Interruptible FFI calls with pthread_kill and CancelSynchronousIO. v4
[ghc.git] / rts / RaiseAsync.c
1 /* ---------------------------------------------------------------------------
2 *
3 * (c) The GHC Team, 1998-2006
4 *
5 * Asynchronous exceptions
6 *
7 * --------------------------------------------------------------------------*/
8
9 #include "PosixSource.h"
10 #include "Rts.h"
11
12 #include "sm/Storage.h"
13 #include "Threads.h"
14 #include "Trace.h"
15 #include "RaiseAsync.h"
16 #include "Schedule.h"
17 #include "Updates.h"
18 #include "STM.h"
19 #include "sm/Sanity.h"
20 #include "Profiling.h"
21 #include "Messages.h"
22 #if defined(mingw32_HOST_OS)
23 #include "win32/IOManager.h"
24 #endif
25
26 static void raiseAsync (Capability *cap,
27 StgTSO *tso,
28 StgClosure *exception,
29 rtsBool stop_at_atomically,
30 StgUpdateFrame *stop_here);
31
32 static void removeFromQueues(Capability *cap, StgTSO *tso);
33
34 static void removeFromMVarBlockedQueue (StgTSO *tso);
35
36 static void blockedThrowTo (Capability *cap,
37 StgTSO *target, MessageThrowTo *msg);
38
39 static void throwToSendMsg (Capability *cap USED_IF_THREADS,
40 Capability *target_cap USED_IF_THREADS,
41 MessageThrowTo *msg USED_IF_THREADS);
42
43 /* -----------------------------------------------------------------------------
44 throwToSingleThreaded
45
46 This version of throwTo is safe to use if and only if one of the
47 following holds:
48
49 - !THREADED_RTS
50
51 - all the other threads in the system are stopped (eg. during GC).
52
53 - we surely own the target TSO (eg. we just took it from the
54 run queue of the current capability, or we are running it).
55
56 It doesn't cater for blocking the source thread until the exception
57 has been raised.
58 -------------------------------------------------------------------------- */
59
60 void
61 throwToSingleThreaded(Capability *cap, StgTSO *tso, StgClosure *exception)
62 {
63 throwToSingleThreaded_(cap, tso, exception, rtsFalse);
64 }
65
66 void
67 throwToSingleThreaded_(Capability *cap, StgTSO *tso, StgClosure *exception,
68 rtsBool stop_at_atomically)
69 {
70 tso = deRefTSO(tso);
71
72 // Thread already dead?
73 if (tso->what_next == ThreadComplete || tso->what_next == ThreadKilled) {
74 return;
75 }
76
77 // Remove it from any blocking queues
78 removeFromQueues(cap,tso);
79
80 raiseAsync(cap, tso, exception, stop_at_atomically, NULL);
81 }
82
83 void
84 suspendComputation(Capability *cap, StgTSO *tso, StgUpdateFrame *stop_here)
85 {
86 tso = deRefTSO(tso);
87
88 // Thread already dead?
89 if (tso->what_next == ThreadComplete || tso->what_next == ThreadKilled) {
90 return;
91 }
92
93 // Remove it from any blocking queues
94 removeFromQueues(cap,tso);
95
96 raiseAsync(cap, tso, NULL, rtsFalse, stop_here);
97 }
98
99 /* -----------------------------------------------------------------------------
100 throwTo
101
102 This function may be used to throw an exception from one thread to
103 another, during the course of normal execution. This is a tricky
104 task: the target thread might be running on another CPU, or it
105 may be blocked and could be woken up at any point by another CPU.
106 We have some delicate synchronisation to do.
107
108 The underlying scheme when multiple Capabilities are in use is
109 message passing: when the target of a throwTo is on another
110 Capability, we send a message (a MessageThrowTo closure) to that
111 Capability.
112
113 If the throwTo needs to block because the target TSO is masking
114 exceptions (the TSO_BLOCKEX flag), then the message is placed on
115 the blocked_exceptions queue attached to the target TSO. When the
116 target TSO enters the unmasked state again, it must check the
117 queue. The blocked_exceptions queue is not locked; only the
118 Capability owning the TSO may modify it.
119
120 To make things simpler for throwTo, we always create the message
121 first before deciding what to do. The message may get sent, or it
122 may get attached to a TSO's blocked_exceptions queue, or the
123 exception may get thrown immediately and the message dropped,
124 depending on the current state of the target.
