1 /* -----------------------------------------------------------------------------
3 * (c) The GHC Team, 1998-2004
5 * Out-of-line primitive operations
7 * This file contains the implementations of all the primitive
8 * operations ("primops") which are not expanded inline. See
9 * ghc/compiler/prelude/primops.txt.pp for a list of all the primops;
10 * this file contains code for most of those with the attribute
13 * Entry convention: the entry convention for a primop is that all the
14 * args are in Stg registers (R1, R2, etc.). This is to make writing
15 * the primops easier. (see compiler/codeGen/CgCallConv.hs).
17 * Return convention: results from a primop are generally returned
18 * using the ordinary unboxed tuple return convention. The C-- parser
19 * implements the RET_xxxx() macros to perform unboxed-tuple returns
20 * based on the prevailing return convention.
22 * This file is written in a subset of C--, extended with various
23 * features specific to GHC. It is compiled by GHC directly. For the
24 * syntax of .cmm files, see the parser in ghc/compiler/cmm/CmmParse.y.
26 * ---------------------------------------------------------------------------*/
31 import pthread_mutex_lock;
32 import pthread_mutex_unlock;
34 import base_ControlziExceptionziBase_nestedAtomically_closure;
35 import EnterCriticalSection;
36 import LeaveCriticalSection;
37 import ghczmprim_GHCziBool_False_closure;
39 /*-----------------------------------------------------------------------------
42 Basically just new*Array - the others are all inline macros.
44 The size arg is always passed in R1, and the result returned in R1.
46 The slow entry point is for returning from a heap check, the saved
47 size argument must be re-loaded from the stack.
48 -------------------------------------------------------------------------- */
50 /* for objects that are *less* than the size of a word, make sure we
51 * round up to the nearest word for the size of the array.
56 W_ words, payload_words, n, p;
57 MAYBE_GC(NO_PTRS,stg_newByteArrayzh);
59 payload_words = ROUNDUP_BYTES_TO_WDS(n);
60 words = BYTES_TO_WDS(SIZEOF_StgArrWords) + payload_words;
61 ("ptr" p) = foreign "C" allocate(MyCapability() "ptr",words) [];
62 TICK_ALLOC_PRIM(SIZEOF_StgArrWords,WDS(payload_words),0);
63 SET_HDR(p, stg_ARR_WORDS_info, W_[CCCS]);
64 StgArrWords_words(p) = payload_words;
69 #define BA_MASK (BA_ALIGN-1)
71 stg_newPinnedByteArrayzh
73 W_ words, bytes, payload_words, p;
75 MAYBE_GC(NO_PTRS,stg_newPinnedByteArrayzh);
77 /* payload_words is what we will tell the profiler we had to allocate */
78 payload_words = ROUNDUP_BYTES_TO_WDS(bytes);
79 /* When we actually allocate memory, we need to allow space for the
81 bytes = bytes + SIZEOF_StgArrWords;
82 /* And we want to align to BA_ALIGN bytes, so we need to allow space
83 to shift up to BA_ALIGN - 1 bytes: */
84 bytes = bytes + BA_ALIGN - 1;
85 /* Now we convert to a number of words: */
86 words = ROUNDUP_BYTES_TO_WDS(bytes);
88 ("ptr" p) = foreign "C" allocatePinned(MyCapability() "ptr", words) [];
89 TICK_ALLOC_PRIM(SIZEOF_StgArrWords,WDS(payload_words),0);
91 /* Now we need to move p forward so that the payload is aligned
93 p = p + ((-p - SIZEOF_StgArrWords) & BA_MASK);
95 SET_HDR(p, stg_ARR_WORDS_info, W_[CCCS]);
96 StgArrWords_words(p) = payload_words;
100 stg_newAlignedPinnedByteArrayzh
102 W_ words, bytes, payload_words, p, alignment;
104 MAYBE_GC(NO_PTRS,stg_newAlignedPinnedByteArrayzh);
108 /* payload_words is what we will tell the profiler we had to allocate */
109 payload_words = ROUNDUP_BYTES_TO_WDS(bytes);
111 /* When we actually allocate memory, we need to allow space for the
113 bytes = bytes + SIZEOF_StgArrWords;
114 /* And we want to align to <alignment> bytes, so we need to allow space
115 to shift up to <alignment - 1> bytes: */
116 bytes = bytes + alignment - 1;
117 /* Now we convert to a number of words: */
118 words = ROUNDUP_BYTES_TO_WDS(bytes);
120 ("ptr" p) = foreign "C" allocatePinned(MyCapability() "ptr", words) [];
121 TICK_ALLOC_PRIM(SIZEOF_StgArrWords,WDS(payload_words),0);
123 /* Now we need to move p forward so that the payload is aligned
124 to <alignment> bytes. Note that we are assuming that
125 <alignment> is a power of 2, which is technically not guaranteed */
126 p = p + ((-p - SIZEOF_StgArrWords) & (alignment - 1));
128 SET_HDR(p, stg_ARR_WORDS_info, W_[CCCS]);
129 StgArrWords_words(p) = payload_words;
135 W_ words, n, init, arr, p;
136 /* Args: R1 = words, R2 = initialisation value */
139 MAYBE_GC(R2_PTR,stg_newArrayzh);
141 words = BYTES_TO_WDS(SIZEOF_StgMutArrPtrs) + n;
142 ("ptr" arr) = foreign "C" allocate(MyCapability() "ptr",words) [R2];
143 TICK_ALLOC_PRIM(SIZEOF_StgMutArrPtrs, WDS(n), 0);
145 SET_HDR(arr, stg_MUT_ARR_PTRS_DIRTY_info, W_[CCCS]);
146 StgMutArrPtrs_ptrs(arr) = n;
148 // Initialise all elements of the the array with the value in R2
150 p = arr + SIZEOF_StgMutArrPtrs;
152 if (p < arr + WDS(words)) {
161 stg_unsafeThawArrayzh
163 // SUBTLETY TO DO WITH THE OLD GEN MUTABLE LIST
165 // A MUT_ARR_PTRS lives on the mutable list, but a MUT_ARR_PTRS_FROZEN
166 // normally doesn't. However, when we freeze a MUT_ARR_PTRS, we leave
167 // it on the mutable list for the GC to remove (removing something from
168 // the mutable list is not easy, because the mut_list is only singly-linked).
170 // So that we can tell whether a MUT_ARR_PTRS_FROZEN is on the mutable list,
171 // when we freeze it we set the info ptr to be MUT_ARR_PTRS_FROZEN0
172 // to indicate that it is still on the mutable list.
174 // So, when we thaw a MUT_ARR_PTRS_FROZEN, we must cope with two cases:
175 // either it is on a mut_list, or it isn't. We adopt the convention that
176 // the closure type is MUT_ARR_PTRS_FROZEN0 if it is on the mutable list,
177 // and MUT_ARR_PTRS_FROZEN otherwise. In fact it wouldn't matter if
178 // we put it on the mutable list more than once, but it would get scavenged
179 // multiple times during GC, which would be unnecessarily slow.
181 if (StgHeader_info(R1) != stg_MUT_ARR_PTRS_FROZEN0_info) {
182 SET_INFO(R1,stg_MUT_ARR_PTRS_DIRTY_info);
183 recordMutable(R1, R1);
184 // must be done after SET_INFO, because it ASSERTs closure_MUTABLE()
187 SET_INFO(R1,stg_MUT_ARR_PTRS_DIRTY_info);
192 /* -----------------------------------------------------------------------------
194 -------------------------------------------------------------------------- */
199 /* Args: R1 = initialisation value */
201 ALLOC_PRIM( SIZEOF_StgMutVar, R1_PTR, stg_newMutVarzh);
203 mv = Hp - SIZEOF_StgMutVar + WDS(1);
204 SET_HDR(mv,stg_MUT_VAR_DIRTY_info,W_[CCCS]);
205 StgMutVar_var(mv) = R1;
210 stg_atomicModifyMutVarzh
212 W_ mv, f, z, x, y, r, h;
213 /* Args: R1 :: MutVar#, R2 :: a -> (a,b) */
215 /* If x is the current contents of the MutVar#, then
216 We want to make the new contents point to
220 and the return value is
224 obviously we can share (f x).
