for local sparks */
898 /* the spark pool for the current PE */
899 pool = &(MainRegTable.rSparks); // generalise to cap = &MainRegTable
900 if (advisory_thread_count < RtsFlags.ParFlags.maxThreads &&
901 pool->hd < pool->tl) {
903 * ToDo: add GC code check that we really have enough heap afterwards!!
905 * If we're here (no runnable threads) and we have pending
906 * sparks, we must have a space problem. Get enough space
907 * to turn one of those pending sparks into a
911 spark = findSpark(rtsFalse); /* get a spark */
912 if (spark != (rtsSpark) NULL) {
913 tso = activateSpark(spark); /* turn the spark into a thread */
914 IF_PAR_DEBUG(schedule,
915 belch("==== schedule: Created TSO %d (%p); %d threads active",
916 tso->id, tso, advisory_thread_count));
918 if (tso==END_TSO_QUEUE) { /* failed to activate spark->back to loop */
919 belch("==^^ failed to activate spark");
921 } /* otherwise fall through & pick-up new tso */
923 IF_PAR_DEBUG(verbose,
924 belch("==^^ no local sparks (spark pool contains only NFs: %d)",
925 spark_queue_len(pool)));
930 /* If we still have no work we need to send a FISH to get a spark
933 if (EMPTY_RUN_QUEUE()) {
934 /* =8-[ no local sparks => look for work on other PEs */
936 * We really have absolutely no work. Send out a fish
937 * (there may be some out there already), and wait for
938 * something to arrive. We clearly can't run any threads
939 * until a SCHEDULE or RESUME arrives, and so that's what
940 * we're hoping to see. (Of course, we still have to
941 * respond to other types of messages.)
943 TIME now = msTime() /*CURRENT_TIME*/;
944 IF_PAR_DEBUG(verbose,
945 belch("-- now=%ld", now));
946 IF_PAR_DEBUG(verbose,
947 if (outstandingFishes < RtsFlags.ParFlags.maxFishes &&
948 (last_fish_arrived_at!=0 &&
949 last_fish_arrived_at+RtsFlags.ParFlags.fishDelay > now)) {
950 belch("--$$ delaying FISH until %ld (last fish %ld, delay %ld, now %ld)",
951 last_fish_arrived_at+RtsFlags.ParFlags.fishDelay,
952 last_fish_arrived_at,
953 RtsFlags.ParFlags.fishDelay, now);
956 if (outstandingFishes < RtsFlags.ParFlags.maxFishes &&
957 (last_fish_arrived_at==0 ||
958 (last_fish_arrived_at+RtsFlags.ParFlags.fishDelay <= now))) {
959 /* outstandingFishes is set in sendFish, processFish;
960 avoid flooding system with fishes via delay */
962 sendFish(pe, mytid, NEW_FISH_AGE, NEW_FISH_HISTORY,
965 // Global statistics: count no. of fishes
966 if (RtsFlags.ParFlags.ParStats.Global &&
967 RtsFlags.GcFlags.giveStats > NO_GC_STATS) {
968 globalParStats.tot_fish_mess++;
972 receivedFinish = processMessages();
975 } else if (PacketsWaiting()) { /* Look for incoming messages */
976 receivedFinish = processMessages();
979 /* Now we are sure that we have some work available */
980 ASSERT(run_queue_hd != END_TSO_QUEUE);
982 /* Take a thread from the run queue, if we have work */
983 t = POP_RUN_QUEUE(); // take_off_run_queue(END_TSO_QUEUE);
984 IF_DEBUG(sanity,checkTSO(t));
986 /* ToDo: write something to the log-file
987 if (RTSflags.ParFlags.granSimStats && !sameThread)
988 DumpGranEvent(GR_SCHEDULE, RunnableThreadsHd);
992 /* the spark pool for the current PE */
993 pool = &(MainRegTable.rSparks); // generalise to cap = &MainRegTable
996 belch("--=^ %d threads, %d sparks on [%#x]",
997 run_queue_len(), spark_queue_len(pool), CURRENT_PROC));
1000 if (0 && RtsFlags.ParFlags.ParStats.Full &&
1001 t && LastTSO && t->id != LastTSO->id &&
1002 LastTSO->why_blocked == NotBlocked &&
1003 LastTSO->what_next != ThreadComplete) {
1004 // if previously scheduled TSO not blocked we have to record the context switch
1005 DumpVeryRawGranEvent(TimeOfLastYield, CURRENT_PROC, CURRENT_PROC,
1006 GR_DESCHEDULE, LastTSO, (StgClosure *)NULL, 0, 0);
1009 if (RtsFlags.ParFlags.ParStats.Full &&
1010 (emitSchedule /* forced emit */ ||
1011 (t && LastTSO && t->id != LastTSO->id))) {
1013 we are running a different TSO, so write a schedule event to log file
1014 NB: If we use fair scheduling we also have to write a deschedule
1015 event for LastTSO; with unfair scheduling we know that the
1016 previous tso has blocked whenever we switch to another tso, so
1017 we don't need it in GUM for now
1019 DumpRawGranEvent(CURRENT_PROC, CURRENT_PROC,
1020 GR_SCHEDULE, t, (StgClosure *)NULL, 0, 0);
1021 emitSchedule = rtsFalse;
1025 #else /* !GRAN && !PAR */
1027 /* grab a thread from the run queue
1029 ASSERT(run_queue_hd != END_TSO_QUEUE);
1030 t = POP_RUN_QUEUE();
1031 // Sanity check the thread we're about to run. This can be
1032 // expensive if there is lots of thread switching going on...
1033 IF_DEBUG(sanity,checkTSO(t));
1036 grabCapability(&cap);
1037 cap->r.rCurrentTSO = t;
1039 /* context switches are now initiated by the timer signal, unless
1040 * the user specified "context switch as often as possible", with
1045 RtsFlags.ProfFlags.profileInterval == 0 ||
1047 (RtsFlags.ConcFlags.ctxtSwitchTicks == 0
1048 && (run_queue_hd != END_TSO_QUEUE
1049 || blocked_queue_hd != END_TSO_QUEUE
1050 || sleeping_queue != END_TSO_QUEUE)))
1055 RELEASE_LOCK(&sched_mutex);
1057 IF_DEBUG(scheduler, sched_belch("-->> Running TSO %ld (%p) %s ...",
1058 t->id, t, whatNext_strs[t->what_next]));
1061 startHeapProfTimer();
1064 /* +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ */
1065 /* Run the current thread
1067 switch (cap->r.rCurrentTSO->what_next) {
1069 case ThreadComplete:
1070 /* Thread already finished, return to scheduler. */
1071 ret = ThreadFinished;
1073 case ThreadEnterGHC:
1074 ret = StgRun((StgFunPtr) stg_enterStackTop, &cap->r);
1077 ret = StgRun((StgFunPtr) stg_returnToStackTop, &cap->r);
1079 case ThreadEnterInterp:
1080 ret = interpretBCO(cap);
1083 barf("schedule: invalid what_next field");
1085 /* +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ */
1087 /* Costs for the scheduler are assigned to CCS_SYSTEM */
1089 stopHeapProfTimer();
1093 ACQUIRE_LOCK(&sched_mutex);
1096 IF_DEBUG(scheduler,fprintf(stderr,"scheduler (task %ld): ", osThreadId()););
1097 #elif !defined(GRAN) && !defined(PAR)
1098 IF_DEBUG(scheduler,fprintf(stderr,"scheduler: "););
1100 t = cap->r.rCurrentTSO;
1103 /* HACK 675: if the last thread didn't yield, make sure to print a
1104 SCHEDULE event to the log file when StgRunning the next thread, even
1105 if it is the same one as before */
1107 TimeOfLastYield = CURRENT_TIME;
1113 IF_DEBUG(gran, DumpGranEvent(GR_DESCHEDULE, t));
1114 globalGranStats.tot_heapover++;
1116 globalParStats.tot_heapover++;
1119 // did the task ask for a large block?
1120 if (cap->r.rHpAlloc > BLOCK_SIZE_W) {
1121 // if so, get one and push it on the front of the nursery.
1125 blocks = (nat)BLOCK_ROUND_UP(cap->r.rHpAlloc * sizeof(W_)) / BLOCK_SIZE;
1127 IF_DEBUG(scheduler,belch("--<< thread %ld (%p; %s) stopped: requesting a large block (size %d)",
1129 whatNext_strs[t->what_next], blocks));
1131 // don't do this if it would push us over the
1132 // alloc_blocks_lim limit; we'll GC first.
1133 if (alloc_blocks + blocks < alloc_blocks_lim) {
1135 alloc_blocks += blocks;
1136 bd = allocGroup( blocks );
1138 // link the new group into the list
1139 bd->link = cap->r.rCurrentNursery;
1140 bd->u.back = cap->r.rCurrentNursery->u.back;
1141 if (cap->r.rCurrentNursery->u.back != NULL) {
1142 cap->r.rCurrentNursery->u.back->link = bd;
1144 ASSERT(g0s0->blocks == cap->r.rCurrentNursery &&
1145 g0s0->blocks == cap->r.rNursery);
1146 cap->r.rNursery = g0s0->blocks = bd;
1148 cap->r.rCurrentNursery->u.back = bd;
1150 // initialise it as a nursery block
1154 bd->free = bd->start;
1156 // don't forget to update the block count in g0s0.
1157 g0s0->n_blocks += blocks;
1158 ASSERT(countBlocks(g0s0->blocks) == g0s0->n_blocks);
1160 // now update the nursery to point to the new block
1161 cap->r.rCurrentNursery = bd;
1163 // we might be unlucky and have another thread get on the
1164 // run queue before us and steal the large block, but in that
1165 // case the thread will just end up requesting another large
1167 PUSH_ON_RUN_QUEUE(t);
1172 /* make all the running tasks block on a condition variable,
1173 * maybe set context_switch and wait till they all pile in,
1174 * then have them wait on a GC condition variable.
1176 IF_DEBUG(scheduler,belch("--<< thread %ld (%p; %s) stopped: HeapOverflow",
1177 t->id, t, whatNext_strs[t->what_next]));
1180 ASSERT(!is_on_queue(t,CurrentProc));
1182 /* Currently we emit a DESCHEDULE event before GC in GUM.
1183 ToDo: either add separate event to distinguish SYSTEM time from rest
1184 or just nuke this DESCHEDULE (and the following SCHEDULE) */
1185 if (0 && RtsFlags.ParFlags.ParStats.Full) {
1186 DumpRawGranEvent(CURRENT_PROC, CURRENT_PROC,
1187 GR_DESCHEDULE, t, (StgClosure *)NULL, 0, 0);
1188 emitSchedule = rtsTrue;
1192 ready_to_gc = rtsTrue;
1193 context_switch = 1; /* stop other threads ASAP */
1194 PUSH_ON_RUN_QUEUE(t);
1195 /* actual GC is done at the end of the while loop */
1201 DumpGranEvent(GR_DESCHEDULE, t));
1202 globalGranStats.tot_stackover++;
1205 // DumpGranEvent(GR_DESCHEDULE, t);
1206 globalParStats.tot_stackover++;
1208 IF_DEBUG(scheduler,belch("--<< thread %ld (%p; %s) stopped, StackOverflow",
1209 t->id, t, whatNext_strs[t->what_next]));
1210 /* just adjust the stack for this thread, then pop it back
1216 /* enlarge the stack */
1217 StgTSO *new_t = threadStackOverflow(t);
1219 /* This TSO has moved, so update any pointers to it from the
1220 * main thread stack. It better not be on any other queues...