125
126 Currently we send a message if the target belongs to another
127 Capability, and it is
128
129 - NotBlocked, BlockedOnMsgThrowTo,
130 BlockedOnCCall_Interruptible
131
132 - or it is masking exceptions (TSO_BLOCKEX)
133
134 Currently, if the target is BlockedOnMVar, BlockedOnSTM, or
135 BlockedOnBlackHole then we acquire ownership of the TSO by locking
136 its parent container (e.g. the MVar) and then raise the exception.
137 We might change these cases to be more message-passing-like in the
138 future.
139
140 Returns:
141
142 NULL exception was raised, ok to continue
143
144 MessageThrowTo * exception was not raised; the source TSO
145 should now put itself in the state
146 BlockedOnMsgThrowTo, and when it is ready
147 it should unlock the mssage using
148 unlockClosure(msg, &stg_MSG_THROWTO_info);
149 If it decides not to raise the exception after
150 all, it can revoke it safely with
151 unlockClosure(msg, &stg_MSG_NULL_info);
152
153 -------------------------------------------------------------------------- */
154
155 MessageThrowTo *
156 throwTo (Capability *cap, // the Capability we hold
157 StgTSO *source, // the TSO sending the exception (or NULL)
158 StgTSO *target, // the TSO receiving the exception
159 StgClosure *exception) // the exception closure
160 {
161 MessageThrowTo *msg;
162
163 msg = (MessageThrowTo *) allocate(cap, sizeofW(MessageThrowTo));
164 // message starts locked; the caller has to unlock it when it is
165 // ready.
166 SET_HDR(msg, &stg_WHITEHOLE_info, CCS_SYSTEM);
167 msg->source = source;
168 msg->target = target;
169 msg->exception = exception;
170
171 switch (throwToMsg(cap, msg))
172 {
173 case THROWTO_SUCCESS:
174 return NULL;
175 case THROWTO_BLOCKED:
176 default:
177 return msg;
178 }
179 }
180
181
182 nat
183 throwToMsg (Capability *cap, MessageThrowTo *msg)
184 {
185 StgWord status;
186 StgTSO *target = msg->target;
187 Capability *target_cap;
188
189 goto check_target;
190
191 retry:
192 write_barrier();
193 debugTrace(DEBUG_sched, "throwTo: retrying...");
194
195 check_target:
196 ASSERT(target != END_TSO_QUEUE);
197
198 // follow ThreadRelocated links in the target first
199 target = deRefTSO(target);
200
201 // Thread already dead?
202 if (target->what_next == ThreadComplete
203 || target->what_next == ThreadKilled) {
204 return THROWTO_SUCCESS;
205 }
206
207 debugTraceCap(DEBUG_sched, cap,
208 "throwTo: from thread %lu to thread %lu",
209 (unsigned long)msg->source->id,
210 (unsigned long)msg->target->id);
211
212 #ifdef DEBUG
213 traceThreadStatus(DEBUG_sched, target);
214 #endif
215
216 target_cap = target->cap;
217 if (target->cap != cap) {
218 throwToSendMsg(cap, target_cap, msg);
219 return THROWTO_BLOCKED;
220 }
221
222 status = target->why_blocked;
223
224 switch (status) {
225 case NotBlocked:
226 {
227 if ((target->flags & TSO_BLOCKEX) == 0) {
228 // It's on our run queue and not blocking exceptions
229 raiseAsync(cap, target, msg->exception, rtsFalse, NULL);
230 return THROWTO_SUCCESS;
231 } else {
232 blockedThrowTo(cap,target,msg);
233 return THROWTO_BLOCKED;
234 }
235 }
236
237 case BlockedOnMsgThrowTo:
238 {
239 const StgInfoTable *i;
240 MessageThrowTo *m;
241
242 m = target->block_info.throwto;
243
244 // target is local to this cap, but has sent a throwto
245 // message to another cap.
246 //
247 // The source message is locked. We need to revoke the
248 // target's message so that we can raise the exception, so
249 // we attempt to lock it.
250
251 // There's a possibility of a deadlock if two threads are both
252 // trying to throwTo each other (or more generally, a cycle of
253 // threads). To break the symmetry we compare the addresses
254 // of the MessageThrowTo objects, and the one for which m <
255 // msg gets to spin, while the other can only try to lock
256 // once, but must then back off and unlock both before trying
257 // again.