226 z = [stg_ap_2 f x] (max (HS + 2) MIN_UPD_SIZE)
227 y = [stg_sel_0 z] (max (HS + 1) MIN_UPD_SIZE)
228 r = [stg_sel_1 z] (max (HS + 1) MIN_UPD_SIZE)
232 #define THUNK_1_SIZE (SIZEOF_StgThunkHeader + WDS(MIN_UPD_SIZE))
233 #define TICK_ALLOC_THUNK_1() TICK_ALLOC_UP_THK(WDS(1),WDS(MIN_UPD_SIZE-1))
235 #define THUNK_1_SIZE (SIZEOF_StgThunkHeader + WDS(1))
236 #define TICK_ALLOC_THUNK_1() TICK_ALLOC_UP_THK(WDS(1),0)
240 #define THUNK_2_SIZE (SIZEOF_StgThunkHeader + WDS(MIN_UPD_SIZE))
241 #define TICK_ALLOC_THUNK_2() TICK_ALLOC_UP_THK(WDS(2),WDS(MIN_UPD_SIZE-2))
243 #define THUNK_2_SIZE (SIZEOF_StgThunkHeader + WDS(2))
244 #define TICK_ALLOC_THUNK_2() TICK_ALLOC_UP_THK(WDS(2),0)
247 #define SIZE (THUNK_2_SIZE + THUNK_1_SIZE + THUNK_1_SIZE)
249 HP_CHK_GEN_TICKY(SIZE, R1_PTR & R2_PTR, stg_atomicModifyMutVarzh);
254 TICK_ALLOC_THUNK_2();
255 CCCS_ALLOC(THUNK_2_SIZE);
256 z = Hp - THUNK_2_SIZE + WDS(1);
257 SET_HDR(z, stg_ap_2_upd_info, W_[CCCS]);
258 LDV_RECORD_CREATE(z);
259 StgThunk_payload(z,0) = f;
261 TICK_ALLOC_THUNK_1();
262 CCCS_ALLOC(THUNK_1_SIZE);
263 y = z - THUNK_1_SIZE;
264 SET_HDR(y, stg_sel_0_upd_info, W_[CCCS]);
265 LDV_RECORD_CREATE(y);
266 StgThunk_payload(y,0) = z;
268 TICK_ALLOC_THUNK_1();
269 CCCS_ALLOC(THUNK_1_SIZE);
270 r = y - THUNK_1_SIZE;
271 SET_HDR(r, stg_sel_1_upd_info, W_[CCCS]);
272 LDV_RECORD_CREATE(r);
273 StgThunk_payload(r,0) = z;
276 x = StgMutVar_var(mv);
277 StgThunk_payload(z,1) = x;
279 (h) = foreign "C" cas(mv + SIZEOF_StgHeader + OFFSET_StgMutVar_var, x, y) [];
280 if (h != x) { goto retry; }
282 StgMutVar_var(mv) = y;
285 if (GET_INFO(mv) == stg_MUT_VAR_CLEAN_info) {
286 foreign "C" dirty_MUT_VAR(BaseReg "ptr", mv "ptr") [];
292 /* -----------------------------------------------------------------------------
293 Weak Pointer Primitives
294 -------------------------------------------------------------------------- */
296 STRING(stg_weak_msg,"New weak pointer at %p\n")
302 R3 = finalizer (or NULL)
307 R3 = stg_NO_FINALIZER_closure;
310 ALLOC_PRIM( SIZEOF_StgWeak, R1_PTR & R2_PTR & R3_PTR, stg_mkWeakzh );
312 w = Hp - SIZEOF_StgWeak + WDS(1);
313 SET_HDR(w, stg_WEAK_info, W_[CCCS]);
315 // We don't care about cfinalizer here.
316 // Should StgWeak_cfinalizer(w) be stg_NO_FINALIZER_closure or
320 StgWeak_value(w) = R2;
321 StgWeak_finalizer(w) = R3;
322 StgWeak_cfinalizer(w) = stg_NO_FINALIZER_closure;
324 ACQUIRE_LOCK(sm_mutex);
325 StgWeak_link(w) = W_[weak_ptr_list];
326 W_[weak_ptr_list] = w;
327 RELEASE_LOCK(sm_mutex);
329 IF_DEBUG(weak, foreign "C" debugBelch(stg_weak_msg,w) []);
334 stg_mkWeakForeignEnvzh
340 R5 = has environment (0 or 1)
343 W_ w, payload_words, words, p;
345 W_ key, val, fptr, ptr, flag, eptr;
354 ALLOC_PRIM( SIZEOF_StgWeak, R1_PTR & R2_PTR, stg_mkWeakForeignEnvzh );
356 w = Hp - SIZEOF_StgWeak + WDS(1);
357 SET_HDR(w, stg_WEAK_info, W_[CCCS]);
360 words = BYTES_TO_WDS(SIZEOF_StgArrWords) + payload_words;
361 ("ptr" p) = foreign "C" allocate(MyCapability() "ptr", words) [];
363 TICK_ALLOC_PRIM(SIZEOF_StgArrWords,WDS(payload_words),0);
364 SET_HDR(p, stg_ARR_WORDS_info, W_[CCCS]);
366 StgArrWords_words(p) = payload_words;
367 StgArrWords_payload(p,0) = fptr;
368 StgArrWords_payload(p,1) = ptr;
369 StgArrWords_payload(p,2) = eptr;
370 StgArrWords_payload(p,3) = flag;
372 // We don't care about the value here.
373 // Should StgWeak_value(w) be stg_NO_FINALIZER_closure or something else?
375 StgWeak_key(w) = key;
376 StgWeak_value(w) = val;
377 StgWeak_finalizer(w) = stg_NO_FINALIZER_closure;
378 StgWeak_cfinalizer(w) = p;
380 ACQUIRE_LOCK(sm_mutex);
381 StgWeak_link(w) = W_[weak_ptr_list];
382 W_[weak_ptr_list] = w;
383 RELEASE_LOCK(sm_mutex);
385 IF_DEBUG(weak, foreign "C" debugBelch(stg_weak_msg,w) []);
399 if (GET_INFO(w) == stg_DEAD_WEAK_info) {
400 RET_NP(0,stg_NO_FINALIZER_closure);
406 // A weak pointer is inherently used, so we do not need to call
407 // LDV_recordDead_FILL_SLOP_DYNAMIC():
408 // LDV_recordDead_FILL_SLOP_DYNAMIC((StgClosure *)w);
409 // or, LDV_recordDead():
410 // LDV_recordDead((StgClosure *)w, sizeofW(StgWeak) - sizeofW(StgProfHeader));
411 // Furthermore, when PROFILING is turned on, dead weak pointers are exactly as
412 // large as weak pointers, so there is no need to fill the slop, either.
413 // See stg_DEAD_WEAK_info in StgMiscClosures.hc.
417 // Todo: maybe use SET_HDR() and remove LDV_recordCreate()?