1221 * (it shouldn't be).
1223 for (m = main_threads; m != NULL; m = m->link) {
1228 threadPaused(new_t);
1229 PUSH_ON_RUN_QUEUE(new_t);
1233 case ThreadYielding:
1236 DumpGranEvent(GR_DESCHEDULE, t));
1237 globalGranStats.tot_yields++;
1240 // DumpGranEvent(GR_DESCHEDULE, t);
1241 globalParStats.tot_yields++;
1243 /* put the thread back on the run queue. Then, if we're ready to
1244 * GC, check whether this is the last task to stop. If so, wake
1245 * up the GC thread. getThread will block during a GC until the
1249 if (t->what_next == ThreadEnterInterp) {
1250 /* ToDo: or maybe a timer expired when we were in Hugs?
1251 * or maybe someone hit ctrl-C
1253 belch("--<< thread %ld (%p; %s) stopped to switch to Hugs",
1254 t->id, t, whatNext_strs[t->what_next]);
1256 belch("--<< thread %ld (%p; %s) stopped, yielding",
1257 t->id, t, whatNext_strs[t->what_next]);
1264 //belch("&& Doing sanity check on yielding TSO %ld.", t->id);
1266 ASSERT(t->link == END_TSO_QUEUE);
1268 ASSERT(!is_on_queue(t,CurrentProc));
1271 //belch("&& Doing sanity check on all ThreadQueues (and their TSOs).");
1272 checkThreadQsSanity(rtsTrue));
1275 if (RtsFlags.ParFlags.doFairScheduling) {
1276 /* this does round-robin scheduling; good for concurrency */
1277 APPEND_TO_RUN_QUEUE(t);
1279 /* this does unfair scheduling; good for parallelism */
1280 PUSH_ON_RUN_QUEUE(t);
1283 /* this does round-robin scheduling; good for concurrency */
1284 APPEND_TO_RUN_QUEUE(t);
1287 /* add a ContinueThread event to actually process the thread */
1288 new_event(CurrentProc, CurrentProc, CurrentTime[CurrentProc],
1290 t, (StgClosure*)NULL, (rtsSpark*)NULL);
1292 belch("GRAN: eventq and runnableq after adding yielded thread to queue again:");
1301 belch("--<< thread %ld (%p; %s) stopped, blocking on node %p [PE %d] with BQ: ",
1302 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)));
1303 if (t->block_info.closure!=(StgClosure*)NULL) print_bq(t->block_info.closure));
1305 // ??? needed; should emit block before
1307 DumpGranEvent(GR_DESCHEDULE, t));
1308 prune_eventq(t, (StgClosure *)NULL); // prune ContinueThreads for t
1311 ASSERT(procStatus[CurrentProc]==Busy ||
1312 ((procStatus[CurrentProc]==Fetching) &&
1313 (t->block_info.closure!=(StgClosure*)NULL)));
1314 if (run_queue_hds[CurrentProc] == END_TSO_QUEUE &&
1315 !(!RtsFlags.GranFlags.DoAsyncFetch &&
1316 procStatus[CurrentProc]==Fetching))
1317 procStatus[CurrentProc] = Idle;
1321 belch("--<< thread %ld (%p; %s) stopped, blocking on node %p with BQ: ",
1322 t->id, t, whatNext_strs[t->what_next], t->block_info.closure));
1325 if (t->block_info.closure!=(StgClosure*)NULL)
1326 print_bq(t->block_info.closure));
1328 /* Send a fetch (if BlockedOnGA) and dump event to log file */
1331 /* whatever we schedule next, we must log that schedule */
1332 emitSchedule = rtsTrue;
1335 /* don't need to do anything. Either the thread is blocked on
1336 * I/O, in which case we'll have called addToBlockedQueue
1337 * previously, or it's blocked on an MVar or Blackhole, in which
1338 * case it'll be on the relevant queue already.
1341 fprintf(stderr, "--<< thread %d (%p) stopped: ", t->id, t);
1342 printThreadBlockage(t);
1343 fprintf(stderr, "\n"));
1345 /* Only for dumping event to log file
1346 ToDo: do I need this in GranSim, too?
1353 case ThreadFinished:
1354 /* Need to check whether this was a main thread, and if so, signal
1355 * the task that started it with the return value. If we have no
1356 * more main threads, we probably need to stop all the tasks until
1359 /* We also end up here if the thread kills itself with an
1360 * uncaught exception, see Exception.hc.
1362 IF_DEBUG(scheduler,belch("--++ thread %d (%p) finished", t->id, t));
1364 endThread(t, CurrentProc); // clean-up the thread
1366 /* For now all are advisory -- HWL */
1367 //if(t->priority==AdvisoryPriority) ??
1368 advisory_thread_count--;
1371 if(t->dist.priority==RevalPriority)
1375 if (RtsFlags.ParFlags.ParStats.Full &&
1376 !RtsFlags.ParFlags.ParStats.Suppressed)
1377 DumpEndEvent(CURRENT_PROC, t, rtsFalse /* not mandatory */);
1382 barf("schedule: invalid thread return code %d", (int)ret);
1386 grabCapability(&cap);
1390 if (RtsFlags.ProfFlags.profileInterval==0 || performHeapProfile) {
1391 GarbageCollect(GetRoots, rtsTrue);
1393 performHeapProfile = rtsFalse;
1394 ready_to_gc = rtsFalse; // we already GC'd
1399 if (ready_to_gc && allFreeCapabilities() )
1404 /* everybody back, start the GC.
1405 * Could do it in this thread, or signal a condition var
1406 * to do it in another thread. Either way, we need to
1407 * broadcast on gc_pending_cond afterward.
1409 #if defined(RTS_SUPPORTS_THREADS)
1410 IF_DEBUG(scheduler,sched_belch("doing GC"));
1412 GarbageCollect(GetRoots,rtsFalse);
1413 ready_to_gc = rtsFalse;
1415 broadcastCondition(&gc_pending_cond);
1418 /* add a ContinueThread event to continue execution of current thread */
1419 new_event(CurrentProc, CurrentProc, CurrentTime[CurrentProc],
1421 t, (StgClosure*)NULL, (rtsSpark*)NULL);
1423 fprintf(stderr, "GRAN: eventq and runnableq after Garbage collection:\n");
1431 IF_GRAN_DEBUG(unused,
1432 print_eventq(EventHd));
1434 event = get_next_event();
1437 /* ToDo: wait for next message to arrive rather than busy wait */
1440 } /* end of while(1) */
1442 IF_PAR_DEBUG(verbose,
1443 belch("== Leaving schedule() after having received Finish"));
1446 /* ---------------------------------------------------------------------------
1447 * deleteAllThreads(): kill all the live threads.
1449 * This is used when we catch a user interrupt (^C), before performing
1450 * any necessary cleanups and running finalizers.
1451 * ------------------------------------------------------------------------- */
1453 void deleteAllThreads ( void )
1456 IF_DEBUG(scheduler,sched_belch("deleting all threads"));
1457 for (t = run_queue_hd; t != END_TSO_QUEUE; t = next) {
1461 for (t = blocked_queue_hd; t != END_TSO_QUEUE; t = next) {
1465 for (t = sleeping_queue; t != END_TSO_QUEUE; t = next) {
1469 run_queue_hd = run_queue_tl = END_TSO_QUEUE;
1470 blocked_queue_hd = blocked_queue_tl = END_TSO_QUEUE;
1471 sleeping_queue = END_TSO_QUEUE;
1474 /* startThread and insertThread are now in GranSim.c -- HWL */
1477 //@node Suspend and Resume, Run queue code, Main scheduling loop, Main scheduling code
1478 //@subsection Suspend and Resume
1480 /* ---------------------------------------------------------------------------
1481 * Suspending & resuming Haskell threads.
1483 * When making a "safe" call to C (aka _ccall_GC), the task gives back
1484 * its capability before calling the C function. This allows another
1485 * task to pick up the capability and carry on running Haskell
1486 * threads. It also means that if the C call blocks, it won't lock
1489 * The Haskell thread making the C call is put to sleep for the
1490 * duration of the call, on the susepended_ccalling_threads queue. We
1491 * give out a token to the task, which it can use to resume the thread
1492 * on return from the C function.
1493 * ------------------------------------------------------------------------- */
1496 suspendThread( StgRegTable *reg )
1501 /* assume that *reg is a pointer to the StgRegTable part
1504 cap = (Capability *)((void *)reg - sizeof(StgFunTable));
1506 ACQUIRE_LOCK(&sched_mutex);
1509 sched_belch("thread %d did a _ccall_gc", cap->r.rCurrentTSO->id));
1511 threadPaused(cap->r.rCurrentTSO);
1512 cap->r.rCurrentTSO->link = suspended_ccalling_threads;
1513 suspended_ccalling_threads = cap->r.rCurrentTSO;
1515 /* Use the thread ID as the token; it should be unique */
1516 tok = cap->r.rCurrentTSO->id;
1518 /* Hand back capability */
1519 releaseCapability(&cap);
1521 #if defined(RTS_SUPPORTS_THREADS) && !defined(SMP)
1522 IF_DEBUG(scheduler, sched_belch("thread %d leaving RTS\n", tok));
1523 startTask(taskStart);
1526 RELEASE_LOCK(&sched_mutex);
1527 RELEASE_LOCK(&rts_mutex);
1532 resumeThread( StgInt tok )
1534 StgTSO *tso, **prev;
1537 #if defined(THREADED_RTS)
1538 IF_DEBUG(scheduler, sched_belch("thread %d returning, waiting for sched. lock.\n", tok));
1539 ACQUIRE_LOCK(&sched_mutex);
1540 ext_threads_waiting++;
1541 IF_DEBUG(scheduler, sched_belch("thread %d returning, ext_thread count: %d.\n", tok, ext_threads_waiting));
1542 RELEASE_LOCK(&sched_mutex);
1544 IF_DEBUG(scheduler, sched_belch("thread %d waiting for RTS lock...\n", tok));
1545 ACQUIRE_LOCK(&rts_mutex);
1546 ext_threads_waiting--;
1547 IF_DEBUG(scheduler, sched_belch("thread %d acquired RTS lock...\n", tok));
1550 #if defined(THREADED_RTS)
1551 /* Free up any RTS-blocked threads. */
1552 broadcastCondition(&thread_ready_cond);
1555 /* Remove the thread off of the suspended list */
1556 prev = &suspended_ccalling_threads;
1557 for (tso = suspended_ccalling_threads;
1558 tso != END_TSO_QUEUE;
1559 prev = &tso->link, tso = tso->link) {
1560 if (tso->id == (StgThreadID)tok) {
1565 if (tso == END_TSO_QUEUE) {
1566 barf("resumeThread: thread not found");
1568 tso->link = END_TSO_QUEUE;
1571 while ( noFreeCapabilities() ) {
1572 IF_DEBUG(scheduler, sched_belch("waiting to resume"));
1573 waitCondition(&thread_ready_cond, &sched_mutex);
1574 IF_DEBUG(scheduler, sched_belch("resuming thread %d", tso->id));
1578 grabCapability(&cap);
1580 cap->r.rCurrentTSO = tso;
1586 /* ---------------------------------------------------------------------------
1588 * ------------------------------------------------------------------------ */
1589 static void unblockThread(StgTSO *tso);
1591 /* ---------------------------------------------------------------------------
1592 * Comparing Thread ids.