258 if (m < msg) {
259 i = lockClosure((StgClosure *)m);
260 } else {
261 i = tryLockClosure((StgClosure *)m);
262 if (i == NULL) {
263 // debugBelch("collision\n");
264 throwToSendMsg(cap, target->cap, msg);
265 return THROWTO_BLOCKED;
266 }
267 }
268
269 if (i == &stg_MSG_NULL_info) {
270 // we know there's a MSG_TRY_WAKEUP on the way, so we
271 // might as well just do it now. The message will
272 // be a no-op when it arrives.
273 unlockClosure((StgClosure*)m, i);
274 tryWakeupThread_(cap, target);
275 goto retry;
276 }
277
278 if (i != &stg_MSG_THROWTO_info) {
279 // if it's a MSG_NULL, this TSO has been woken up by another Cap
280 unlockClosure((StgClosure*)m, i);
281 goto retry;
282 }
283
284 if ((target->flags & TSO_BLOCKEX) &&
285 ((target->flags & TSO_INTERRUPTIBLE) == 0)) {
286 unlockClosure((StgClosure*)m, i);
287 blockedThrowTo(cap,target,msg);
288 return THROWTO_BLOCKED;
289 }
290
291 // nobody else can wake up this TSO after we claim the message
292 unlockClosure((StgClosure*)m, &stg_MSG_NULL_info);
293
294 raiseAsync(cap, target, msg->exception, rtsFalse, NULL);
295 return THROWTO_SUCCESS;
296 }
297
298 case BlockedOnMVar:
299 {
300 /*
301 To establish ownership of this TSO, we need to acquire a
302 lock on the MVar that it is blocked on.
303 */
304 StgMVar *mvar;
305 StgInfoTable *info USED_IF_THREADS;
306
307 mvar = (StgMVar *)target->block_info.closure;
308
309 // ASSUMPTION: tso->block_info must always point to a
310 // closure. In the threaded RTS it does.
311 switch (get_itbl(mvar)->type) {
312 case MVAR_CLEAN:
313 case MVAR_DIRTY:
314 break;
315 default:
316 goto retry;
317 }
318
319 info = lockClosure((StgClosure *)mvar);
320
321 if (target->what_next == ThreadRelocated) {
322 target = target->_link;
323 unlockClosure((StgClosure *)mvar,info);
324 goto retry;
325 }
326 // we have the MVar, let's check whether the thread
327 // is still blocked on the same MVar.
328 if (target->why_blocked != BlockedOnMVar
329 || (StgMVar *)target->block_info.closure != mvar) {
330 unlockClosure((StgClosure *)mvar, info);
331 goto retry;
332 }
333
334 if (target->_link == END_TSO_QUEUE) {
335 // the MVar operation has already completed. There is a
336 // MSG_TRY_WAKEUP on the way, but we can just wake up the
337 // thread now anyway and ignore the message when it
338 // arrives.
339 unlockClosure((StgClosure *)mvar, info);
340 tryWakeupThread_(cap, target);
341 goto retry;
342 }
343
344 if ((target->flags & TSO_BLOCKEX) &&
345 ((target->flags & TSO_INTERRUPTIBLE) == 0)) {
346 blockedThrowTo(cap,target,msg);
347 unlockClosure((StgClosure *)mvar, info);
348 return THROWTO_BLOCKED;
349 } else {
350 // revoke the MVar operation
351 removeFromMVarBlockedQueue(target);
352 raiseAsync(cap, target, msg->exception, rtsFalse, NULL);
353 unlockClosure((StgClosure *)mvar, info);
354 return THROWTO_SUCCESS;
355 }
356 }
357
358 case BlockedOnBlackHole:
359 {
360 if (target->flags & TSO_BLOCKEX) {
361 // BlockedOnBlackHole is not interruptible.
362 blockedThrowTo(cap,target,msg);
363 return THROWTO_BLOCKED;
364 } else {
365 // Revoke the message by replacing it with IND. We're not
366 // locking anything here, so we might still get a TRY_WAKEUP
367 // message from the owner of the blackhole some time in the
368 // future, but that doesn't matter.
369 ASSERT(target->block_info.bh->header.info == &stg_MSG_BLACKHOLE_info);
370 OVERWRITE_INFO(target->block_info.bh, &stg_IND_info);
371 raiseAsync(cap, target, msg->exception, rtsFalse, NULL);
372 return THROWTO_SUCCESS;
373 }
374 }
375
376 case BlockedOnSTM:
377 lockTSO(target);
378 // Unblocking BlockedOnSTM threads requires the TSO to be
379 // locked; see STM.c:unpark_tso().