419 SET_INFO(w,stg_DEAD_WEAK_info);
420 LDV_RECORD_CREATE(w);
422 f = StgWeak_finalizer(w);
423 arr = StgWeak_cfinalizer(w);
425 StgDeadWeak_link(w) = StgWeak_link(w);
427 if (arr != stg_NO_FINALIZER_closure) {
428 foreign "C" runCFinalizer(StgArrWords_payload(arr,0),
429 StgArrWords_payload(arr,1),
430 StgArrWords_payload(arr,2),
431 StgArrWords_payload(arr,3)) [];
434 /* return the finalizer */
435 if (f == stg_NO_FINALIZER_closure) {
436 RET_NP(0,stg_NO_FINALIZER_closure);
448 if (GET_INFO(w) == stg_WEAK_info) {
450 val = StgWeak_value(w);
458 /* -----------------------------------------------------------------------------
459 Floating point operations.
460 -------------------------------------------------------------------------- */
462 stg_decodeFloatzuIntzh
469 STK_CHK_GEN( WDS(2), NO_PTRS, stg_decodeFloatzuIntzh );
471 mp_tmp1 = Sp - WDS(1);
472 mp_tmp_w = Sp - WDS(2);
474 /* arguments: F1 = Float# */
477 /* Perform the operation */
478 foreign "C" __decodeFloat_Int(mp_tmp1 "ptr", mp_tmp_w "ptr", arg) [];
480 /* returns: (Int# (mantissa), Int# (exponent)) */
481 RET_NN(W_[mp_tmp1], W_[mp_tmp_w]);
484 stg_decodeDoublezu2Intzh
493 STK_CHK_GEN( WDS(4), NO_PTRS, stg_decodeDoublezu2Intzh );
495 mp_tmp1 = Sp - WDS(1);
496 mp_tmp2 = Sp - WDS(2);
497 mp_result1 = Sp - WDS(3);
498 mp_result2 = Sp - WDS(4);
500 /* arguments: D1 = Double# */
503 /* Perform the operation */
504 foreign "C" __decodeDouble_2Int(mp_tmp1 "ptr", mp_tmp2 "ptr",
505 mp_result1 "ptr", mp_result2 "ptr",
509 (Int# (mant sign), Word# (mant high), Word# (mant low), Int# (expn)) */
510 RET_NNNN(W_[mp_tmp1], W_[mp_tmp2], W_[mp_result1], W_[mp_result2]);
513 /* -----------------------------------------------------------------------------
514 * Concurrency primitives
515 * -------------------------------------------------------------------------- */
519 /* args: R1 = closure to spark */
521 MAYBE_GC(R1_PTR, stg_forkzh);
527 ("ptr" threadid) = foreign "C" createIOThread( MyCapability() "ptr",
528 RtsFlags_GcFlags_initialStkSize(RtsFlags),
531 /* start blocked if the current thread is blocked */
532 StgTSO_flags(threadid) =
533 StgTSO_flags(threadid) | (StgTSO_flags(CurrentTSO) &
534 (TSO_BLOCKEX::I32 | TSO_INTERRUPTIBLE::I32));
536 foreign "C" scheduleThread(MyCapability() "ptr", threadid "ptr") [];
538 // context switch soon, but not immediately: we don't want every
539 // forkIO to force a context-switch.
540 Capability_context_switch(MyCapability()) = 1 :: CInt;
547 /* args: R1 = cpu, R2 = closure to spark */
549 MAYBE_GC(R2_PTR, stg_forkOnzh);
557 ("ptr" threadid) = foreign "C" createIOThread( MyCapability() "ptr",
558 RtsFlags_GcFlags_initialStkSize(RtsFlags),
561 /* start blocked if the current thread is blocked */
562 StgTSO_flags(threadid) =
563 StgTSO_flags(threadid) | (StgTSO_flags(CurrentTSO) &
564 (TSO_BLOCKEX::I32 | TSO_INTERRUPTIBLE::I32));
566 foreign "C" scheduleThreadOn(MyCapability() "ptr", cpu, threadid "ptr") [];
568 // context switch soon, but not immediately: we don't want every
569 // forkIO to force a context-switch.
570 Capability_context_switch(MyCapability()) = 1 :: CInt;
577 jump stg_yield_noregs;
592 foreign "C" labelThread(R1 "ptr", R2 "ptr") [];
594 jump %ENTRY_CODE(Sp(0));
597 stg_isCurrentThreadBoundzh
601 (r) = foreign "C" isThreadBound(CurrentTSO) [];
607 /* args: R1 :: ThreadId# */
615 if (TO_W_(StgTSO_what_next(tso)) == ThreadRelocated) {
616 tso = StgTSO__link(tso);
620 what_next = TO_W_(StgTSO_what_next(tso));
621 why_blocked = TO_W_(StgTSO_why_blocked(tso));
622 // Note: these two reads are not atomic, so they might end up
623 // being inconsistent. It doesn't matter, since we
624 // only return one or the other. If we wanted to return the
625 // contents of block_info too, then we'd have to do some synchronisation.
627 if (what_next == ThreadComplete) {
628 ret = 16; // NB. magic, matches up with GHC.Conc.threadStatus
630 if (what_next == ThreadKilled) {
639 /* -----------------------------------------------------------------------------
641 * -------------------------------------------------------------------------- */
645 // Catch retry frame ------------------------------------------------------------
647 INFO_TABLE_RET(stg_catch_retry_frame, CATCH_RETRY_FRAME,
648 #if defined(PROFILING)
649 W_ unused1, W_ unused2,
651 W_ unused3, P_ unused4, P_ unused5)
653 W_ r, frame, trec, outer;
656 trec = StgTSO_trec(CurrentTSO);
657 outer = StgTRecHeader_enclosing_trec(trec);
658 (r) = foreign "C" stmCommitNestedTransaction(MyCapability() "ptr", trec "ptr") [];
660 /* Succeeded (either first branch or second branch) */
661 StgTSO_trec(CurrentTSO) = outer;
662 Sp = Sp + SIZEOF_StgCatchRetryFrame;
663 jump %ENTRY_CODE(Sp(SP_OFF));
665 /* Did not commit: re-execute */
667 ("ptr" new_trec) = foreign "C" stmStartTransaction(MyCapability() "ptr", outer "ptr") [];
668 StgTSO_trec(CurrentTSO) = new_trec;
669 if (StgCatchRetryFrame_running_alt_code(frame) != 0::I32) {
670 R1 = StgCatchRetryFrame_alt_code(frame);
672 R1 = StgCatchRetryFrame_first_code(frame);
679 // Atomically frame ------------------------------------------------------------
681 INFO_TABLE_RET(stg_atomically_frame, ATOMICALLY_FRAME,
682 #if defined(PROFILING)
683 W_ unused1, W_ unused2,
685 P_ code, P_ next_invariant_to_check, P_ result)
687 W_ frame, trec, valid, next_invariant, q, outer;
690 trec = StgTSO_trec(CurrentTSO);
692 outer = StgTRecHeader_enclosing_trec(trec);
694 if (outer == NO_TREC) {
695 /* First time back at the atomically frame -- pick up invariants */
696 ("ptr" q) = foreign "C" stmGetInvariantsToCheck(MyCapability() "ptr", trec "ptr") [];
697 StgAtomicallyFrame_next_invariant_to_check(frame) = q;
698 StgAtomicallyFrame_result(frame) = result;
701 /* Second/subsequent time back at the atomically frame -- abort the
702 * tx that's checking the invariant and move on to the next one */
703 StgTSO_trec(CurrentTSO) = outer;
704 q = StgAtomicallyFrame_next_invariant_to_check(frame);
705 StgInvariantCheckQueue_my_execution(q) = trec;
706 foreign "C" stmAbortTransaction(MyCapability() "ptr", trec "ptr") [];
707 /* Don't free trec -- it's linked from q and will be stashed in the
708 * invariant if we eventually commit. */
709 q = StgInvariantCheckQueue_next_queue_entry(q);
710 StgAtomicallyFrame_next_invariant_to_check(frame) = q;
714 q = StgAtomicallyFrame_next_invariant_to_check(frame);
716 if (q != END_INVARIANT_CHECK_QUEUE) {
717 /* We can't commit yet: another invariant to check */
718 ("ptr" trec) = foreign "C" stmStartTransaction(MyCapability() "ptr", trec "ptr") [];
719 StgTSO_trec(CurrentTSO) = trec;
721 next_invariant = StgInvariantCheckQueue_invariant(q);
722 R1 = StgAtomicInvariant_code(next_invariant);
727 /* We've got no more invariants to check, try to commit */
728 (valid) = foreign "C" stmCommitTransaction(MyCapability() "ptr", trec "ptr") [];
730 /* Transaction was valid: commit succeeded */
731 StgTSO_trec(CurrentTSO) = NO_TREC;
732 R1 = StgAtomicallyFrame_result(frame);
733 Sp = Sp + SIZEOF_StgAtomicallyFrame;
734 jump %ENTRY_CODE(Sp(SP_OFF));
736 /* Transaction was not valid: try again */
737 ("ptr" trec) = foreign "C" stmStartTransaction(MyCapability() "ptr", NO_TREC "ptr") [];
738 StgTSO_trec(CurrentTSO) = trec;
739 StgAtomicallyFrame_next_invariant_to_check(frame) = END_INVARIANT_CHECK_QUEUE;
740 R1 = StgAtomicallyFrame_code(frame);
746 INFO_TABLE_RET(stg_atomically_waiting_frame, ATOMICALLY_FRAME,
747 #if defined(PROFILING)
748 W_ unused1, W_ unused2,
750 P_ code, P_ next_invariant_to_check, P_ result)
752 W_ frame, trec, valid;
756 /* The TSO is currently waiting: should we stop waiting? */
757 (valid) = foreign "C" stmReWait(MyCapability() "ptr", CurrentTSO "ptr") [];
759 /* Previous attempt is still valid: no point trying again yet */
760 jump stg_block_noregs;
762 /* Previous attempt is no longer valid: try again */
763 ("ptr" trec) = foreign "C" stmStartTransaction(MyCapability() "ptr", NO_TREC "ptr") [];
764 StgTSO_trec(CurrentTSO) = trec;
765 StgHeader_info(frame) = stg_atomically_frame_info;
766 R1 = StgAtomicallyFrame_code(frame);
771 // STM catch frame --------------------------------------------------------------
775 /* Catch frames are very similar to update frames, but when entering
776 * one we just pop the frame off the stack and perform the correct
777 * kind of return to the activation record underneath us on the stack.