1594 * This is used from STG land in the implementation of the
1595 * instances of Eq/Ord for ThreadIds.
1596 * ------------------------------------------------------------------------ */
1598 int cmp_thread(const StgTSO *tso1, const StgTSO *tso2)
1600 StgThreadID id1 = tso1->id;
1601 StgThreadID id2 = tso2->id;
1603 if (id1 < id2) return (-1);
1604 if (id1 > id2) return 1;
1608 /* ---------------------------------------------------------------------------
1609 * Fetching the ThreadID from an StgTSO.
1611 * This is used in the implementation of Show for ThreadIds.
1612 * ------------------------------------------------------------------------ */
1613 int rts_getThreadId(const StgTSO *tso)
1618 /* ---------------------------------------------------------------------------
1619 Create a new thread.
1621 The new thread starts with the given stack size. Before the
1622 scheduler can run, however, this thread needs to have a closure
1623 (and possibly some arguments) pushed on its stack. See
1624 pushClosure() in Schedule.h.
1626 createGenThread() and createIOThread() (in SchedAPI.h) are
1627 convenient packaged versions of this function.
1629 currently pri (priority) is only used in a GRAN setup -- HWL
1630 ------------------------------------------------------------------------ */
1631 //@cindex createThread
1633 /* currently pri (priority) is only used in a GRAN setup -- HWL */
1635 createThread(nat stack_size, StgInt pri)
1637 return createThread_(stack_size, rtsFalse, pri);
1641 createThread_(nat size, rtsBool have_lock, StgInt pri)
1645 createThread(nat stack_size)
1647 return createThread_(stack_size, rtsFalse);
1651 createThread_(nat size, rtsBool have_lock)
1658 /* First check whether we should create a thread at all */
1660 /* check that no more than RtsFlags.ParFlags.maxThreads threads are created */
1661 if (advisory_thread_count >= RtsFlags.ParFlags.maxThreads) {
1663 belch("{createThread}Daq ghuH: refusing to create another thread; no more than %d threads allowed (currently %d)",
1664 RtsFlags.ParFlags.maxThreads, advisory_thread_count);
1665 return END_TSO_QUEUE;
1671 ASSERT(!RtsFlags.GranFlags.Light || CurrentProc==0);
1674 // ToDo: check whether size = stack_size - TSO_STRUCT_SIZEW
1676 /* catch ridiculously small stack sizes */
1677 if (size < MIN_STACK_WORDS + TSO_STRUCT_SIZEW) {
1678 size = MIN_STACK_WORDS + TSO_STRUCT_SIZEW;
1681 stack_size = size - TSO_STRUCT_SIZEW;
1683 tso = (StgTSO *)allocate(size);
1684 TICK_ALLOC_TSO(stack_size, 0);
1686 SET_HDR(tso, &stg_TSO_info, CCS_SYSTEM);
1688 SET_GRAN_HDR(tso, ThisPE);
1690 tso->what_next = ThreadEnterGHC;
1692 /* tso->id needs to be unique. For now we use a heavyweight mutex to
1693 * protect the increment operation on next_thread_id.
1694 * In future, we could use an atomic increment instead.
1696 if (!have_lock) { ACQUIRE_LOCK(&sched_mutex); }
1697 tso->id = next_thread_id++;
1698 if (!have_lock) { RELEASE_LOCK(&sched_mutex); }
1700 tso->why_blocked = NotBlocked;
1701 tso->blocked_exceptions = NULL;
1703 tso->stack_size = stack_size;
1704 tso->max_stack_size = round_to_mblocks(RtsFlags.GcFlags.maxStkSize)
1706 tso->sp = (P_)&(tso->stack) + stack_size;
1709 tso->prof.CCCS = CCS_MAIN;
1712 /* put a stop frame on the stack */
1713 tso->sp -= sizeofW(StgStopFrame);
1714 SET_HDR((StgClosure*)tso->sp,(StgInfoTable *)&stg_stop_thread_info,CCS_SYSTEM);
1715 tso->su = (StgUpdateFrame*)tso->sp;
1719 tso->link = END_TSO_QUEUE;
1720 /* uses more flexible routine in GranSim */
1721 insertThread(tso, CurrentProc);
1723 /* In a non-GranSim setup the pushing of a TSO onto the runq is separated
1729 if (RtsFlags.GranFlags.GranSimStats.Full)
1730 DumpGranEvent(GR_START,tso);
1732 if (RtsFlags.ParFlags.ParStats.Full)
1733 DumpGranEvent(GR_STARTQ,tso);
1734 /* HACk to avoid SCHEDULE
1738 /* Link the new thread on the global thread list.
1740 tso->global_link = all_threads;
1744 tso->dist.priority = MandatoryPriority; //by default that is...
1748 tso->gran.pri = pri;
1750 tso->gran.magic = TSO_MAGIC; // debugging only
1752 tso->gran.sparkname = 0;
1753 tso->gran.startedat = CURRENT_TIME;
1754 tso->gran.exported = 0;
1755 tso->gran.basicblocks = 0;
1756 tso->gran.allocs = 0;
1757 tso->gran.exectime = 0;
1758 tso->gran.fetchtime = 0;
1759 tso->gran.fetchcount = 0;
1760 tso->gran.blocktime = 0;
1761 tso->gran.blockcount = 0;
1762 tso->gran.blockedat = 0;
1763 tso->gran.globalsparks = 0;
1764 tso->gran.localsparks = 0;
1765 if (RtsFlags.GranFlags.Light)
1766 tso->gran.clock = Now; /* local clock */
1768 tso->gran.clock = 0;
1770 IF_DEBUG(gran,printTSO(tso));
1773 tso->par.magic = TSO_MAGIC; // debugging only
1775 tso->par.sparkname = 0;
1776 tso->par.startedat = CURRENT_TIME;
1777 tso->par.exported = 0;
1778 tso->par.basicblocks = 0;
1779 tso->par.allocs = 0;
1780 tso->par.exectime = 0;
1781 tso->par.fetchtime = 0;
1782 tso->par.fetchcount = 0;
1783 tso->par.blocktime = 0;
1784 tso->par.blockcount = 0;
1785 tso->par.blockedat = 0;
1786 tso->par.globalsparks = 0;
1787 tso->par.localsparks = 0;
1791 globalGranStats.tot_threads_created++;
1792 globalGranStats.threads_created_on_PE[CurrentProc]++;
1793 globalGranStats.tot_sq_len += spark_queue_len(CurrentProc);
1794 globalGranStats.tot_sq_probes++;
1796 // collect parallel global statistics (currently done together with GC stats)
1797 if (RtsFlags.ParFlags.ParStats.Global &&
1798 RtsFlags.GcFlags.giveStats > NO_GC_STATS) {
1799 //fprintf(stderr, "Creating thread %d @ %11.2f\n", tso->id, usertime());
1800 globalParStats.tot_threads_created++;
1806 belch("==__ schedule: Created TSO %d (%p);",
1807 CurrentProc, tso, tso->id));
1809 IF_PAR_DEBUG(verbose,
1810 belch("==__ schedule: Created TSO %d (%p); %d threads active",
1811 tso->id, tso, advisory_thread_count));
1813 IF_DEBUG(scheduler,sched_belch("created thread %ld, stack size = %lx words",
1814 tso->id, tso->stack_size));
1821 all parallel thread creation calls should fall through the following routine.
1824 createSparkThread(rtsSpark spark)
1826 ASSERT(spark != (rtsSpark)NULL);
1827 if (advisory_thread_count >= RtsFlags.ParFlags.maxThreads)
1829 barf("{createSparkThread}Daq ghuH: refusing to create another thread; no more than %d threads allowed (currently %d)",
1830 RtsFlags.ParFlags.maxThreads, advisory_thread_count);
1831 return END_TSO_QUEUE;
1835 tso = createThread_(RtsFlags.GcFlags.initialStkSize, rtsTrue);
1836 if (tso==END_TSO_QUEUE)
1837 barf("createSparkThread: Cannot create TSO");
1839 tso->priority = AdvisoryPriority;
1841 pushClosure(tso,spark);
1842 PUSH_ON_RUN_QUEUE(tso);
1843 advisory_thread_count++;
1850 Turn a spark into a thread.
1851 ToDo: fix for SMP (needs to acquire SCHED_MUTEX!)
1854 //@cindex activateSpark
1856 activateSpark (rtsSpark spark)
1860 tso = createSparkThread(spark);
1861 if (RtsFlags.ParFlags.ParStats.Full) {
1862 //ASSERT(run_queue_hd == END_TSO_QUEUE); // I think ...
1863 IF_PAR_DEBUG(verbose,
1864 belch("==^^ activateSpark: turning spark of closure %p (%s) into a thread",
1865 (StgClosure *)spark, info_type((StgClosure *)spark)));
1867 // ToDo: fwd info on local/global spark to thread -- HWL
1868 // tso->gran.exported = spark->exported;
1869 // tso->gran.locked = !spark->global;
1870 // tso->gran.sparkname = spark->name;
1876 /* ---------------------------------------------------------------------------
1879 * scheduleThread puts a thread on the head of the runnable queue.
1880 * This will usually be done immediately after a thread is created.
1881 * The caller of scheduleThread must create the thread using e.g.
1882 * createThread and push an appropriate closure
1883 * on this thread's stack before the scheduler is invoked.
1884 * ------------------------------------------------------------------------ */
1887 scheduleThread(StgTSO *tso)
1889 ACQUIRE_LOCK(&sched_mutex);
1891 /* Put the new thread on the head of the runnable queue. The caller
1892 * better push an appropriate closure on this thread's stack
1893 * beforehand. In the SMP case, the thread may start running as
1894 * soon as we release the scheduler lock below.