380 if (target->why_blocked != BlockedOnSTM) {
381 unlockTSO(target);
382 goto retry;
383 }
384 if ((target->flags & TSO_BLOCKEX) &&
385 ((target->flags & TSO_INTERRUPTIBLE) == 0)) {
386 blockedThrowTo(cap,target,msg);
387 unlockTSO(target);
388 return THROWTO_BLOCKED;
389 } else {
390 raiseAsync(cap, target, msg->exception, rtsFalse, NULL);
391 unlockTSO(target);
392 return THROWTO_SUCCESS;
393 }
394
395 case BlockedOnCCall_Interruptible:
396 #ifdef THREADED_RTS
397 {
398 Task *task = NULL;
399 // walk suspended_ccalls to find the correct worker thread
400 InCall *incall;
401 for (incall = cap->suspended_ccalls; incall != NULL; incall = incall->next) {
402 if (incall->suspended_tso == target) {
403 task = incall->task;
404 break;
405 }
406 }
407 if (task != NULL) {
408 raiseAsync(cap, target, msg->exception, rtsFalse, NULL);
409 interruptWorkerTask(task);
410 return THROWTO_SUCCESS;
411 } else {
412 debugTraceCap(DEBUG_sched, cap, "throwTo: could not find worker thread to kill");
413 }
414 // fall to next
415 }
416 #endif
417 case BlockedOnCCall:
418 blockedThrowTo(cap,target,msg);
419 return THROWTO_BLOCKED;
420
421 #ifndef THREADEDED_RTS
422 case BlockedOnRead:
423 case BlockedOnWrite:
424 case BlockedOnDelay:
425 #if defined(mingw32_HOST_OS)
426 case BlockedOnDoProc:
427 #endif
428 if ((target->flags & TSO_BLOCKEX) &&
429 ((target->flags & TSO_INTERRUPTIBLE) == 0)) {
430 blockedThrowTo(cap,target,msg);
431 return THROWTO_BLOCKED;
432 } else {
433 removeFromQueues(cap,target);
434 raiseAsync(cap, target, msg->exception, rtsFalse, NULL);
435 return THROWTO_SUCCESS;
436 }
437 #endif
438
439 default:
440 barf("throwTo: unrecognised why_blocked value");
441 }
442 barf("throwTo");
443 }
444
445 static void
446 throwToSendMsg (Capability *cap STG_UNUSED,
447 Capability *target_cap USED_IF_THREADS,
448 MessageThrowTo *msg USED_IF_THREADS)
449
450 {
451 #ifdef THREADED_RTS
452 debugTraceCap(DEBUG_sched, cap, "throwTo: sending a throwto message to cap %lu", (unsigned long)target_cap->no);
453
454 sendMessage(cap, target_cap, (Message*)msg);
455 #endif
456 }
457
458 // Block a throwTo message on the target TSO's blocked_exceptions
459 // queue. The current Capability must own the target TSO in order to
460 // modify the blocked_exceptions queue.
461 static void
462 blockedThrowTo (Capability *cap, StgTSO *target, MessageThrowTo *msg)
463 {
464 debugTraceCap(DEBUG_sched, cap, "throwTo: blocking on thread %lu",
465 (unsigned long)target->id);
466
467 ASSERT(target->cap == cap);
468
469 msg->link = target->blocked_exceptions;
470 target->blocked_exceptions = msg;
471 dirty_TSO(cap,target); // we modified the blocked_exceptions queue
472 }
473
474 /* -----------------------------------------------------------------------------
475 Waking up threads blocked in throwTo
476
477 There are two ways to do this: maybePerformBlockedException() will
478 perform the throwTo() for the thread at the head of the queue
479 immediately, and leave the other threads on the queue.
480 maybePerformBlockedException() also checks the TSO_BLOCKEX flag
481 before raising an exception.
482
483 awakenBlockedExceptionQueue() will wake up all the threads in the
484 queue, but not perform any throwTo() immediately. This might be
485 more appropriate when the target thread is the one actually running
486 (see Exception.cmm).