780 INFO_TABLE_RET(stg_catch_stm_frame, CATCH_STM_FRAME,
781 #if defined(PROFILING)
782 W_ unused1, W_ unused2,
784 P_ unused3, P_ unused4)
786 W_ r, frame, trec, outer;
788 trec = StgTSO_trec(CurrentTSO);
789 outer = StgTRecHeader_enclosing_trec(trec);
790 (r) = foreign "C" stmCommitNestedTransaction(MyCapability() "ptr", trec "ptr") [];
792 /* Commit succeeded */
793 StgTSO_trec(CurrentTSO) = outer;
794 Sp = Sp + SIZEOF_StgCatchSTMFrame;
799 ("ptr" new_trec) = foreign "C" stmStartTransaction(MyCapability() "ptr", outer "ptr") [];
800 StgTSO_trec(CurrentTSO) = new_trec;
801 R1 = StgCatchSTMFrame_code(frame);
807 // Primop definition ------------------------------------------------------------
815 // stmStartTransaction may allocate
816 MAYBE_GC (R1_PTR, stg_atomicallyzh);
818 /* Args: R1 = m :: STM a */
819 STK_CHK_GEN(SIZEOF_StgAtomicallyFrame + WDS(1), R1_PTR, stg_atomicallyzh);
821 old_trec = StgTSO_trec(CurrentTSO);
823 /* Nested transactions are not allowed; raise an exception */
824 if (old_trec != NO_TREC) {
825 R1 = base_ControlziExceptionziBase_nestedAtomically_closure;
829 /* Set up the atomically frame */
830 Sp = Sp - SIZEOF_StgAtomicallyFrame;
833 SET_HDR(frame,stg_atomically_frame_info, W_[CCCS]);
834 StgAtomicallyFrame_code(frame) = R1;
835 StgAtomicallyFrame_result(frame) = NO_TREC;
836 StgAtomicallyFrame_next_invariant_to_check(frame) = END_INVARIANT_CHECK_QUEUE;
838 /* Start the memory transcation */
839 ("ptr" new_trec) = foreign "C" stmStartTransaction(MyCapability() "ptr", old_trec "ptr") [R1];
840 StgTSO_trec(CurrentTSO) = new_trec;
842 /* Apply R1 to the realworld token */
851 /* Args: R1 :: STM a */
852 /* Args: R2 :: Exception -> STM a */
853 STK_CHK_GEN(SIZEOF_StgCatchSTMFrame + WDS(1), R1_PTR & R2_PTR, stg_catchSTMzh);
855 /* Set up the catch frame */
856 Sp = Sp - SIZEOF_StgCatchSTMFrame;
859 SET_HDR(frame, stg_catch_stm_frame_info, W_[CCCS]);
860 StgCatchSTMFrame_handler(frame) = R2;
861 StgCatchSTMFrame_code(frame) = R1;
863 /* Start a nested transaction to run the body of the try block in */
866 cur_trec = StgTSO_trec(CurrentTSO);
867 ("ptr" new_trec) = foreign "C" stmStartTransaction(MyCapability() "ptr", cur_trec "ptr");
868 StgTSO_trec(CurrentTSO) = new_trec;
870 /* Apply R1 to the realworld token */
881 // stmStartTransaction may allocate
882 MAYBE_GC (R1_PTR & R2_PTR, stg_catchRetryzh);
884 /* Args: R1 :: STM a */
885 /* Args: R2 :: STM a */
886 STK_CHK_GEN(SIZEOF_StgCatchRetryFrame + WDS(1), R1_PTR & R2_PTR, stg_catchRetryzh);
888 /* Start a nested transaction within which to run the first code */
889 trec = StgTSO_trec(CurrentTSO);
890 ("ptr" new_trec) = foreign "C" stmStartTransaction(MyCapability() "ptr", trec "ptr") [R1,R2];
891 StgTSO_trec(CurrentTSO) = new_trec;
893 /* Set up the catch-retry frame */
894 Sp = Sp - SIZEOF_StgCatchRetryFrame;
897 SET_HDR(frame, stg_catch_retry_frame_info, W_[CCCS]);
898 StgCatchRetryFrame_running_alt_code(frame) = 0 :: CInt; // false;
899 StgCatchRetryFrame_first_code(frame) = R1;
900 StgCatchRetryFrame_alt_code(frame) = R2;
902 /* Apply R1 to the realworld token */
915 MAYBE_GC (NO_PTRS, stg_retryzh); // STM operations may allocate
917 // Find the enclosing ATOMICALLY_FRAME or CATCH_RETRY_FRAME
919 StgTSO_sp(CurrentTSO) = Sp;
920 (frame_type) = foreign "C" findRetryFrameHelper(CurrentTSO "ptr") [];
921 Sp = StgTSO_sp(CurrentTSO);
923 trec = StgTSO_trec(CurrentTSO);
924 outer = StgTRecHeader_enclosing_trec(trec);
926 if (frame_type == CATCH_RETRY_FRAME) {
927 // The retry reaches a CATCH_RETRY_FRAME before the atomic frame
928 ASSERT(outer != NO_TREC);
929 // Abort the transaction attempting the current branch
930 foreign "C" stmAbortTransaction(MyCapability() "ptr", trec "ptr") [];
931 foreign "C" stmFreeAbortedTRec(MyCapability() "ptr", trec "ptr") [];
932 if (!StgCatchRetryFrame_running_alt_code(frame) != 0::I32) {
933 // Retry in the first branch: try the alternative
934 ("ptr" trec) = foreign "C" stmStartTransaction(MyCapability() "ptr", outer "ptr") [];
935 StgTSO_trec(CurrentTSO) = trec;
936 StgCatchRetryFrame_running_alt_code(frame) = 1 :: CInt; // true;
937 R1 = StgCatchRetryFrame_alt_code(frame);
940 // Retry in the alternative code: propagate the retry
941 StgTSO_trec(CurrentTSO) = outer;
942 Sp = Sp + SIZEOF_StgCatchRetryFrame;
943 goto retry_pop_stack;
947 // We've reached the ATOMICALLY_FRAME: attempt to wait
948 ASSERT(frame_type == ATOMICALLY_FRAME);
949 if (outer != NO_TREC) {
950 // We called retry while checking invariants, so abort the current
951 // invariant check (merging its TVar accesses into the parents read
952 // set so we'll wait on them)
953 foreign "C" stmAbortTransaction(MyCapability() "ptr", trec "ptr") [];
954 foreign "C" stmFreeAbortedTRec(MyCapability() "ptr", trec "ptr") [];
956 StgTSO_trec(CurrentTSO) = trec;
957 outer = StgTRecHeader_enclosing_trec(trec);
959 ASSERT(outer == NO_TREC);
961 (r) = foreign "C" stmWait(MyCapability() "ptr", CurrentTSO "ptr", trec "ptr") [];
963 // Transaction was valid: stmWait put us on the TVars' queues, we now block
964 StgHeader_info(frame) = stg_atomically_waiting_frame_info;
966 // Fix up the stack in the unregisterised case: the return convention is different.