1896 PUSH_ON_RUN_QUEUE(tso);
1900 IF_DEBUG(scheduler,printTSO(tso));
1902 RELEASE_LOCK(&sched_mutex);
1905 /* ---------------------------------------------------------------------------
1908 * Initialise the scheduler. This resets all the queues - if the
1909 * queues contained any threads, they'll be garbage collected at the
1912 * ------------------------------------------------------------------------ */
1916 term_handler(int sig STG_UNUSED)
1919 ACQUIRE_LOCK(&term_mutex);
1921 RELEASE_LOCK(&term_mutex);
1932 for (i=0; i<=MAX_PROC; i++) {
1933 run_queue_hds[i] = END_TSO_QUEUE;
1934 run_queue_tls[i] = END_TSO_QUEUE;
1935 blocked_queue_hds[i] = END_TSO_QUEUE;
1936 blocked_queue_tls[i] = END_TSO_QUEUE;
1937 ccalling_threadss[i] = END_TSO_QUEUE;
1938 sleeping_queue = END_TSO_QUEUE;
1941 run_queue_hd = END_TSO_QUEUE;
1942 run_queue_tl = END_TSO_QUEUE;
1943 blocked_queue_hd = END_TSO_QUEUE;
1944 blocked_queue_tl = END_TSO_QUEUE;
1945 sleeping_queue = END_TSO_QUEUE;
1948 suspended_ccalling_threads = END_TSO_QUEUE;
1950 main_threads = NULL;
1951 all_threads = END_TSO_QUEUE;
1956 RtsFlags.ConcFlags.ctxtSwitchTicks =
1957 RtsFlags.ConcFlags.ctxtSwitchTime / TICK_MILLISECS;
1959 #if defined(RTS_SUPPORTS_THREADS)
1960 /* Initialise the mutex and condition variables used by
1962 initMutex(&rts_mutex);
1963 initMutex(&sched_mutex);
1964 initMutex(&term_mutex);
1965 #if defined(THREADED_RTS)
1966 initMutex(&thread_ready_aux_mutex);
1969 initCondition(&thread_ready_cond);
1970 initCondition(&gc_pending_cond);
1973 #if defined(THREADED_RTS)
1975 ACQUIRE_LOCK(&rts_mutex);
1977 sched_belch("worker thread (%d): acquired RTS lock\n", osThreadId()));
1980 /* Install the SIGHUP handler */
1983 struct sigaction action,oact;
1985 action.sa_handler = term_handler;
1986 sigemptyset(&action.sa_mask);
1987 action.sa_flags = 0;
1988 if (sigaction(SIGTERM, &action, &oact) != 0) {
1989 barf("can't install TERM handler");
1994 /* A capability holds the state a native thread needs in
1995 * order to execute STG code. At least one capability is
1996 * floating around (only SMP builds have more than one).
2000 #if defined(RTS_SUPPORTS_THREADS)
2001 /* start our haskell execution tasks */
2003 startTaskManager(RtsFlags.ParFlags.nNodes, taskStart);
2005 startTaskManager(0,taskStart);
2009 #if /* defined(SMP) ||*/ defined(PAR)
2015 exitScheduler( void )
2017 #if defined(RTS_SUPPORTS_THREADS)
2022 /* -----------------------------------------------------------------------------
2023 Managing the per-task allocation areas.
2025 Each capability comes with an allocation area. These are
2026 fixed-length block lists into which allocation can be done.
2028 ToDo: no support for two-space collection at the moment???
2029 -------------------------------------------------------------------------- */
2031 /* -----------------------------------------------------------------------------
2032 * waitThread is the external interface for running a new computation
2033 * and waiting for the result.
2035 * In the non-SMP case, we create a new main thread, push it on the
2036 * main-thread stack, and invoke the scheduler to run it. The
2037 * scheduler will return when the top main thread on the stack has
2038 * completed or died, and fill in the necessary fields of the
2039 * main_thread structure.
2041 * In the SMP case, we create a main thread as before, but we then
2042 * create a new condition variable and sleep on it. When our new
2043 * main thread has completed, we'll be woken up and the status/result
2044 * will be in the main_thread struct.
2045 * -------------------------------------------------------------------------- */
2048 howManyThreadsAvail ( void )
2052 for (q = run_queue_hd; q != END_TSO_QUEUE; q = q->link)
2054 for (q = blocked_queue_hd; q != END_TSO_QUEUE; q = q->link)
2056 for (q = sleeping_queue; q != END_TSO_QUEUE; q = q->link)
2062 finishAllThreads ( void )
2065 while (run_queue_hd != END_TSO_QUEUE) {
2066 waitThread ( run_queue_hd, NULL );
2068 while (blocked_queue_hd != END_TSO_QUEUE) {
2069 waitThread ( blocked_queue_hd, NULL );
2071 while (sleeping_queue != END_TSO_QUEUE) {
2072 waitThread ( blocked_queue_hd, NULL );
2075 (blocked_queue_hd != END_TSO_QUEUE ||
2076 run_queue_hd != END_TSO_QUEUE ||
2077 sleeping_queue != END_TSO_QUEUE);
2081 waitThread(StgTSO *tso, /*out*/StgClosure **ret)
2084 SchedulerStatus stat;
2086 ACQUIRE_LOCK(&sched_mutex);
2088 m = stgMallocBytes(sizeof(StgMainThread), "waitThread");
2093 #if defined(RTS_SUPPORTS_THREADS)
2094 initCondition(&m->wakeup);
2097 m->link = main_threads;
2100 IF_DEBUG(scheduler, fprintf(stderr, "== scheduler: new main thread (%d)\n",
2105 waitCondition(&m->wakeup, &sched_mutex);
2106 } while (m->stat == NoStatus);
2108 /* GranSim specific init */
2109 CurrentTSO = m->tso; // the TSO to run
2110 procStatus[MainProc] = Busy; // status of main PE
2111 CurrentProc = MainProc; // PE to run it on
2115 RELEASE_LOCK(&sched_mutex);
2117 ASSERT(m->stat != NoStatus);
2122 #if defined(RTS_SUPPORTS_THREADS)
2123 closeCondition(&m->wakeup);
2126 IF_DEBUG(scheduler, fprintf(stderr, "== scheduler: main thread (%d) finished\n",
2130 RELEASE_LOCK(&sched_mutex);
2135 //@node Run queue code, Garbage Collextion Routines, Suspend and Resume, Main scheduling code
2136 //@subsection Run queue code
2140 NB: In GranSim we have many run queues; run_queue_hd is actually a macro
2141 unfolding to run_queue_hds[CurrentProc], thus CurrentProc is an
2142 implicit global variable that has to be correct when calling these
2146 /* Put the new thread on the head of the runnable queue.
2147 * The caller of createThread better push an appropriate closure
2148 * on this thread's stack before the scheduler is invoked.
2150 static /* inline */ void
2151 add_to_run_queue(tso)
2154 ASSERT(tso!=run_queue_hd && tso!=run_queue_tl);
2155 tso->link = run_queue_hd;
2157 if (run_queue_tl == END_TSO_QUEUE) {
2162 /* Put the new thread at the end of the runnable queue. */
2163 static /* inline */ void
2164 push_on_run_queue(tso)
2167 ASSERT(get_itbl((StgClosure *)tso)->type == TSO);
2168 ASSERT(run_queue_hd!=NULL && run_queue_tl!=NULL);
2169 ASSERT(tso!=run_queue_hd && tso!=run_queue_tl);
2170 if (run_queue_hd == END_TSO_QUEUE) {
2173 run_queue_tl->link = tso;
2179 Should be inlined because it's used very often in schedule. The tso
2180 argument is actually only needed in GranSim, where we want to have the
2181 possibility to schedule *any* TSO on the run queue, irrespective of the
2182 actual ordering. Therefore, if tso is not the nil TSO then we traverse
2183 the run queue and dequeue the tso, adjusting the links in the queue.
2185 //@cindex take_off_run_queue
2186 static /* inline */ StgTSO*
2187 take_off_run_queue(StgTSO *tso) {
2191 qetlaHbogh Qu' ngaSbogh ghomDaQ {tso} yIteq!
2193 if tso is specified, unlink that tso from the run_queue (doesn't have
2194 to be at the beginning of the queue); GranSim only
2196 if (tso!=END_TSO_QUEUE) {
2197 /* find tso in queue */
2198 for (t=run_queue_hd, prev=END_TSO_QUEUE;
2199 t!=END_TSO_QUEUE && t!=tso;
2203 /* now actually dequeue the tso */
2204 if (prev!=END_TSO_QUEUE) {
2205 ASSERT(run_queue_hd!=t);
2206 prev->link = t->link;
2208 /* t is at beginning of thread queue */
2209 ASSERT(run_queue_hd==t);
2210 run_queue_hd = t->link;
2212 /* t is at end of thread queue */
2213 if (t->link==END_TSO_QUEUE) {
2214 ASSERT(t==run_queue_tl);
2215 run_queue_tl = prev;
2217 ASSERT(run_queue_tl!=t);
2219 t->link = END_TSO_QUEUE;
2221 /* take tso from the beginning of the queue; std concurrent code */
2223 if (t != END_TSO_QUEUE) {
2224 run_queue_hd = t->link;
2225 t->link = END_TSO_QUEUE;
2226 if (run_queue_hd == END_TSO_QUEUE) {
2227 run_queue_tl = END_TSO_QUEUE;
2236 //@node Garbage Collextion Routines, Blocking Queue Routines, Run queue code, Main scheduling code
2237 //@subsection Garbage Collextion Routines
2239 /* ---------------------------------------------------------------------------
2240 Where are the roots that we know about?
2242 - all the threads on the runnable queue
2243 - all the threads on the blocked queue
2244 - all the threads on the sleeping queue
2245 - all the thread currently executing a _ccall_GC
2246 - all the "main threads"
2248 ------------------------------------------------------------------------ */
2250 /* This has to be protected either by the scheduler monitor, or by the
2251 garbage collection monitor (probably the latter).
2256 GetRoots(evac_fn evac)
2263 for (i=0; i<=RtsFlags.GranFlags.proc; i++) {
2264 if ((run_queue_hds[i] != END_TSO_QUEUE) && ((run_queue_hds[i] != NULL)))
2265 evac((StgClosure **)&run_queue_hds[i]);
2266 if ((run_queue_tls[i] != END_TSO_QUEUE) && ((run_queue_tls[i] != NULL)))
2267 evac((StgClosure **)&run_queue_tls[i]);
2269 if ((blocked_queue_hds[i] != END_TSO_QUEUE) && ((blocked_queue_hds[i] != NULL)))
2270 evac((StgClosure **)&blocked_queue_hds[i]);
2271 if ((blocked_queue_tls[i] != END_TSO_QUEUE) && ((blocked_queue_tls[i] != NULL)))
2272 evac((StgClosure **)&blocked_queue_tls[i]);
2273 if ((ccalling_threadss[i] != END_TSO_QUEUE) && ((ccalling_threadss[i] != NULL)))
2274 evac((StgClosure **)&ccalling_threads[i]);
2281 if (run_queue_hd != END_TSO_QUEUE) {
2282 ASSERT(run_queue_tl != END_TSO_QUEUE);
2283 evac((StgClosure **)&run_queue_hd);
2284 evac((StgClosure **)&run_queue_tl);
2287 if (blocked_queue_hd != END_TSO_QUEUE) {
2288 ASSERT(blocked_queue_tl != END_TSO_QUEUE);
2289 evac((StgClosure **)&blocked_queue_hd);
2290 evac((StgClosure **)&blocked_queue_tl);
2293 if (sleeping_queue != END_TSO_QUEUE) {
2294 evac((StgClosure **)&sleeping_queue);
2298 for (m = main_threads; m != NULL; m = m->link) {
2299 evac((StgClosure **)&m->tso);
2301 if (suspended_ccalling_threads != END_TSO_QUEUE) {
2302 evac((StgClosure **)&suspended_ccalling_threads);
2305 #if defined(PAR) || defined(GRAN)
2306 markSparkQueue(evac);
2310 /* -----------------------------------------------------------------------------
2313 This is the interface to the garbage collector from Haskell land.
2314 We provide this so that external C code can allocate and garbage
2315 collect when called from Haskell via _ccall_GC.