487
488 Returns: non-zero if an exception was raised, zero otherwise.
489 -------------------------------------------------------------------------- */
490
491 int
492 maybePerformBlockedException (Capability *cap, StgTSO *tso)
493 {
494 MessageThrowTo *msg;
495 const StgInfoTable *i;
496
497 if (tso->what_next == ThreadComplete || tso->what_next == ThreadFinished) {
498 if (tso->blocked_exceptions != END_BLOCKED_EXCEPTIONS_QUEUE) {
499 awakenBlockedExceptionQueue(cap,tso);
500 return 1;
501 } else {
502 return 0;
503 }
504 }
505
506 if (tso->blocked_exceptions != END_BLOCKED_EXCEPTIONS_QUEUE &&
507 (tso->flags & TSO_BLOCKEX) != 0) {
508 debugTraceCap(DEBUG_sched, cap, "throwTo: thread %lu has blocked exceptions but is inside block", (unsigned long)tso->id);
509 }
510
511 if (tso->blocked_exceptions != END_BLOCKED_EXCEPTIONS_QUEUE
512 && ((tso->flags & TSO_BLOCKEX) == 0
513 || ((tso->flags & TSO_INTERRUPTIBLE) && interruptible(tso)))) {
514
515 // We unblock just the first thread on the queue, and perform
516 // its throw immediately.
517 loop:
518 msg = tso->blocked_exceptions;
519 if (msg == END_BLOCKED_EXCEPTIONS_QUEUE) return 0;
520 i = lockClosure((StgClosure*)msg);
521 tso->blocked_exceptions = (MessageThrowTo*)msg->link;
522 if (i == &stg_MSG_NULL_info) {
523 unlockClosure((StgClosure*)msg,i);
524 goto loop;
525 }
526
527 throwToSingleThreaded(cap, msg->target, msg->exception);
528 unlockClosure((StgClosure*)msg,&stg_MSG_NULL_info);
529 tryWakeupThread(cap, msg->source);
530 return 1;
531 }
532 return 0;
533 }
534
535 // awakenBlockedExceptionQueue(): Just wake up the whole queue of
536 // blocked exceptions.
537
538 void
539 awakenBlockedExceptionQueue (Capability *cap, StgTSO *tso)
540 {
541 MessageThrowTo *msg;
542 const StgInfoTable *i;
543
544 for (msg = tso->blocked_exceptions; msg != END_BLOCKED_EXCEPTIONS_QUEUE;
545 msg = (MessageThrowTo*)msg->link) {
546 i = lockClosure((StgClosure *)msg);
547 if (i != &stg_MSG_NULL_info) {
548 unlockClosure((StgClosure *)msg,&stg_MSG_NULL_info);
549 tryWakeupThread(cap, msg->source);
550 } else {
551 unlockClosure((StgClosure *)msg,i);
552 }
553 }
554 tso->blocked_exceptions = END_BLOCKED_EXCEPTIONS_QUEUE;
555 }
556
557 /* -----------------------------------------------------------------------------
558 Remove a thread from blocking queues.
559
560 This is for use when we raise an exception in another thread, which
561 may be blocked.
562
563 Precondition: we have exclusive access to the TSO, via the same set
564 of conditions as throwToSingleThreaded() (c.f.).
565 -------------------------------------------------------------------------- */
566
567 static void
568 removeFromMVarBlockedQueue (StgTSO *tso)
569 {
570 StgMVar *mvar = (StgMVar*)tso->block_info.closure;
571 StgMVarTSOQueue *q = (StgMVarTSOQueue*)tso->_link;
572
573 if (q == (StgMVarTSOQueue*)END_TSO_QUEUE) {
574 // already removed from this MVar
575 return;
576 }
577
578 // Assume the MVar is locked. (not assertable; sometimes it isn't
579 // actually WHITEHOLE'd).
580
581 // We want to remove the MVAR_TSO_QUEUE object from the queue. It
582 // isn't doubly-linked so we can't actually remove it; instead we
583 // just overwrite it with an IND if possible and let the GC short
584 // it out. However, we have to be careful to maintain the deque
585 // structure:
586
587 if (mvar->head == q) {
588 mvar->head = q->link;
589 q->header.info = &stg_IND_info;
590 if (mvar->tail == q) {
591 mvar->tail = (StgMVarTSOQueue*)END_TSO_QUEUE;
592 }
593 }
594 else if (mvar->tail == q) {
595 // we can't replace it with an IND in this case, because then
596 // we lose the tail pointer when the GC shorts out the IND.