967 R3 = trec; // passing to stmWaitUnblock()
968 jump stg_block_stmwait;
970 // Transaction was not valid: retry immediately
971 ("ptr" trec) = foreign "C" stmStartTransaction(MyCapability() "ptr", outer "ptr") [];
972 StgTSO_trec(CurrentTSO) = trec;
973 R1 = StgAtomicallyFrame_code(frame);
984 /* Args: R1 = invariant closure */
985 MAYBE_GC (R1_PTR, stg_checkzh);
987 trec = StgTSO_trec(CurrentTSO);
989 foreign "C" stmAddInvariantToCheck(MyCapability() "ptr",
993 jump %ENTRY_CODE(Sp(0));
1002 /* Args: R1 = initialisation value */
1004 MAYBE_GC (R1_PTR, stg_newTVarzh);
1006 ("ptr" tv) = foreign "C" stmNewTVar(MyCapability() "ptr", new_value "ptr") [];
1017 /* Args: R1 = TVar closure */
1019 MAYBE_GC (R1_PTR, stg_readTVarzh); // Call to stmReadTVar may allocate
1020 trec = StgTSO_trec(CurrentTSO);
1022 ("ptr" result) = foreign "C" stmReadTVar(MyCapability() "ptr", trec "ptr", tvar "ptr") [];
1032 result = StgTVar_current_value(R1);
1033 if (%INFO_PTR(result) == stg_TREC_HEADER_info) {
1045 /* Args: R1 = TVar closure */
1046 /* R2 = New value */
1048 MAYBE_GC (R1_PTR & R2_PTR, stg_writeTVarzh); // Call to stmWriteTVar may allocate
1049 trec = StgTSO_trec(CurrentTSO);
1052 foreign "C" stmWriteTVar(MyCapability() "ptr", trec "ptr", tvar "ptr", new_value "ptr") [];
1054 jump %ENTRY_CODE(Sp(0));
1058 /* -----------------------------------------------------------------------------
1061 * take & putMVar work as follows. Firstly, an important invariant:
1063 * If the MVar is full, then the blocking queue contains only
1064 * threads blocked on putMVar, and if the MVar is empty then the
1065 * blocking queue contains only threads blocked on takeMVar.
1068 * MVar empty : then add ourselves to the blocking queue
1069 * MVar full : remove the value from the MVar, and
1070 * blocking queue empty : return
1071 * blocking queue non-empty : perform the first blocked putMVar
1072 * from the queue, and wake up the
1073 * thread (MVar is now full again)
1075 * putMVar is just the dual of the above algorithm.
1077 * How do we "perform a putMVar"? Well, we have to fiddle around with
1078 * the stack of the thread waiting to do the putMVar. See
1079 * stg_block_putmvar and stg_block_takemvar in HeapStackCheck.c for
1080 * the stack layout, and the PerformPut and PerformTake macros below.
1082 * It is important that a blocked take or put is woken up with the
1083 * take/put already performed, because otherwise there would be a
1084 * small window of vulnerability where the thread could receive an
1085 * exception and never perform its take or put, and we'd end up with a
1088 * -------------------------------------------------------------------------- */
1092 /* args: R1 = MVar closure */
1094 if (StgMVar_value(R1) == stg_END_TSO_QUEUE_closure) {
1106 ALLOC_PRIM ( SIZEOF_StgMVar, NO_PTRS, stg_newMVarzh );
1108 mvar = Hp - SIZEOF_StgMVar + WDS(1);
1109 SET_HDR(mvar,stg_MVAR_DIRTY_info,W_[CCCS]);
1110 // MVARs start dirty: generation 0 has no mutable list
1111 StgMVar_head(mvar) = stg_END_TSO_QUEUE_closure;
1112 StgMVar_tail(mvar) = stg_END_TSO_QUEUE_closure;
1113 StgMVar_value(mvar) = stg_END_TSO_QUEUE_closure;
1118 #define PerformTake(tso, value) \
1119 W_[StgTSO_sp(tso) + WDS(1)] = value; \
1120 W_[StgTSO_sp(tso) + WDS(0)] = stg_gc_unpt_r1_info;
1122 #define PerformPut(tso,lval) \
1123 StgTSO_sp(tso) = StgTSO_sp(tso) + WDS(3); \
1124 lval = W_[StgTSO_sp(tso) - WDS(1)];
1128 W_ mvar, val, info, tso;
1130 /* args: R1 = MVar closure */
1133 #if defined(THREADED_RTS)
1134 ("ptr" info) = foreign "C" lockClosure(mvar "ptr") [];
1136 info = GET_INFO(mvar);
1139 if (info == stg_MVAR_CLEAN_info) {
1140 foreign "C" dirty_MVAR(BaseReg "ptr", mvar "ptr") [];
1143 /* If the MVar is empty, put ourselves on its blocking queue,
1144 * and wait until we're woken up.
1146 if (StgMVar_value(mvar) == stg_END_TSO_QUEUE_closure) {
1147 if (StgMVar_head(mvar) == stg_END_TSO_QUEUE_closure) {
1148 StgMVar_head(mvar) = CurrentTSO;
1150 foreign "C" setTSOLink(MyCapability() "ptr",
1151 StgMVar_tail(mvar) "ptr",
1154 StgTSO__link(CurrentTSO) = stg_END_TSO_QUEUE_closure;
1155 StgTSO_block_info(CurrentTSO) = mvar;
1156 // write barrier for throwTo(), which looks at block_info
1157 // if why_blocked==BlockedOnMVar.
1158 prim %write_barrier() [];
1159 StgTSO_why_blocked(CurrentTSO) = BlockedOnMVar::I16;
1160 StgMVar_tail(mvar) = CurrentTSO;
1163 jump stg_block_takemvar;
1166 /* we got the value... */
1167 val = StgMVar_value(mvar);
1169 if (StgMVar_head(mvar) != stg_END_TSO_QUEUE_closure)
1171 /* There are putMVar(s) waiting...