2317 It might be useful to provide an interface whereby the programmer
2318 can specify more roots (ToDo).
2320 This needs to be protected by the GC condition variable above. KH.
2321 -------------------------------------------------------------------------- */
2323 void (*extra_roots)(evac_fn);
2328 GarbageCollect(GetRoots,rtsFalse);
2332 performMajorGC(void)
2334 GarbageCollect(GetRoots,rtsTrue);
2338 AllRoots(evac_fn evac)
2340 GetRoots(evac); // the scheduler's roots
2341 extra_roots(evac); // the user's roots
2345 performGCWithRoots(void (*get_roots)(evac_fn))
2347 extra_roots = get_roots;
2348 GarbageCollect(AllRoots,rtsFalse);
2351 /* -----------------------------------------------------------------------------
2354 If the thread has reached its maximum stack size, then raise the
2355 StackOverflow exception in the offending thread. Otherwise
2356 relocate the TSO into a larger chunk of memory and adjust its stack
2358 -------------------------------------------------------------------------- */
2361 threadStackOverflow(StgTSO *tso)
2363 nat new_stack_size, new_tso_size, diff, stack_words;
2367 IF_DEBUG(sanity,checkTSO(tso));
2368 if (tso->stack_size >= tso->max_stack_size) {
2371 belch("@@ threadStackOverflow of TSO %d (%p): stack too large (now %ld; max is %ld",
2372 tso->id, tso, tso->stack_size, tso->max_stack_size);
2373 /* If we're debugging, just print out the top of the stack */
2374 printStackChunk(tso->sp, stg_min(tso->stack+tso->stack_size,
2377 /* Send this thread the StackOverflow exception */
2378 raiseAsync(tso, (StgClosure *)stackOverflow_closure);
2382 /* Try to double the current stack size. If that takes us over the
2383 * maximum stack size for this thread, then use the maximum instead.
2384 * Finally round up so the TSO ends up as a whole number of blocks.
2386 new_stack_size = stg_min(tso->stack_size * 2, tso->max_stack_size);
2387 new_tso_size = (nat)BLOCK_ROUND_UP(new_stack_size * sizeof(W_) +
2388 TSO_STRUCT_SIZE)/sizeof(W_);
2389 new_tso_size = round_to_mblocks(new_tso_size); /* Be MBLOCK-friendly */
2390 new_stack_size = new_tso_size - TSO_STRUCT_SIZEW;
2392 IF_DEBUG(scheduler, fprintf(stderr,"== scheduler: increasing stack size from %d words to %d.\n", tso->stack_size, new_stack_size));
2394 dest = (StgTSO *)allocate(new_tso_size);
2395 TICK_ALLOC_TSO(new_stack_size,0);
2397 /* copy the TSO block and the old stack into the new area */
2398 memcpy(dest,tso,TSO_STRUCT_SIZE);
2399 stack_words = tso->stack + tso->stack_size - tso->sp;
2400 new_sp = (P_)dest + new_tso_size - stack_words;
2401 memcpy(new_sp, tso->sp, stack_words * sizeof(W_));
2403 /* relocate the stack pointers... */
2404 diff = (P_)new_sp - (P_)tso->sp; /* In *words* */
2405 dest->su = (StgUpdateFrame *) ((P_)dest->su + diff);
2407 dest->stack_size = new_stack_size;
2409 /* and relocate the update frame list */
2410 relocate_stack(dest, diff);
2412 /* Mark the old TSO as relocated. We have to check for relocated
2413 * TSOs in the garbage collector and any primops that deal with TSOs.
2415 * It's important to set the sp and su values to just beyond the end
2416 * of the stack, so we don't attempt to scavenge any part of the
2419 tso->what_next = ThreadRelocated;
2421 tso->sp = (P_)&(tso->stack[tso->stack_size]);
2422 tso->su = (StgUpdateFrame *)tso->sp;
2423 tso->why_blocked = NotBlocked;
2424 dest->mut_link = NULL;
2426 IF_PAR_DEBUG(verbose,
2427 belch("@@ threadStackOverflow of TSO %d (now at %p): stack size increased to %ld",
2428 tso->id, tso, tso->stack_size);
2429 /* If we're debugging, just print out the top of the stack */
2430 printStackChunk(tso->sp, stg_min(tso->stack+tso->stack_size,
2433 IF_DEBUG(sanity,checkTSO(tso));
2435 IF_DEBUG(scheduler,printTSO(dest));
2441 //@node Blocking Queue Routines, Exception Handling Routines, Garbage Collextion Routines, Main scheduling code
2442 //@subsection Blocking Queue Routines
2444 /* ---------------------------------------------------------------------------
2445 Wake up a queue that was blocked on some resource.
2446 ------------------------------------------------------------------------ */
2450 unblockCount ( StgBlockingQueueElement *bqe, StgClosure *node )
2455 unblockCount ( StgBlockingQueueElement *bqe, StgClosure *node )
2457 /* write RESUME events to log file and
2458 update blocked and fetch time (depending on type of the orig closure) */
2459 if (RtsFlags.ParFlags.ParStats.Full) {
2460 DumpRawGranEvent(CURRENT_PROC, CURRENT_PROC,
2461 GR_RESUMEQ, ((StgTSO *)bqe), ((StgTSO *)bqe)->block_info.closure,
2462 0, 0 /* spark_queue_len(ADVISORY_POOL) */);
2463 if (EMPTY_RUN_QUEUE())
2464 emitSchedule = rtsTrue;
2466 switch (get_itbl(node)->type) {
2468 ((StgTSO *)bqe)->par.fetchtime += CURRENT_TIME-((StgTSO *)bqe)->par.blockedat;
2473 ((StgTSO *)bqe)->par.blocktime += CURRENT_TIME-((StgTSO *)bqe)->par.blockedat;
2480 barf("{unblockOneLocked}Daq Qagh: unexpected closure in blocking queue");
2487 static StgBlockingQueueElement *
2488 unblockOneLocked(StgBlockingQueueElement *bqe, StgClosure *node)
2491 PEs node_loc, tso_loc;
2493 node_loc = where_is(node); // should be lifted out of loop
2494 tso = (StgTSO *)bqe; // wastes an assignment to get the type right
2495 tso_loc = where_is((StgClosure *)tso);
2496 if (IS_LOCAL_TO(PROCS(node),tso_loc)) { // TSO is local
2497 /* !fake_fetch => TSO is on CurrentProc is same as IS_LOCAL_TO */
2498 ASSERT(CurrentProc!=node_loc || tso_loc==CurrentProc);
2499 CurrentTime[CurrentProc] += RtsFlags.GranFlags.Costs.lunblocktime;
2500 // insertThread(tso, node_loc);
2501 new_event(tso_loc, tso_loc, CurrentTime[CurrentProc],
2503 tso, node, (rtsSpark*)NULL);
2504 tso->link = END_TSO_QUEUE; // overwrite link just to be sure
2507 } else { // TSO is remote (actually should be FMBQ)
2508 CurrentTime[CurrentProc] += RtsFlags.GranFlags.Costs.mpacktime +
2509 RtsFlags.GranFlags.Costs.gunblocktime +
2510 RtsFlags.GranFlags.Costs.latency;
2511 new_event(tso_loc, CurrentProc, CurrentTime[CurrentProc],
2513 tso, node, (rtsSpark*)NULL);
2514 tso->link = END_TSO_QUEUE; // overwrite link just to be sure
2517 /* the thread-queue-overhead is accounted for in either Resume or UnblockThread */
2519 fprintf(stderr," %s TSO %d (%p) [PE %d] (block_info.closure=%p) (next=%p) ,",
2520 (node_loc==tso_loc ? "Local" : "Global"),
2521 tso->id, tso, CurrentProc, tso->block_info.closure, tso->link));
2522 tso->block_info.closure = NULL;
2523 IF_DEBUG(scheduler,belch("-- Waking up thread %ld (%p)",
2527 static StgBlockingQueueElement *
2528 unblockOneLocked(StgBlockingQueueElement *bqe, StgClosure *node)
2530 StgBlockingQueueElement *next;
2532 switch (get_itbl(bqe)->type) {
2534 ASSERT(((StgTSO *)bqe)->why_blocked != NotBlocked);
2535 /* if it's a TSO just push it onto the run_queue */
2537 // ((StgTSO *)bqe)->link = END_TSO_QUEUE; // debugging?