597 // So we use MSG_NULL as a kind of non-dupable indirection;
598 // these are ignored by takeMVar/putMVar.
599 q->header.info = &stg_MSG_NULL_info;
600 }
601 else {
602 q->header.info = &stg_IND_info;
603 }
604
605 // revoke the MVar operation
606 tso->_link = END_TSO_QUEUE;
607 }
608
609 static void
610 removeFromQueues(Capability *cap, StgTSO *tso)
611 {
612 switch (tso->why_blocked) {
613
614 case NotBlocked:
615 case ThreadMigrating:
616 return;
617
618 case BlockedOnSTM:
619 // Be careful: nothing to do here! We tell the scheduler that the
620 // thread is runnable and we leave it to the stack-walking code to
621 // abort the transaction while unwinding the stack. We should
622 // perhaps have a debugging test to make sure that this really
623 // happens and that the 'zombie' transaction does not get
624 // committed.
625 goto done;
626
627 case BlockedOnMVar:
628 removeFromMVarBlockedQueue(tso);
629 goto done;
630
631 case BlockedOnBlackHole:
632 // nothing to do
633 goto done;
634
635 case BlockedOnMsgThrowTo:
636 {
637 MessageThrowTo *m = tso->block_info.throwto;
638 // The message is locked by us, unless we got here via
639 // deleteAllThreads(), in which case we own all the
640 // capabilities.
641 // ASSERT(m->header.info == &stg_WHITEHOLE_info);
642
643 // unlock and revoke it at the same time
644 unlockClosure((StgClosure*)m,&stg_MSG_NULL_info);
645 break;
646 }
647
648 #if !defined(THREADED_RTS)
649 case BlockedOnRead:
650 case BlockedOnWrite:
651 #if defined(mingw32_HOST_OS)
652 case BlockedOnDoProc:
653 #endif
654 removeThreadFromDeQueue(cap, &blocked_queue_hd, &blocked_queue_tl, tso);
655 #if defined(mingw32_HOST_OS)
656 /* (Cooperatively) signal that the worker thread should abort
657 * the request.
658 */
659 abandonWorkRequest(tso->block_info.async_result->reqID);
660 #endif
661 goto done;
662
663 case BlockedOnDelay:
664 removeThreadFromQueue(cap, &sleeping_queue, tso);
665 goto done;
666 #endif
667
668 default:
669 barf("removeFromQueues: %d", tso->why_blocked);
670 }
671
672 done:
673 tso->why_blocked = NotBlocked;
674 appendToRunQueue(cap, tso);
675 }
676
677 /* -----------------------------------------------------------------------------
678 * raiseAsync()
679 *
680 * The following function implements the magic for raising an
681 * asynchronous exception in an existing thread.
682 *
683 * We first remove the thread from any queue on which it might be
684 * blocked. The possible blockages are MVARs, BLOCKING_QUEUESs, and
685 * TSO blocked_exception queues.
686 *
687 * We strip the stack down to the innermost CATCH_FRAME, building
688 * thunks in the heap for all the active computations, so they can
689 * be restarted if necessary. When we reach a CATCH_FRAME, we build
690 * an application of the handler to the exception, and push it on
691 * the top of the stack.
692 *
693 * How exactly do we save all the active computations? We create an
694 * AP_STACK for every UpdateFrame on the stack. Entering one of these
695 * AP_STACKs pushes everything from the corresponding update frame
696 * upwards onto the stack. (Actually, it pushes everything up to the
697 * next update frame plus a pointer to the next AP_STACK object.
698 * Entering the next AP_STACK object pushes more onto the stack until we
699 * reach the last AP_STACK object - at which point the stack should look
700 * exactly as it did when we killed the TSO and we can continue
701 * execution by entering the closure on top of the stack.
702 *
703 * We can also kill a thread entirely - this happens if either (a) the
704 * exception passed to raiseAsync is NULL, or (b) there's no
705 * CATCH_FRAME on the stack. In either case, we strip the entire
706 * stack and replace the thread with a zombie.
707 *
708 * ToDo: in THREADED_RTS mode, this function is only safe if either
709 * (a) we hold all the Capabilities (eg. in GC, or if there is only
710 * one Capability), or (b) we own the Capability that the TSO is
711 * currently blocked on or on the run queue of.