1172 * wake up the first thread on the queue
1174 ASSERT(StgTSO_why_blocked(StgMVar_head(mvar)) == BlockedOnMVar::I16);
1176 /* actually perform the putMVar for the thread that we just woke up */
1177 tso = StgMVar_head(mvar);
1178 PerformPut(tso,StgMVar_value(mvar));
1180 if (TO_W_(StgTSO_dirty(tso)) == 0) {
1181 foreign "C" dirty_TSO(MyCapability() "ptr", tso "ptr") [];
1184 ("ptr" tso) = foreign "C" unblockOne_(MyCapability() "ptr",
1185 StgMVar_head(mvar) "ptr", 1) [];
1186 StgMVar_head(mvar) = tso;
1188 if (StgMVar_head(mvar) == stg_END_TSO_QUEUE_closure) {
1189 StgMVar_tail(mvar) = stg_END_TSO_QUEUE_closure;
1192 #if defined(THREADED_RTS)
1193 unlockClosure(mvar, stg_MVAR_DIRTY_info);
1195 SET_INFO(mvar,stg_MVAR_DIRTY_info);
1201 /* No further putMVars, MVar is now empty */
1202 StgMVar_value(mvar) = stg_END_TSO_QUEUE_closure;
1204 #if defined(THREADED_RTS)
1205 unlockClosure(mvar, stg_MVAR_DIRTY_info);
1207 SET_INFO(mvar,stg_MVAR_DIRTY_info);
1217 W_ mvar, val, info, tso;
1219 /* args: R1 = MVar closure */
1223 #if defined(THREADED_RTS)
1224 ("ptr" info) = foreign "C" lockClosure(mvar "ptr") [];
1226 info = GET_INFO(mvar);
1229 if (StgMVar_value(mvar) == stg_END_TSO_QUEUE_closure) {
1230 #if defined(THREADED_RTS)
1231 unlockClosure(mvar, info);
1233 /* HACK: we need a pointer to pass back,
1234 * so we abuse NO_FINALIZER_closure
1236 RET_NP(0, stg_NO_FINALIZER_closure);
1239 if (info == stg_MVAR_CLEAN_info) {
1240 foreign "C" dirty_MVAR(BaseReg "ptr", mvar "ptr");
1243 /* we got the value... */
1244 val = StgMVar_value(mvar);
1246 if (StgMVar_head(mvar) != stg_END_TSO_QUEUE_closure) {
1248 /* There are putMVar(s) waiting...
1249 * wake up the first thread on the queue
1251 ASSERT(StgTSO_why_blocked(StgMVar_head(mvar)) == BlockedOnMVar::I16);
1253 /* actually perform the putMVar for the thread that we just woke up */
1254 tso = StgMVar_head(mvar);
1255 PerformPut(tso,StgMVar_value(mvar));
1256 if (TO_W_(StgTSO_dirty(tso)) == 0) {
1257 foreign "C" dirty_TSO(MyCapability() "ptr", tso "ptr") [];
1260 ("ptr" tso) = foreign "C" unblockOne_(MyCapability() "ptr",
1261 StgMVar_head(mvar) "ptr", 1) [];
1262 StgMVar_head(mvar) = tso;
1264 if (StgMVar_head(mvar) == stg_END_TSO_QUEUE_closure) {
1265 StgMVar_tail(mvar) = stg_END_TSO_QUEUE_closure;
1267 #if defined(THREADED_RTS)
1268 unlockClosure(mvar, stg_MVAR_DIRTY_info);
1270 SET_INFO(mvar,stg_MVAR_DIRTY_info);
1275 /* No further putMVars, MVar is now empty */
1276 StgMVar_value(mvar) = stg_END_TSO_QUEUE_closure;
1277 #if defined(THREADED_RTS)
1278 unlockClosure(mvar, stg_MVAR_DIRTY_info);
1280 SET_INFO(mvar,stg_MVAR_DIRTY_info);
1290 W_ mvar, val, info, tso;
1292 /* args: R1 = MVar, R2 = value */
1296 #if defined(THREADED_RTS)
1297 ("ptr" info) = foreign "C" lockClosure(mvar "ptr") [];
1299 info = GET_INFO(mvar);
1302 if (info == stg_MVAR_CLEAN_info) {
1303 foreign "C" dirty_MVAR(BaseReg "ptr", mvar "ptr");
1306 if (StgMVar_value(mvar) != stg_END_TSO_QUEUE_closure) {
1307 if (StgMVar_head(mvar) == stg_END_TSO_QUEUE_closure) {
1308 StgMVar_head(mvar) = CurrentTSO;
1310 foreign "C" setTSOLink(MyCapability() "ptr",
1311 StgMVar_tail(mvar) "ptr",
1314 StgTSO__link(CurrentTSO) = stg_END_TSO_QUEUE_closure;
1315 StgTSO_block_info(CurrentTSO) = mvar;
1316 // write barrier for throwTo(), which looks at block_info
1317 // if why_blocked==BlockedOnMVar.
1318 prim %write_barrier() [];
1319 StgTSO_why_blocked(CurrentTSO) = BlockedOnMVar::I16;
1320 StgMVar_tail(mvar) = CurrentTSO;
1324 jump stg_block_putmvar;
1327 if (StgMVar_head(mvar) != stg_END_TSO_QUEUE_closure) {
1329 /* There are takeMVar(s) waiting: wake up the first one
1331 ASSERT(StgTSO_why_blocked(StgMVar_head(mvar)) == BlockedOnMVar::I16);
1333 /* actually perform the takeMVar */
1334 tso = StgMVar_head(mvar);
1335 PerformTake(tso, val);
1336 if (TO_W_(StgTSO_dirty(tso)) == 0) {
1337 foreign "C" dirty_TSO(MyCapability() "ptr", tso "ptr") [];
1340 ("ptr" tso) = foreign "C" unblockOne_(MyCapability() "ptr",
1341 StgMVar_head(mvar) "ptr", 1) [];
1342 StgMVar_head(mvar) = tso;
1344 if (StgMVar_head(mvar) == stg_END_TSO_QUEUE_closure) {
1345 StgMVar_tail(mvar) = stg_END_TSO_QUEUE_closure;
1348 #if defined(THREADED_RTS)
1349 unlockClosure(mvar, stg_MVAR_DIRTY_info);
1351 SET_INFO(mvar,stg_MVAR_DIRTY_info);
1353 jump %ENTRY_CODE(Sp(0));
1357 /* No further takes, the MVar is now full. */
1358 StgMVar_value(mvar) = val;
1360 #if defined(THREADED_RTS)
1361 unlockClosure(mvar, stg_MVAR_DIRTY_info);
1363 SET_INFO(mvar,stg_MVAR_DIRTY_info);
1365 jump %ENTRY_CODE(Sp(0));
1368 /* ToDo: yield afterward for better communication performance? */
1376 /* args: R1 = MVar, R2 = value */
1379 #if defined(THREADED_RTS)
1380 ("ptr" info) = foreign "C" lockClosure(mvar "ptr") [R2];
1382 info = GET_INFO(mvar);
1385 if (StgMVar_value(mvar) != stg_END_TSO_QUEUE_closure) {
1386 #if defined(THREADED_RTS)
1387 unlockClosure(mvar, info);
1392 if (info == stg_MVAR_CLEAN_info) {
1393 foreign "C" dirty_MVAR(BaseReg "ptr", mvar "ptr");
1396 if (StgMVar_head(mvar) != stg_END_TSO_QUEUE_closure) {
1398 /* There are takeMVar(s) waiting: wake up the first one
1400 ASSERT(StgTSO_why_blocked(StgMVar_head(mvar)) == BlockedOnMVar::I16);
1402 /* actually perform the takeMVar */
1403 tso = StgMVar_head(mvar);
1404 PerformTake(tso, R2);
1405 if (TO_W_(StgTSO_dirty(tso)) == 0) {
1406 foreign "C" dirty_TSO(MyCapability() "ptr", tso "ptr") [];
1409 ("ptr" tso) = foreign "C" unblockOne_(MyCapability() "ptr",
1410 StgMVar_head(mvar) "ptr", 1) [];
1411 StgMVar_head(mvar) = tso;
1413 if (StgMVar_head(mvar) == stg_END_TSO_QUEUE_closure) {
1414 StgMVar_tail(mvar) = stg_END_TSO_QUEUE_closure;
1417 #if defined(THREADED_RTS)
1418 unlockClosure(mvar, stg_MVAR_DIRTY_info);
1420 SET_INFO(mvar,stg_MVAR_DIRTY_info);
1425 /* No further takes, the MVar is now full. */
1426 StgMVar_value(mvar) = R2;
1428 #if defined(THREADED_RTS)
1429 unlockClosure(mvar, stg_MVAR_DIRTY_info);
1431 SET_INFO(mvar,stg_MVAR_DIRTY_info);
1436 /* ToDo: yield afterward for better communication performance? */
1440 /* -----------------------------------------------------------------------------
1441 Stable pointer primitives
1442 ------------------------------------------------------------------------- */
1444 stg_makeStableNamezh
1448 ALLOC_PRIM( SIZEOF_StgStableName, R1_PTR, stg_makeStableNamezh );
1450 (index) = foreign "C" lookupStableName(R1 "ptr") [];
1452 /* Is there already a StableName for this heap object?