2538 PUSH_ON_RUN_QUEUE((StgTSO *)bqe);
2540 unblockCount(bqe, node);
2541 /* reset blocking status after dumping event */
2542 ((StgTSO *)bqe)->why_blocked = NotBlocked;
2546 /* if it's a BLOCKED_FETCH put it on the PendingFetches list */
2548 bqe->link = (StgBlockingQueueElement *)PendingFetches;
2549 PendingFetches = (StgBlockedFetch *)bqe;
2553 /* can ignore this case in a non-debugging setup;
2554 see comments on RBHSave closures above */
2556 /* check that the closure is an RBHSave closure */
2557 ASSERT(get_itbl((StgClosure *)bqe) == &stg_RBH_Save_0_info ||
2558 get_itbl((StgClosure *)bqe) == &stg_RBH_Save_1_info ||
2559 get_itbl((StgClosure *)bqe) == &stg_RBH_Save_2_info);
2563 barf("{unblockOneLocked}Daq Qagh: Unexpected IP (%#lx; %s) in blocking queue at %#lx\n",
2564 get_itbl((StgClosure *)bqe), info_type((StgClosure *)bqe),
2568 IF_PAR_DEBUG(bq, fprintf(stderr, ", %p (%s)", bqe, info_type((StgClosure*)bqe)));
2572 #else /* !GRAN && !PAR */
2574 unblockOneLocked(StgTSO *tso)
2578 ASSERT(get_itbl(tso)->type == TSO);
2579 ASSERT(tso->why_blocked != NotBlocked);
2580 tso->why_blocked = NotBlocked;
2582 PUSH_ON_RUN_QUEUE(tso);
2584 IF_DEBUG(scheduler,sched_belch("waking up thread %ld", tso->id));
2589 #if defined(GRAN) || defined(PAR)
2590 inline StgBlockingQueueElement *
2591 unblockOne(StgBlockingQueueElement *bqe, StgClosure *node)
2593 ACQUIRE_LOCK(&sched_mutex);
2594 bqe = unblockOneLocked(bqe, node);
2595 RELEASE_LOCK(&sched_mutex);
2600 unblockOne(StgTSO *tso)
2602 ACQUIRE_LOCK(&sched_mutex);
2603 tso = unblockOneLocked(tso);
2604 RELEASE_LOCK(&sched_mutex);
2611 awakenBlockedQueue(StgBlockingQueueElement *q, StgClosure *node)
2613 StgBlockingQueueElement *bqe;
2618 belch("##-_ AwBQ for node %p on PE %d @ %ld by TSO %d (%p): ", \
2619 node, CurrentProc, CurrentTime[CurrentProc],
2620 CurrentTSO->id, CurrentTSO));
2622 node_loc = where_is(node);
2624 ASSERT(q == END_BQ_QUEUE ||
2625 get_itbl(q)->type == TSO || // q is either a TSO or an RBHSave
2626 get_itbl(q)->type == CONSTR); // closure (type constructor)
2627 ASSERT(is_unique(node));
2629 /* FAKE FETCH: magically copy the node to the tso's proc;
2630 no Fetch necessary because in reality the node should not have been
2631 moved to the other PE in the first place
2633 if (CurrentProc!=node_loc) {
2635 belch("## node %p is on PE %d but CurrentProc is %d (TSO %d); assuming fake fetch and adjusting bitmask (old: %#x)",
2636 node, node_loc, CurrentProc, CurrentTSO->id,
2637 // CurrentTSO, where_is(CurrentTSO),
2638 node->header.gran.procs));
2639 node->header.gran.procs = (node->header.gran.procs) | PE_NUMBER(CurrentProc);
2641 belch("## new bitmask of node %p is %#x",
2642 node, node->header.gran.procs));
2643 if (RtsFlags.GranFlags.GranSimStats.Global) {
2644 globalGranStats.tot_fake_fetches++;
2649 // ToDo: check: ASSERT(CurrentProc==node_loc);
2650 while (get_itbl(bqe)->type==TSO) { // q != END_TSO_QUEUE) {
2653 bqe points to the current element in the queue
2654 next points to the next element in the queue
2656 //tso = (StgTSO *)bqe; // wastes an assignment to get the type right
2657 //tso_loc = where_is(tso);
2659 bqe = unblockOneLocked(bqe, node);
2662 /* if this is the BQ of an RBH, we have to put back the info ripped out of
2663 the closure to make room for the anchor of the BQ */
2664 if (bqe!=END_BQ_QUEUE) {
2665 ASSERT(get_itbl(node)->type == RBH && get_itbl(bqe)->type == CONSTR);
2667 ASSERT((info_ptr==&RBH_Save_0_info) ||
2668 (info_ptr==&RBH_Save_1_info) ||
2669 (info_ptr==&RBH_Save_2_info));
2671 /* cf. convertToRBH in RBH.c for writing the RBHSave closure */
2672 ((StgRBH *)node)->blocking_queue = (StgBlockingQueueElement *)((StgRBHSave *)bqe)->payload[0];
2673 ((StgRBH *)node)->mut_link = (StgMutClosure *)((StgRBHSave *)bqe)->payload[1];
2676 belch("## Filled in RBH_Save for %p (%s) at end of AwBQ",
2677 node, info_type(node)));
2680 /* statistics gathering */
2681 if (RtsFlags.GranFlags.GranSimStats.Global) {
2682 // globalGranStats.tot_bq_processing_time += bq_processing_time;
2683 globalGranStats.tot_bq_len += len; // total length of all bqs awakened
2684 // globalGranStats.tot_bq_len_local += len_local; // same for local TSOs only
2685 globalGranStats.tot_awbq++; // total no. of bqs awakened
2688 fprintf(stderr,"## BQ Stats of %p: [%d entries] %s\n",
2689 node, len, (bqe!=END_BQ_QUEUE) ? "RBH" : ""));
2693 awakenBlockedQueue(StgBlockingQueueElement *q, StgClosure *node)
2695 StgBlockingQueueElement *bqe;
2697 ACQUIRE_LOCK(&sched_mutex);
2699 IF_PAR_DEBUG(verbose,
2700 belch("##-_ AwBQ for node %p on [%x]: ",
2704 if(get_itbl(q)->type == CONSTR || q==END_BQ_QUEUE) {
2705 IF_PAR_DEBUG(verbose, belch("## ... nothing to unblock so lets just return. RFP (BUG?)"));
2710 ASSERT(q == END_BQ_QUEUE ||
2711 get_itbl(q)->type == TSO ||
2712 get_itbl(q)->type == BLOCKED_FETCH ||
2713 get_itbl(q)->type == CONSTR);
2716 while (get_itbl(bqe)->type==TSO ||
2717 get_itbl(bqe)->type==BLOCKED_FETCH) {
2718 bqe = unblockOneLocked(bqe, node);
2720 RELEASE_LOCK(&sched_mutex);
2723 #else /* !GRAN && !PAR */
2725 awakenBlockedQueue(StgTSO *tso)
2727 ACQUIRE_LOCK(&sched_mutex);
2728 while (tso != END_TSO_QUEUE) {
2729 tso = unblockOneLocked(tso);
2731 RELEASE_LOCK(&sched_mutex);
2735 //@node Exception Handling Routines, Debugging Routines, Blocking Queue Routines, Main scheduling code
2736 //@subsection Exception Handling Routines
2738 /* ---------------------------------------------------------------------------
2740 - usually called inside a signal handler so it mustn't do anything fancy.
2741 ------------------------------------------------------------------------ */
2744 interruptStgRts(void)
2750 /* -----------------------------------------------------------------------------
2753 This is for use when we raise an exception in another thread, which
2755 This has nothing to do with the UnblockThread event in GranSim. -- HWL
2756 -------------------------------------------------------------------------- */
2758 #if defined(GRAN) || defined(PAR)
2760 NB: only the type of the blocking queue is different in GranSim and GUM
2761 the operations on the queue-elements are the same
2762 long live polymorphism!
2765 unblockThread(StgTSO *tso)
2767 StgBlockingQueueElement *t, **last;
2769 ACQUIRE_LOCK(&sched_mutex);
2770 switch (tso->why_blocked) {
2773 return; /* not blocked */
2776 ASSERT(get_itbl(tso->block_info.closure)->type == MVAR);
2778 StgBlockingQueueElement *last_tso = END_BQ_QUEUE;
2779 StgMVar *mvar = (StgMVar *)(tso->block_info.closure);
2781 last = (StgBlockingQueueElement **)&mvar->head;
2782 for (t = (StgBlockingQueueElement *)mvar->head;
2784 last = &t->link, last_tso = t, t = t->link) {
2785 if (t == (StgBlockingQueueElement *)tso) {
2786 *last = (StgBlockingQueueElement *)tso->link;
2787 if (mvar->tail == tso) {
2788 mvar->tail = (StgTSO *)last_tso;
2793 barf("unblockThread (MVAR): TSO not found");
2796 case BlockedOnBlackHole:
2797 ASSERT(get_itbl(tso->block_info.closure)->type == BLACKHOLE_BQ);
2799 StgBlockingQueue *bq = (StgBlockingQueue *)(tso->block_info.closure);
2801 last = &bq->blocking_queue;
2802 for (t = bq->blocking_queue;
2804 last = &t->link, t = t->link) {
2805 if (t == (StgBlockingQueueElement *)tso) {
2806 *last = (StgBlockingQueueElement *)tso->link;
2810 barf("unblockThread (BLACKHOLE): TSO not found");
2813 case BlockedOnException:
2815 StgTSO *target = tso->block_info.tso;
2817 ASSERT(get_itbl(target)->type == TSO);
2819 if (target->what_next == ThreadRelocated) {
2820 target = target->link;
2821 ASSERT(get_itbl(target)->type == TSO);
2824 ASSERT(target->blocked_exceptions != NULL);
2826 last = (StgBlockingQueueElement **)&target->blocked_exceptions;
2827 for (t = (StgBlockingQueueElement *)target->blocked_exceptions;
2829 last = &t->link, t = t->link) {
2830 ASSERT(get_itbl(t)->type == TSO);
2831 if (t == (StgBlockingQueueElement *)tso) {
2832 *last = (StgBlockingQueueElement *)tso->link;
2836 barf("unblockThread (Exception): TSO not found");
2840 case BlockedOnWrite:
2842 /* take TSO off blocked_queue */
2843 StgBlockingQueueElement *prev = NULL;
2844 for (t = (StgBlockingQueueElement *)blocked_queue_hd; t != END_BQ_QUEUE;
2845 prev = t, t = t->link) {
2846 if (t == (StgBlockingQueueElement *)tso) {
2848 blocked_queue_hd = (StgTSO *)t->link;
2849 if ((StgBlockingQueueElement *)blocked_queue_tl == t) {
2850 blocked_queue_tl = END_TSO_QUEUE;
2853 prev->link = t->link;
2854 if ((StgBlockingQueueElement *)blocked_queue_tl == t) {
2855 blocked_queue_tl = (StgTSO *)prev;
2861 barf("unblockThread (I/O): TSO not found");
2864 case BlockedOnDelay:
2866 /* take TSO off sleeping_queue */
2867 StgBlockingQueueElement *prev = NULL;
2868 for (t = (StgBlockingQueueElement *)sleeping_queue; t != END_BQ_QUEUE;
2869 prev = t, t = t->link) {
2870 if (t == (StgBlockingQueueElement *)tso) {
2872 sleeping_queue = (StgTSO *)t->link;
2874 prev->link = t->link;
2879 barf("unblockThread (I/O): TSO not found");
2883 barf("unblockThread");
2887 tso->link = END_TSO_QUEUE;
2888 tso->why_blocked = NotBlocked;
2889 tso->block_info.closure = NULL;
2890 PUSH_ON_RUN_QUEUE(tso);
2891 RELEASE_LOCK(&sched_mutex);
2895 unblockThread(StgTSO *tso)
2899 ACQUIRE_LOCK(&sched_mutex);
2900 switch (tso->why_blocked) {
2903 return; /* not blocked */
2906 ASSERT(get_itbl(tso->block_info.closure)->type == MVAR);
2908 StgTSO *last_tso = END_TSO_QUEUE;
2909 StgMVar *mvar = (StgMVar *)(tso->block_info.closure);
2912 for (t = mvar->head; t != END_TSO_QUEUE;
2913 last = &t->link, last_tso = t, t = t->link) {
2916 if (mvar->tail == tso) {
2917 mvar->tail = last_tso;
2922 barf("unblockThread (MVAR): TSO not found");
2925 case BlockedOnBlackHole:
2926 ASSERT(get_itbl(tso->block_info.closure)->type == BLACKHOLE_BQ);
2928 StgBlockingQueue *bq = (StgBlockingQueue *)(tso->block_info.closure);
2930 last = &bq->blocking_queue;
2931 for (t = bq->blocking_queue; t != END_TSO_QUEUE;
2932 last = &t->link, t = t->link) {
2938 barf("unblockThread (BLACKHOLE): TSO not found");
2941 case BlockedOnException:
2943 StgTSO *target = tso->block_info.tso;
2945 ASSERT(get_itbl(target)->type == TSO);
2947 while (target->what_next == ThreadRelocated) {
2948 target = target->link;
2949 ASSERT(get_itbl(target)->type == TSO);
2952 ASSERT(target->blocked_exceptions != NULL);
2954 last = &target->blocked_exceptions;
2955 for (t = target->blocked_exceptions; t != END_TSO_QUEUE;
2956 last = &t->link, t = t->link) {
2957 ASSERT(get_itbl(t)->type == TSO);
2963 barf("unblockThread (Exception): TSO not found");
2967 case BlockedOnWrite:
2969 StgTSO *prev = NULL;
2970 for (t = blocked_queue_hd; t != END_TSO_QUEUE;
2971 prev = t, t = t->link) {
2974 blocked_queue_hd = t->link;
2975 if (blocked_queue_tl == t) {
2976 blocked_queue_tl = END_TSO_QUEUE;
2979 prev->link = t->link;
2980 if (blocked_queue_tl == t) {
2981 blocked_queue_tl = prev;
2987 barf("unblockThread (I/O): TSO not found");
2990 case BlockedOnDelay:
2992 StgTSO *prev = NULL;
2993 for (t = sleeping_queue; t != END_TSO_QUEUE;
2994 prev = t, t = t->link) {
2997 sleeping_queue = t->link;
2999 prev->link = t->link;
3004 barf("unblockThread (I/O): TSO not found");
3008 barf("unblockThread");
3012 tso->link = END_TSO_QUEUE;
3013 tso->why_blocked = NotBlocked;
3014 tso->block_info.closure = NULL;
3015 PUSH_ON_RUN_QUEUE(tso);
3016 RELEASE_LOCK(&sched_mutex);
3020 /* -----------------------------------------------------------------------------
3023 * The following function implements the magic for raising an
3024 * asynchronous exception in an existing thread.