712 *
713 * -------------------------------------------------------------------------- */
714
715 static void
716 raiseAsync(Capability *cap, StgTSO *tso, StgClosure *exception,
717 rtsBool stop_at_atomically, StgUpdateFrame *stop_here)
718 {
719 StgRetInfoTable *info;
720 StgPtr sp, frame;
721 StgClosure *updatee;
722 nat i;
723
724 debugTraceCap(DEBUG_sched, cap,
725 "raising exception in thread %ld.", (long)tso->id);
726
727 #if defined(PROFILING)
728 /*
729 * Debugging tool: on raising an exception, show where we are.
730 * See also Exception.cmm:stg_raisezh.
731 * This wasn't done for asynchronous exceptions originally; see #1450
732 */
733 if (RtsFlags.ProfFlags.showCCSOnException)
734 {
735 fprintCCS_stderr(tso->prof.CCCS);
736 }
737 #endif
738 // ASSUMES: the thread is not already complete or dead, or
739 // ThreadRelocated. Upper layers should deal with that.
740 ASSERT(tso->what_next != ThreadComplete &&
741 tso->what_next != ThreadKilled &&
742 tso->what_next != ThreadRelocated);
743
744 // only if we own this TSO (except that deleteThread() calls this
745 ASSERT(tso->cap == cap);
746
747 // wake it up
748 if (tso->why_blocked != NotBlocked) {
749 tso->why_blocked = NotBlocked;
750 appendToRunQueue(cap,tso);
751 }
752
753 // mark it dirty; we're about to change its stack.
754 dirty_TSO(cap, tso);
755
756 sp = tso->sp;
757
758 if (stop_here != NULL) {
759 updatee = stop_here->updatee;
760 } else {
761 updatee = NULL;
762 }
763
764 // The stack freezing code assumes there's a closure pointer on
765 // the top of the stack, so we have to arrange that this is the case...
766 //
767 if (sp[0] == (W_)&stg_enter_info) {
768 sp++;
769 } else {
770 sp--;
771 sp[0] = (W_)&stg_dummy_ret_closure;
772 }
773
774 frame = sp + 1;
775 while (stop_here == NULL || frame < (StgPtr)stop_here) {
776
777 // 1. Let the top of the stack be the "current closure"
778 //
779 // 2. Walk up the stack until we find either an UPDATE_FRAME or a
780 // CATCH_FRAME.
781 //
782 // 3. If it's an UPDATE_FRAME, then make an AP_STACK containing the
783 // current closure applied to the chunk of stack up to (but not
784 // including) the update frame. This closure becomes the "current
785 // closure". Go back to step 2.
786 //
787 // 4. If it's a CATCH_FRAME, then leave the exception handler on
788 // top of the stack applied to the exception.
789 //
790 // 5. If it's a STOP_FRAME, then kill the thread.
791 //
792 // NB: if we pass an ATOMICALLY_FRAME then abort the associated
793 // transaction
794
795 info = get_ret_itbl((StgClosure *)frame);
796
797 switch (info->i.type) {
798
799 case UPDATE_FRAME:
800 {
801 StgAP_STACK * ap;
802 nat words;
803
804 // First build an AP_STACK consisting of the stack chunk above the
805 // current update frame, with the top word on the stack as the
806 // fun field.
807 //
808 words = frame - sp - 1;
809 ap = (StgAP_STACK *)allocate(cap,AP_STACK_sizeW(words));
810
811 ap->size = words;
812 ap->fun = (StgClosure *)sp[0];
813 sp++;
814 for(i=0; i < (nat)words; ++i) {
815 ap->payload[i] = (StgClosure *)*sp++;
816 }
817
818 SET_HDR(ap,&stg_AP_STACK_info,
819 ((StgClosure *)frame)->header.prof.ccs /* ToDo */);
820 TICK_ALLOC_UP_THK(words+1,0);
821
822 //IF_DEBUG(scheduler,
823 // debugBelch("sched: Updating ");
824 // printPtr((P_)((StgUpdateFrame *)frame)->updatee);
825 // debugBelch(" with ");
826 // printObj((StgClosure *)ap);
827 // );
828
829 if (((StgUpdateFrame *)frame)->updatee == updatee) {
830 // If this update frame points to the same closure as
831 // the update frame further down the stack
832 // (stop_here), then don't perform the update. We
833 // want to keep the blackhole in this case, so we can
834 // detect and report the loop (#2783).
835 ap = (StgAP_STACK*)updatee;
836 } else {
837 // Perform the update
838 // TODO: this may waste some work, if the thunk has
839 // already been updated by another thread.