1453 * stable_ptr_table is a pointer to an array of snEntry structs.
1455 if ( snEntry_sn_obj(W_[stable_ptr_table] + index*SIZEOF_snEntry) == NULL ) {
1456 sn_obj = Hp - SIZEOF_StgStableName + WDS(1);
1457 SET_HDR(sn_obj, stg_STABLE_NAME_info, W_[CCCS]);
1458 StgStableName_sn(sn_obj) = index;
1459 snEntry_sn_obj(W_[stable_ptr_table] + index*SIZEOF_snEntry) = sn_obj;
1461 sn_obj = snEntry_sn_obj(W_[stable_ptr_table] + index*SIZEOF_snEntry);
1472 MAYBE_GC(R1_PTR, stg_makeStablePtrzh);
1473 ("ptr" sp) = foreign "C" getStablePtr(R1 "ptr") [];
1477 stg_deRefStablePtrzh
1479 /* Args: R1 = the stable ptr */
1482 r = snEntry_addr(W_[stable_ptr_table] + sp*SIZEOF_snEntry);
1486 /* -----------------------------------------------------------------------------
1487 Bytecode object primitives
1488 ------------------------------------------------------------------------- */
1498 W_ bco, bitmap_arr, bytes, words;
1502 words = BYTES_TO_WDS(SIZEOF_StgBCO) + StgArrWords_words(bitmap_arr);
1505 ALLOC_PRIM( bytes, R1_PTR&R2_PTR&R3_PTR&R5_PTR, stg_newBCOzh );
1507 bco = Hp - bytes + WDS(1);
1508 SET_HDR(bco, stg_BCO_info, W_[CCCS]);
1510 StgBCO_instrs(bco) = R1;
1511 StgBCO_literals(bco) = R2;
1512 StgBCO_ptrs(bco) = R3;
1513 StgBCO_arity(bco) = HALF_W_(R4);
1514 StgBCO_size(bco) = HALF_W_(words);
1516 // Copy the arity/bitmap info into the BCO
1520 if (i < StgArrWords_words(bitmap_arr)) {
1521 StgBCO_bitmap(bco,i) = StgArrWords_payload(bitmap_arr,i);
1532 // R1 = the BCO# for the AP
1536 // This function is *only* used to wrap zero-arity BCOs in an
1537 // updatable wrapper (see ByteCodeLink.lhs). An AP thunk is always
1538 // saturated and always points directly to a FUN or BCO.
1539 ASSERT(%INFO_TYPE(%GET_STD_INFO(R1)) == HALF_W_(BCO) &&
1540 StgBCO_arity(R1) == HALF_W_(0));
1542 HP_CHK_GEN_TICKY(SIZEOF_StgAP, R1_PTR, stg_mkApUpd0zh);
1543 TICK_ALLOC_UP_THK(0, 0);
1544 CCCS_ALLOC(SIZEOF_StgAP);
1546 ap = Hp - SIZEOF_StgAP + WDS(1);
1547 SET_HDR(ap, stg_AP_info, W_[CCCS]);
1549 StgAP_n_args(ap) = HALF_W_(0);
1557 /* args: R1 = closure to analyze */
1558 // TODO: Consider the absence of ptrs or nonptrs as a special case ?
1560 W_ info, ptrs, nptrs, p, ptrs_arr, nptrs_arr;
1561 info = %GET_STD_INFO(UNTAG(R1));
1563 // Some closures have non-standard layout, so we omit those here.
1565 type = TO_W_(%INFO_TYPE(info));
1566 switch [0 .. N_CLOSURE_TYPES] type {
1567 case THUNK_SELECTOR : {
1572 case THUNK, THUNK_1_0, THUNK_0_1, THUNK_2_0, THUNK_1_1,
1573 THUNK_0_2, THUNK_STATIC, AP, PAP, AP_STACK, BCO : {
1579 ptrs = TO_W_(%INFO_PTRS(info));
1580 nptrs = TO_W_(%INFO_NPTRS(info));
1585 W_ ptrs_arr_sz, nptrs_arr_sz;
1586 nptrs_arr_sz = SIZEOF_StgArrWords + WDS(nptrs);
1587 ptrs_arr_sz = SIZEOF_StgMutArrPtrs + WDS(ptrs);
1589 ALLOC_PRIM (ptrs_arr_sz + nptrs_arr_sz, R1_PTR, stg_unpackClosurezh);
1594 ptrs_arr = Hp - nptrs_arr_sz - ptrs_arr_sz + WDS(1);
1595 nptrs_arr = Hp - nptrs_arr_sz + WDS(1);
1597 SET_HDR(ptrs_arr, stg_MUT_ARR_PTRS_FROZEN_info, W_[CCCS]);
1598 StgMutArrPtrs_ptrs(ptrs_arr) = ptrs;
1602 W_[ptrs_arr + SIZEOF_StgMutArrPtrs + WDS(p)] = StgClosure_payload(clos,p);
1607 SET_HDR(nptrs_arr, stg_ARR_WORDS_info, W_[CCCS]);
1608 StgArrWords_words(nptrs_arr) = nptrs;
1612 W_[BYTE_ARR_CTS(nptrs_arr) + WDS(p)] = StgClosure_payload(clos, p+ptrs);
1616 RET_NPP(info, ptrs_arr, nptrs_arr);
1619 /* -----------------------------------------------------------------------------
1620 Thread I/O blocking primitives
1621 -------------------------------------------------------------------------- */
1623 /* Add a thread to the end of the blocked queue. (C-- version of the C
1624 * macro in Schedule.h).
1626 #define APPEND_TO_BLOCKED_QUEUE(tso) \
1627 ASSERT(StgTSO__link(tso) == END_TSO_QUEUE); \
1628 if (W_[blocked_queue_hd] == END_TSO_QUEUE) { \
1629 W_[blocked_queue_hd] = tso; \
1631 foreign "C" setTSOLink(MyCapability() "ptr", W_[blocked_queue_tl] "ptr", tso) []; \
1633 W_[blocked_queue_tl] = tso;
1639 foreign "C" barf("waitRead# on threaded RTS") never returns;
1642 ASSERT(StgTSO_why_blocked(CurrentTSO) == NotBlocked::I16);
1643 StgTSO_why_blocked(CurrentTSO) = BlockedOnRead::I16;
1644 StgTSO_block_info(CurrentTSO) = R1;
1645 // No locking - we're not going to use this interface in the
1646 // threaded RTS anyway.
1647 APPEND_TO_BLOCKED_QUEUE(CurrentTSO);
1648 jump stg_block_noregs;
1656 foreign "C" barf("waitWrite# on threaded RTS") never returns;
1659 ASSERT(StgTSO_why_blocked(CurrentTSO) == NotBlocked::I16);
1660 StgTSO_why_blocked(CurrentTSO) = BlockedOnWrite::I16;
1661 StgTSO_block_info(CurrentTSO) = R1;
1662 // No locking - we're not going to use this interface in the
1663 // threaded RTS anyway.