3026 * We first remove the thread from any queue on which it might be
3027 * blocked. The possible blockages are MVARs and BLACKHOLE_BQs.
3029 * We strip the stack down to the innermost CATCH_FRAME, building
3030 * thunks in the heap for all the active computations, so they can
3031 * be restarted if necessary. When we reach a CATCH_FRAME, we build
3032 * an application of the handler to the exception, and push it on
3033 * the top of the stack.
3035 * How exactly do we save all the active computations? We create an
3036 * AP_UPD for every UpdateFrame on the stack. Entering one of these
3037 * AP_UPDs pushes everything from the corresponding update frame
3038 * upwards onto the stack. (Actually, it pushes everything up to the
3039 * next update frame plus a pointer to the next AP_UPD object.
3040 * Entering the next AP_UPD object pushes more onto the stack until we
3041 * reach the last AP_UPD object - at which point the stack should look
3042 * exactly as it did when we killed the TSO and we can continue
3043 * execution by entering the closure on top of the stack.
3045 * We can also kill a thread entirely - this happens if either (a) the
3046 * exception passed to raiseAsync is NULL, or (b) there's no
3047 * CATCH_FRAME on the stack. In either case, we strip the entire
3048 * stack and replace the thread with a zombie.
3050 * -------------------------------------------------------------------------- */
3053 deleteThread(StgTSO *tso)
3055 raiseAsync(tso,NULL);
3059 raiseAsync(StgTSO *tso, StgClosure *exception)
3061 StgUpdateFrame* su = tso->su;
3062 StgPtr sp = tso->sp;
3064 /* Thread already dead? */
3065 if (tso->what_next == ThreadComplete || tso->what_next == ThreadKilled) {
3069 IF_DEBUG(scheduler, sched_belch("raising exception in thread %ld.", tso->id));
3071 /* Remove it from any blocking queues */
3074 /* The stack freezing code assumes there's a closure pointer on
3075 * the top of the stack. This isn't always the case with compiled
3076 * code, so we have to push a dummy closure on the top which just
3077 * returns to the next return address on the stack.
3079 if ( LOOKS_LIKE_GHC_INFO((void*)*sp) ) {
3080 *(--sp) = (W_)&stg_dummy_ret_closure;
3084 nat words = ((P_)su - (P_)sp) - 1;
3088 /* If we find a CATCH_FRAME, and we've got an exception to raise,
3089 * then build PAP(handler,exception,realworld#), and leave it on
3090 * top of the stack ready to enter.
3092 if (get_itbl(su)->type == CATCH_FRAME && exception != NULL) {
3093 StgCatchFrame *cf = (StgCatchFrame *)su;
3094 /* we've got an exception to raise, so let's pass it to the
3095 * handler in this frame.
3097 ap = (StgAP_UPD *)allocate(sizeofW(StgPAP) + 2);
3098 TICK_ALLOC_UPD_PAP(3,0);
3099 SET_HDR(ap,&stg_PAP_info,cf->header.prof.ccs);
3102 ap->fun = cf->handler; /* :: Exception -> IO a */
3103 ap->payload[0] = exception;
3104 ap->payload[1] = ARG_TAG(0); /* realworld token */
3106 /* throw away the stack from Sp up to and including the
3109 sp = (P_)su + sizeofW(StgCatchFrame) - 1;
3112 /* Restore the blocked/unblocked state for asynchronous exceptions
3113 * at the CATCH_FRAME.
3115 * If exceptions were unblocked at the catch, arrange that they
3116 * are unblocked again after executing the handler by pushing an
3117 * unblockAsyncExceptions_ret stack frame.
3119 if (!cf->exceptions_blocked) {
3120 *(sp--) = (W_)&stg_unblockAsyncExceptionszh_ret_info;
3123 /* Ensure that async exceptions are blocked when running the handler.
3125 if (tso->blocked_exceptions == NULL) {
3126 tso->blocked_exceptions = END_TSO_QUEUE;
3129 /* Put the newly-built PAP on top of the stack, ready to execute
3130 * when the thread restarts.
3134 tso->what_next = ThreadEnterGHC;
3135 IF_DEBUG(sanity, checkTSO(tso));
3139 /* First build an AP_UPD consisting of the stack chunk above the
3140 * current update frame, with the top word on the stack as the
3143 ap = (StgAP_UPD *)allocate(AP_sizeW(words));
3148 ap->fun = (StgClosure *)sp[0];
3150 for(i=0; i < (nat)words; ++i) {
3151 ap->payload[i] = (StgClosure *)*sp++;
3154 switch (get_itbl(su)->type) {
3158 SET_HDR(ap,&stg_AP_UPD_info,su->header.prof.ccs /* ToDo */);
3159 TICK_ALLOC_UP_THK(words+1,0);
3162 fprintf(stderr, "scheduler: Updating ");
3163 printPtr((P_)su->updatee);
3164 fprintf(stderr, " with ");
3165 printObj((StgClosure *)ap);
3168 /* Replace the updatee with an indirection - happily
3169 * this will also wake up any threads currently
3170 * waiting on the result.
3172 * Warning: if we're in a loop, more than one update frame on
3173 * the stack may point to the same object. Be careful not to
3174 * overwrite an IND_OLDGEN in this case, because we'll screw
3175 * up the mutable lists. To be on the safe side, don't
3176 * overwrite any kind of indirection at all. See also
3177 * threadSqueezeStack in GC.c, where we have to make a similar
3180 if (!closure_IND(su->updatee)) {
3181 UPD_IND_NOLOCK(su->updatee,ap); /* revert the black hole */
3184 sp += sizeofW(StgUpdateFrame) -1;
3185 sp[0] = (W_)ap; /* push onto stack */
3191 StgCatchFrame *cf = (StgCatchFrame *)su;
3194 /* We want a PAP, not an AP_UPD. Fortunately, the
3195 * layout's the same.
3197 SET_HDR(ap,&stg_PAP_info,su->header.prof.ccs /* ToDo */);
3198 TICK_ALLOC_UPD_PAP(words+1,0);
3200 /* now build o = FUN(catch,ap,handler) */
3201 o = (StgClosure *)allocate(sizeofW(StgClosure)+2);
3202 TICK_ALLOC_FUN(2,0);
3203 SET_HDR(o,&stg_catch_info,su->header.prof.ccs /* ToDo */);
3204 o->payload[0] = (StgClosure *)ap;
3205 o->payload[1] = cf->handler;
3208 fprintf(stderr, "scheduler: Built ");
3209 printObj((StgClosure *)o);
3212 /* pop the old handler and put o on the stack */
3214 sp += sizeofW(StgCatchFrame) - 1;
3221 StgSeqFrame *sf = (StgSeqFrame *)su;
3224 SET_HDR(ap,&stg_PAP_info,su->header.prof.ccs /* ToDo */);
3225 TICK_ALLOC_UPD_PAP(words+1,0);
3227 /* now build o = FUN(seq,ap) */
3228 o = (StgClosure *)allocate(sizeofW(StgClosure)+1);
3229 TICK_ALLOC_SE_THK(1,0);
3230 SET_HDR(o,&stg_seq_info,su->header.prof.ccs /* ToDo */);
3231 o->payload[0] = (StgClosure *)ap;
3234 fprintf(stderr, "scheduler: Built ");
3235 printObj((StgClosure *)o);
3238 /* pop the old handler and put o on the stack */
3240 sp += sizeofW(StgSeqFrame) - 1;
3246 /* We've stripped the entire stack, the thread is now dead. */
3247 sp += sizeofW(StgStopFrame) - 1;
3248 sp[0] = (W_)exception; /* save the exception */
3249 tso->what_next = ThreadKilled;
3250 tso->su = (StgUpdateFrame *)(sp+1);
3261 /* -----------------------------------------------------------------------------
3262 resurrectThreads is called after garbage collection on the list of
3263 threads found to be garbage. Each of these threads will be woken
3264 up and sent a signal: BlockedOnDeadMVar if the thread was blocked
3265 on an MVar, or NonTermination if the thread was blocked on a Black
3267 -------------------------------------------------------------------------- */
3270 resurrectThreads( StgTSO *threads )
3274 for (tso = threads; tso != END_TSO_QUEUE; tso = next) {
3275 next = tso->global_link;
3276 tso->global_link = all_threads;
3278 IF_DEBUG(scheduler, sched_belch("resurrecting thread %d", tso->id));
3280 switch (tso->why_blocked) {
3282 case BlockedOnException:
3283 raiseAsync(tso,(StgClosure *)BlockedOnDeadMVar_closure);
3285 case BlockedOnBlackHole:
3286 raiseAsync(tso,(StgClosure *)NonTermination_closure);
3289 /* This might happen if the thread was blocked on a black hole
3290 * belonging to a thread that we've just woken up (raiseAsync
3291 * can wake up threads, remember...).
3295 barf("resurrectThreads: thread blocked in a strange way");
3300 /* -----------------------------------------------------------------------------
3301 * Blackhole detection: if we reach a deadlock, test whether any
3302 * threads are blocked on themselves. Any threads which are found to
3303 * be self-blocked get sent a NonTermination exception.
3305 * This is only done in a deadlock situation in order to avoid
3306 * performance overhead in the normal case.
3307 * -------------------------------------------------------------------------- */
3310 detectBlackHoles( void )
3312 StgTSO *t = all_threads;
3313 StgUpdateFrame *frame;
3314 StgClosure *blocked_on;
3316 for (t = all_threads; t != END_TSO_QUEUE; t = t->global_link) {
3318 while (t->what_next == ThreadRelocated) {
3320 ASSERT(get_itbl(t)->type == TSO);
3323 if (t->why_blocked != BlockedOnBlackHole) {
3327 blocked_on = t->block_info.closure;
3329 for (frame = t->su; ; frame = frame->link) {
3330 switch (get_itbl(frame)->type) {
3333 if (frame->updatee == blocked_on) {
3334 /* We are blocking on one of our own computations, so
3335 * send this thread the NonTermination exception.