840 updateThunk(cap, tso,
841 ((StgUpdateFrame *)frame)->updatee, (StgClosure *)ap);
842 }
843
844 sp += sizeofW(StgUpdateFrame) - 1;
845 sp[0] = (W_)ap; // push onto stack
846 frame = sp + 1;
847 continue; //no need to bump frame
848 }
849
850 case STOP_FRAME:
851 {
852 // We've stripped the entire stack, the thread is now dead.
853 tso->what_next = ThreadKilled;
854 tso->sp = frame + sizeofW(StgStopFrame);
855 return;
856 }
857
858 case CATCH_FRAME:
859 // If we find a CATCH_FRAME, and we've got an exception to raise,
860 // then build the THUNK raise(exception), and leave it on
861 // top of the CATCH_FRAME ready to enter.
862 //
863 {
864 StgCatchFrame *cf = (StgCatchFrame *)frame;
865 StgThunk *raise;
866
867 if (exception == NULL) break;
868
869 // we've got an exception to raise, so let's pass it to the
870 // handler in this frame.
871 //
872 raise = (StgThunk *)allocate(cap,sizeofW(StgThunk)+1);
873 TICK_ALLOC_SE_THK(1,0);
874 SET_HDR(raise,&stg_raise_info,cf->header.prof.ccs);
875 raise->payload[0] = exception;
876
877 // throw away the stack from Sp up to the CATCH_FRAME.
878 //
879 sp = frame - 1;
880
881 /* Ensure that async excpetions are blocked now, so we don't get
882 * a surprise exception before we get around to executing the
883 * handler.
884 */
885 tso->flags |= TSO_BLOCKEX;
886 if ((cf->exceptions_blocked & TSO_INTERRUPTIBLE) == 0) {
887 tso->flags &= ~TSO_INTERRUPTIBLE;
888 } else {
889 tso->flags |= TSO_INTERRUPTIBLE;
890 }
891
892 /* Put the newly-built THUNK on top of the stack, ready to execute
893 * when the thread restarts.
894 */
895 sp[0] = (W_)raise;
896 sp[-1] = (W_)&stg_enter_info;
897 tso->sp = sp-1;
898 tso->what_next = ThreadRunGHC;
899 IF_DEBUG(sanity, checkTSO(tso));
900 return;
901 }
902
903 case ATOMICALLY_FRAME:
904 if (stop_at_atomically) {
905 ASSERT(tso->trec->enclosing_trec == NO_TREC);
906 stmCondemnTransaction(cap, tso -> trec);
907 tso->sp = frame - 2;
908 // The ATOMICALLY_FRAME expects to be returned a
909 // result from the transaction, which it stores in the
910 // stack frame. Hence we arrange to return a dummy
911 // result, so that the GC doesn't get upset (#3578).
912 // Perhaps a better way would be to have a different
913 // ATOMICALLY_FRAME instance for condemned
914 // transactions, but I don't fully understand the
915 // interaction with STM invariants.
916 tso->sp[1] = (W_)&stg_NO_TREC_closure;
917 tso->sp[0] = (W_)&stg_gc_unpt_r1_info;
918 tso->what_next = ThreadRunGHC;
919 return;
920 }
921 // Not stop_at_atomically... fall through and abort the
922 // transaction.
923
924 case CATCH_STM_FRAME:
925 case CATCH_RETRY_FRAME:
926 // IF we find an ATOMICALLY_FRAME then we abort the
927 // current transaction and propagate the exception. In
928 // this case (unlike ordinary exceptions) we do not care
929 // whether the transaction is valid or not because its
930 // possible validity cannot have caused the exception
931 // and will not be visible after the abort.
932
933 {
934 StgTRecHeader *trec = tso -> trec;
935 StgTRecHeader *outer = trec -> enclosing_trec;
936 debugTraceCap(DEBUG_stm, cap,
937 "found atomically block delivering async exception");
938 stmAbortTransaction(cap, trec);
939 stmFreeAbortedTRec(cap, trec);
940 tso -> trec = outer;
941 break;
942 };
943
944 default:
945 break;
946 }
947
948 // move on to the next stack frame
949 frame += stack_frame_sizeW((StgClosure *)frame);
950 }
951
952 // if we got here, then we stopped at stop_here
953 ASSERT(stop_here != NULL);
954 }
955
956