1664 APPEND_TO_BLOCKED_QUEUE(CurrentTSO);
1665 jump stg_block_noregs;
1670 STRING(stg_delayzh_malloc_str, "stg_delayzh")
1673 #ifdef mingw32_HOST_OS
1681 foreign "C" barf("delay# on threaded RTS") never returns;
1684 /* args: R1 (microsecond delay amount) */
1685 ASSERT(StgTSO_why_blocked(CurrentTSO) == NotBlocked::I16);
1686 StgTSO_why_blocked(CurrentTSO) = BlockedOnDelay::I16;
1688 #ifdef mingw32_HOST_OS
1690 /* could probably allocate this on the heap instead */
1691 ("ptr" ares) = foreign "C" stgMallocBytes(SIZEOF_StgAsyncIOResult,
1692 stg_delayzh_malloc_str);
1693 (reqID) = foreign "C" addDelayRequest(R1);
1694 StgAsyncIOResult_reqID(ares) = reqID;
1695 StgAsyncIOResult_len(ares) = 0;
1696 StgAsyncIOResult_errCode(ares) = 0;
1697 StgTSO_block_info(CurrentTSO) = ares;
1699 /* Having all async-blocked threads reside on the blocked_queue
1700 * simplifies matters, so change the status to OnDoProc put the
1701 * delayed thread on the blocked_queue.
1703 StgTSO_why_blocked(CurrentTSO) = BlockedOnDoProc::I16;
1704 APPEND_TO_BLOCKED_QUEUE(CurrentTSO);
1705 jump stg_block_async_void;
1711 (time) = foreign "C" getourtimeofday() [R1];
1712 divisor = TO_W_(RtsFlags_MiscFlags_tickInterval(RtsFlags));
1716 divisor = divisor * 1000;
1717 target = ((R1 + divisor - 1) / divisor) /* divide rounding up */
1718 + time + 1; /* Add 1 as getourtimeofday rounds down */
1719 StgTSO_block_info(CurrentTSO) = target;
1721 /* Insert the new thread in the sleeping queue. */
1723 t = W_[sleeping_queue];
1725 if (t != END_TSO_QUEUE && StgTSO_block_info(t) < target) {
1727 t = StgTSO__link(t);
1731 StgTSO__link(CurrentTSO) = t;
1733 W_[sleeping_queue] = CurrentTSO;
1735 foreign "C" setTSOLink(MyCapability() "ptr", prev "ptr", CurrentTSO) [];
1737 jump stg_block_noregs;
1739 #endif /* !THREADED_RTS */
1743 #ifdef mingw32_HOST_OS
1744 STRING(stg_asyncReadzh_malloc_str, "stg_asyncReadzh")
1751 foreign "C" barf("asyncRead# on threaded RTS") never returns;
1754 /* args: R1 = fd, R2 = isSock, R3 = len, R4 = buf */
1755 ASSERT(StgTSO_why_blocked(CurrentTSO) == NotBlocked::I16);
1756 StgTSO_why_blocked(CurrentTSO) = BlockedOnRead::I16;
1758 /* could probably allocate this on the heap instead */
1759 ("ptr" ares) = foreign "C" stgMallocBytes(SIZEOF_StgAsyncIOResult,
1760 stg_asyncReadzh_malloc_str)
1762 (reqID) = foreign "C" addIORequest(R1, 0/*FALSE*/,R2,R3,R4 "ptr") [];
1763 StgAsyncIOResult_reqID(ares) = reqID;
1764 StgAsyncIOResult_len(ares) = 0;
1765 StgAsyncIOResult_errCode(ares) = 0;
1766 StgTSO_block_info(CurrentTSO) = ares;
1767 APPEND_TO_BLOCKED_QUEUE(CurrentTSO);
1768 jump stg_block_async;
1772 STRING(stg_asyncWritezh_malloc_str, "stg_asyncWritezh")
1779 foreign "C" barf("asyncWrite# on threaded RTS") never returns;
1782 /* args: R1 = fd, R2 = isSock, R3 = len, R4 = buf */
1783 ASSERT(StgTSO_why_blocked(CurrentTSO) == NotBlocked::I16);
1784 StgTSO_why_blocked(CurrentTSO) = BlockedOnWrite::I16;
1786 ("ptr" ares) = foreign "C" stgMallocBytes(SIZEOF_StgAsyncIOResult,
1787 stg_asyncWritezh_malloc_str)
1789 (reqID) = foreign "C" addIORequest(R1, 1/*TRUE*/,R2,R3,R4 "ptr") [];
1791 StgAsyncIOResult_reqID(ares) = reqID;
1792 StgAsyncIOResult_len(ares) = 0;
1793 StgAsyncIOResult_errCode(ares) = 0;
1794 StgTSO_block_info(CurrentTSO) = ares;
1795 APPEND_TO_BLOCKED_QUEUE(CurrentTSO);
1796 jump stg_block_async;
1800 STRING(stg_asyncDoProczh_malloc_str, "stg_asyncDoProczh")
1807 foreign "C" barf("asyncDoProc# on threaded RTS") never returns;
1810 /* args: R1 = proc, R2 = param */
1811 ASSERT(StgTSO_why_blocked(CurrentTSO) == NotBlocked::I16);
1812 StgTSO_why_blocked(CurrentTSO) = BlockedOnDoProc::I16;
1814 /* could probably allocate this on the heap instead */
1815 ("ptr" ares) = foreign "C" stgMallocBytes(SIZEOF_StgAsyncIOResult,
1816 stg_asyncDoProczh_malloc_str)
1818 (reqID) = foreign "C" addDoProcRequest(R1 "ptr",R2 "ptr") [];
1819 StgAsyncIOResult_reqID(ares) = reqID;
1820 StgAsyncIOResult_len(ares) = 0;
1821 StgAsyncIOResult_errCode(ares) = 0;
1822 StgTSO_block_info(CurrentTSO) = ares;
1823 APPEND_TO_BLOCKED_QUEUE(CurrentTSO);
1824 jump stg_block_async;
1829 // noDuplicate# tries to ensure that none of the thunks under
1830 // evaluation by the current thread are also under evaluation by
1831 // another thread. It relies on *both* threads doing noDuplicate#;
1832 // the second one will get blocked if they are duplicating some work.
1835 SAVE_THREAD_STATE();
1836 ASSERT(StgTSO_what_next(CurrentTSO) == ThreadRunGHC::I16);
1837 foreign "C" threadPaused (MyCapability() "ptr", CurrentTSO "ptr") [];
1839 if (StgTSO_what_next(CurrentTSO) == ThreadKilled::I16) {
1840 jump stg_threadFinished;
1842 LOAD_THREAD_STATE();
1843 ASSERT(StgTSO_what_next(CurrentTSO) == ThreadRunGHC::I16);
1844 jump %ENTRY_CODE(Sp(0));
1850 W_ ap_stack, offset, val, ok;
1852 /* args: R1 = AP_STACK, R2 = offset */
1856 if (%INFO_PTR(ap_stack) == stg_AP_STACK_info) {
1858 val = StgAP_STACK_payload(ap_stack,offset);
1866 /* -----------------------------------------------------------------------------
1868 -------------------------------------------------------------------------- */
1870 // Write the cost center stack of the first argument on stderr; return
1871 // the second. Possibly only makes sense for already evaluated
1878 ccs = StgHeader_ccs(UNTAG(R1));
1879 foreign "C" fprintCCS_stderr(ccs "ptr") [R2];
1890 #ifndef THREADED_RTS
1891 RET_NP(0,ghczmprim_GHCziBool_False_closure);
1893 (spark) = foreign "C" findSpark(MyCapability());
1897 RET_NP(0,ghczmprim_GHCziBool_False_closure);
1907 #if defined(TRACING) || defined(DEBUG)
1909 foreign "C" traceUserMsg(MyCapability() "ptr", msg "ptr") [];
1911 #elif defined(DTRACE)
1915 // We should go through the macro HASKELLEVENT_USER_MSG_ENABLED from
1916 // RtsProbes.h, but that header file includes unistd.h, which doesn't
1918 (enabled) = foreign "C" __dtrace_isenabled$HaskellEvent$user__msg$v1() [];
1920 foreign "C" dtraceUserMsgWrapper(MyCapability() "ptr", msg "ptr") [];
1924 jump %ENTRY_CODE(Sp(0));