3338 sched_belch("thread %d is blocked on itself", t->id));
3339 raiseAsync(t, (StgClosure *)NonTermination_closure);
3360 //@node Debugging Routines, Index, Exception Handling Routines, Main scheduling code
3361 //@subsection Debugging Routines
3363 /* -----------------------------------------------------------------------------
3364 Debugging: why is a thread blocked
3365 -------------------------------------------------------------------------- */
3370 printThreadBlockage(StgTSO *tso)
3372 switch (tso->why_blocked) {
3374 fprintf(stderr,"is blocked on read from fd %d", tso->block_info.fd);
3376 case BlockedOnWrite:
3377 fprintf(stderr,"is blocked on write to fd %d", tso->block_info.fd);
3379 case BlockedOnDelay:
3380 fprintf(stderr,"is blocked until %d", tso->block_info.target);
3383 fprintf(stderr,"is blocked on an MVar");
3385 case BlockedOnException:
3386 fprintf(stderr,"is blocked on delivering an exception to thread %d",
3387 tso->block_info.tso->id);
3389 case BlockedOnBlackHole:
3390 fprintf(stderr,"is blocked on a black hole");
3393 fprintf(stderr,"is not blocked");
3397 fprintf(stderr,"is blocked on global address; local FM_BQ is %p (%s)",
3398 tso->block_info.closure, info_type(tso->block_info.closure));
3400 case BlockedOnGA_NoSend:
3401 fprintf(stderr,"is blocked on global address (no send); local FM_BQ is %p (%s)",
3402 tso->block_info.closure, info_type(tso->block_info.closure));
3406 barf("printThreadBlockage: strange tso->why_blocked: %d for TSO %d (%d)",
3407 tso->why_blocked, tso->id, tso);
3412 printThreadStatus(StgTSO *tso)
3414 switch (tso->what_next) {
3416 fprintf(stderr,"has been killed");
3418 case ThreadComplete:
3419 fprintf(stderr,"has completed");
3422 printThreadBlockage(tso);
3427 printAllThreads(void)
3432 char time_string[TIME_STR_LEN], node_str[NODE_STR_LEN];
3433 ullong_format_string(TIME_ON_PROC(CurrentProc),
3434 time_string, rtsFalse/*no commas!*/);
3436 sched_belch("all threads at [%s]:", time_string);
3438 char time_string[TIME_STR_LEN], node_str[NODE_STR_LEN];
3439 ullong_format_string(CURRENT_TIME,
3440 time_string, rtsFalse/*no commas!*/);
3442 sched_belch("all threads at [%s]:", time_string);
3444 sched_belch("all threads:");
3447 for (t = all_threads; t != END_TSO_QUEUE; t = t->global_link) {
3448 fprintf(stderr, "\tthread %d ", t->id);
3449 printThreadStatus(t);
3450 fprintf(stderr,"\n");
3455 Print a whole blocking queue attached to node (debugging only).
3460 print_bq (StgClosure *node)
3462 StgBlockingQueueElement *bqe;
3466 fprintf(stderr,"## BQ of closure %p (%s): ",
3467 node, info_type(node));
3469 /* should cover all closures that may have a blocking queue */
3470 ASSERT(get_itbl(node)->type == BLACKHOLE_BQ ||
3471 get_itbl(node)->type == FETCH_ME_BQ ||
3472 get_itbl(node)->type == RBH ||
3473 get_itbl(node)->type == MVAR);
3475 ASSERT(node!=(StgClosure*)NULL); // sanity check
3477 print_bqe(((StgBlockingQueue*)node)->blocking_queue);
3481 Print a whole blocking queue starting with the element bqe.
3484 print_bqe (StgBlockingQueueElement *bqe)
3489 NB: In a parallel setup a BQ of an RBH must end with an RBH_Save closure;
3491 for (end = (bqe==END_BQ_QUEUE);
3492 !end; // iterate until bqe points to a CONSTR
3493 end = (get_itbl(bqe)->type == CONSTR) || (bqe->link==END_BQ_QUEUE),
3494 bqe = end ? END_BQ_QUEUE : bqe->link) {
3495 ASSERT(bqe != END_BQ_QUEUE); // sanity check
3496 ASSERT(bqe != (StgBlockingQueueElement *)NULL); // sanity check
3497 /* types of closures that may appear in a blocking queue */
3498 ASSERT(get_itbl(bqe)->type == TSO ||
3499 get_itbl(bqe)->type == BLOCKED_FETCH ||
3500 get_itbl(bqe)->type == CONSTR);
3501 /* only BQs of an RBH end with an RBH_Save closure */
3502 //ASSERT(get_itbl(bqe)->type != CONSTR || get_itbl(node)->type == RBH);
3504 switch (get_itbl(bqe)->type) {
3506 fprintf(stderr," TSO %u (%x),",
3507 ((StgTSO *)bqe)->id, ((StgTSO *)bqe));
3510 fprintf(stderr," BF (node=%p, ga=((%x, %d, %x)),",
3511 ((StgBlockedFetch *)bqe)->node,
3512 ((StgBlockedFetch *)bqe)->ga.payload.gc.gtid,
3513 ((StgBlockedFetch *)bqe)->ga.payload.gc.slot,
3514 ((StgBlockedFetch *)bqe)->ga.weight);
3517 fprintf(stderr," %s (IP %p),",
3518 (get_itbl(bqe) == &stg_RBH_Save_0_info ? "RBH_Save_0" :
3519 get_itbl(bqe) == &stg_RBH_Save_1_info ? "RBH_Save_1" :
3520 get_itbl(bqe) == &stg_RBH_Save_2_info ? "RBH_Save_2" :
3521 "RBH_Save_?"), get_itbl(bqe));
3524 barf("Unexpected closure type %s in blocking queue", // of %p (%s)",
3525 info_type((StgClosure *)bqe)); // , node, info_type(node));
3529 fputc('\n', stderr);
3531 # elif defined(GRAN)
3533 print_bq (StgClosure *node)
3535 StgBlockingQueueElement *bqe;
3536 PEs node_loc, tso_loc;
3539 /* should cover all closures that may have a blocking queue */
3540 ASSERT(get_itbl(node)->type == BLACKHOLE_BQ ||
3541 get_itbl(node)->type == FETCH_ME_BQ ||
3542 get_itbl(node)->type == RBH);
3544 ASSERT(node!=(StgClosure*)NULL); // sanity check
3545 node_loc = where_is(node);
3547 fprintf(stderr,"## BQ of closure %p (%s) on [PE %d]: ",
3548 node, info_type(node), node_loc);
3551 NB: In a parallel setup a BQ of an RBH must end with an RBH_Save closure;
3553 for (bqe = ((StgBlockingQueue*)node)->blocking_queue, end = (bqe==END_BQ_QUEUE);
3554 !end; // iterate until bqe points to a CONSTR
3555 end = (get_itbl(bqe)->type == CONSTR) || (bqe->link==END_BQ_QUEUE), bqe = end ? END_BQ_QUEUE : bqe->link) {
3556 ASSERT(bqe != END_BQ_QUEUE); // sanity check
3557 ASSERT(bqe != (StgBlockingQueueElement *)NULL); // sanity check
3558 /* types of closures that may appear in a blocking queue */
3559 ASSERT(get_itbl(bqe)->type == TSO ||
3560 get_itbl(bqe)->type == CONSTR);
3561 /* only BQs of an RBH end with an RBH_Save closure */
3562 ASSERT(get_itbl(bqe)->type != CONSTR || get_itbl(node)->type == RBH);
3564 tso_loc = where_is((StgClosure *)bqe);
3565 switch (get_itbl(bqe)->type) {
3567 fprintf(stderr," TSO %d (%p) on [PE %d],",
3568 ((StgTSO *)bqe)->id, (StgTSO *)bqe, tso_loc);
3571 fprintf(stderr," %s (IP %p),",
3572 (get_itbl(bqe) == &stg_RBH_Save_0_info ? "RBH_Save_0" :
3573 get_itbl(bqe) == &stg_RBH_Save_1_info ? "RBH_Save_1" :
3574 get_itbl(bqe) == &stg_RBH_Save_2_info ? "RBH_Save_2" :
3575 "RBH_Save_?"), get_itbl(bqe));
3578 barf("Unexpected closure type %s in blocking queue of %p (%s)",
3579 info_type((StgClosure *)bqe), node, info_type(node));
3583 fputc('\n', stderr);
3587 Nice and easy: only TSOs on the blocking queue
3590 print_bq (StgClosure *node)
3594 ASSERT(node!=(StgClosure*)NULL); // sanity check
3595 for (tso = ((StgBlockingQueue*)node)->blocking_queue;
3596 tso != END_TSO_QUEUE;
3598 ASSERT(tso!=NULL && tso!=END_TSO_QUEUE); // sanity check
3599 ASSERT(get_itbl(tso)->type == TSO); // guess what, sanity check
3600 fprintf(stderr," TSO %d (%p),", tso->id, tso);
3602 fputc('\n', stderr);
3613 for (i=0, tso=run_queue_hd;
3614 tso != END_TSO_QUEUE;
3623 sched_belch(char *s, ...)
3628 fprintf(stderr, "scheduler (task %ld): ", osThreadId());
3630 fprintf(stderr, "== ");
3632 fprintf(stderr, "scheduler: ");
3634 vfprintf(stderr, s, ap);
3635 fprintf(stderr, "\n");
3641 //@node Index, , Debugging Routines, Main scheduling code
3645 //* MainRegTable:: @cindex\s-+MainRegTable
3646 //* StgMainThread:: @cindex\s-+StgMainThread
3647 //* awaken_blocked_queue:: @cindex\s-+awaken_blocked_queue
3648 //* blocked_queue_hd:: @cindex\s-+blocked_queue_hd
3649 //* blocked_queue_tl:: @cindex\s-+blocked_queue_tl
3650 //* context_switch:: @cindex\s-+context_switch
3651 //* createThread:: @cindex\s-+createThread
3652 //* gc_pending_cond:: @cindex\s-+gc_pending_cond
3653 //* initScheduler:: @cindex\s-+initScheduler
3654 //* interrupted:: @cindex\s-+interrupted
3655 //* next_thread_id:: @cindex\s-+next_thread_id
3656 //* print_bq:: @cindex\s-+print_bq
3657 //* run_queue_hd:: @cindex\s-+run_queue_hd
3658 //* run_queue_tl:: @cindex\s-+run_queue_tl
3659 //* sched_mutex:: @cindex\s-+sched_mutex
3660 //* schedule:: @cindex\s-+schedule
3661 //* take_off_run_queue:: @cindex\s-+take_off_run_queue
3662 //* term_mutex:: @cindex\s-+term_mutex
3663 //* thread_ready_cond:: @cindex\s-+thread_ready_cond