3 * Simple generational GC.
6 * Paolo Molaro (lupus@ximian.com)
7 * Rodrigo Kumpera (kumpera@gmail.com)
9 * Copyright 2005-2011 Novell, Inc (http://www.novell.com)
10 * Copyright 2011 Xamarin Inc (http://www.xamarin.com)
12 * Thread start/stop adapted from Boehm's GC:
13 * Copyright (c) 1994 by Xerox Corporation. All rights reserved.
14 * Copyright (c) 1996 by Silicon Graphics. All rights reserved.
15 * Copyright (c) 1998 by Fergus Henderson. All rights reserved.
16 * Copyright (c) 2000-2004 by Hewlett-Packard Company. All rights reserved.
17 * Copyright 2001-2003 Ximian, Inc
18 * Copyright 2003-2010 Novell, Inc.
19 * Copyright 2011 Xamarin, Inc.
20 * Copyright (C) 2012 Xamarin Inc
22 * Licensed under the MIT license. See LICENSE file in the project root for full license information.
24 * Important: allocation provides always zeroed memory, having to do
25 * a memset after allocation is deadly for performance.
26 * Memory usage at startup is currently as follows:
28 * 64 KB internal space
30 * We should provide a small memory config with half the sizes
32 * We currently try to make as few mono assumptions as possible:
33 * 1) 2-word header with no GC pointers in it (first vtable, second to store the
35 * 2) gc descriptor is the second word in the vtable (first word in the class)
36 * 3) 8 byte alignment is the minimum and enough (not true for special structures (SIMD), FIXME)
37 * 4) there is a function to get an object's size and the number of
38 * elements in an array.
39 * 5) we know the special way bounds are allocated for complex arrays
40 * 6) we know about proxies and how to treat them when domains are unloaded
42 * Always try to keep stack usage to a minimum: no recursive behaviour
43 * and no large stack allocs.
45 * General description.
46 * Objects are initially allocated in a nursery using a fast bump-pointer technique.
47 * When the nursery is full we start a nursery collection: this is performed with a
49 * When the old generation is full we start a copying GC of the old generation as well:
50 * this will be changed to mark&sweep with copying when fragmentation becomes to severe
51 * in the future. Maybe we'll even do both during the same collection like IMMIX.
53 * The things that complicate this description are:
54 * *) pinned objects: we can't move them so we need to keep track of them
55 * *) no precise info of the thread stacks and registers: we need to be able to
56 * quickly find the objects that may be referenced conservatively and pin them
57 * (this makes the first issues more important)
58 * *) large objects are too expensive to be dealt with using copying GC: we handle them
59 * with mark/sweep during major collections
60 * *) some objects need to not move even if they are small (interned strings, Type handles):
61 * we use mark/sweep for them, too: they are not allocated in the nursery, but inside
62 * PinnedChunks regions
68 *) we could have a function pointer in MonoClass to implement
69 customized write barriers for value types
71 *) investigate the stuff needed to advance a thread to a GC-safe
72 point (single-stepping, read from unmapped memory etc) and implement it.
73 This would enable us to inline allocations and write barriers, for example,
74 or at least parts of them, like the write barrier checks.
75 We may need this also for handling precise info on stacks, even simple things
76 as having uninitialized data on the stack and having to wait for the prolog
77 to zero it. Not an issue for the last frame that we scan conservatively.
78 We could always not trust the value in the slots anyway.
80 *) modify the jit to save info about references in stack locations:
81 this can be done just for locals as a start, so that at least
82 part of the stack is handled precisely.
84 *) test/fix endianess issues
86 *) Implement a card table as the write barrier instead of remembered
87 sets? Card tables are not easy to implement with our current
88 memory layout. We have several different kinds of major heap
89 objects: Small objects in regular blocks, small objects in pinned
90 chunks and LOS objects. If we just have a pointer we have no way
91 to tell which kind of object it points into, therefore we cannot
92 know where its card table is. The least we have to do to make
93 this happen is to get rid of write barriers for indirect stores.
96 *) Get rid of write barriers for indirect stores. We can do this by
97 telling the GC to wbarrier-register an object once we do an ldloca
98 or ldelema on it, and to unregister it once it's not used anymore
99 (it can only travel downwards on the stack). The problem with
100 unregistering is that it needs to happen eventually no matter
101 what, even if exceptions are thrown, the thread aborts, etc.
102 Rodrigo suggested that we could do only the registering part and
103 let the collector find out (pessimistically) when it's safe to
104 unregister, namely when the stack pointer of the thread that
105 registered the object is higher than it was when the registering
106 happened. This might make for a good first implementation to get
107 some data on performance.
109 *) Some sort of blacklist support? Blacklists is a concept from the
110 Boehm GC: if during a conservative scan we find pointers to an
111 area which we might use as heap, we mark that area as unusable, so
112 pointer retention by random pinning pointers is reduced.
114 *) experiment with max small object size (very small right now - 2kb,
115 because it's tied to the max freelist size)
117 *) add an option to mmap the whole heap in one chunk: it makes for many
118 simplifications in the checks (put the nursery at the top and just use a single
119 check for inclusion/exclusion): the issue this has is that on 32 bit systems it's
120 not flexible (too much of the address space may be used by default or we can't
121 increase the heap as needed) and we'd need a race-free mechanism to return memory
122 back to the system (mprotect(PROT_NONE) will still keep the memory allocated if it
123 was written to, munmap is needed, but the following mmap may not find the same segment
126 *) memzero the major fragments after restarting the world and optionally a smaller
129 *) investigate having fragment zeroing threads
131 *) separate locks for finalization and other minor stuff to reduce
134 *) try a different copying order to improve memory locality
136 *) a thread abort after a store but before the write barrier will
137 prevent the write barrier from executing
139 *) specialized dynamically generated markers/copiers
141 *) Dynamically adjust TLAB size to the number of threads. If we have
142 too many threads that do allocation, we might need smaller TLABs,
143 and we might get better performance with larger TLABs if we only
144 have a handful of threads. We could sum up the space left in all
145 assigned TLABs and if that's more than some percentage of the
146 nursery size, reduce the TLAB size.
148 *) Explore placing unreachable objects on unused nursery memory.
149 Instead of memset'ng a region to zero, place an int[] covering it.
150 A good place to start is add_nursery_frag. The tricky thing here is
151 placing those objects atomically outside of a collection.
153 *) Allocation should use asymmetric Dekker synchronization:
154 http://blogs.oracle.com/dave/resource/Asymmetric-Dekker-Synchronization.txt
155 This should help weak consistency archs.
162 #define _XOPEN_SOURCE
163 #define _DARWIN_C_SOURCE
169 #ifdef HAVE_PTHREAD_H
172 #ifdef HAVE_PTHREAD_NP_H
173 #include <pthread_np.h>
181 #include "mono/sgen/sgen-gc.h"
182 #include "mono/sgen/sgen-cardtable.h"
183 #include "mono/sgen/sgen-protocol.h"
184 #include "mono/sgen/sgen-memory-governor.h"
185 #include "mono/sgen/sgen-hash-table.h"
186 #include "mono/sgen/sgen-pinning.h"
187 #include "mono/sgen/sgen-workers.h"
188 #include "mono/sgen/sgen-client.h"
189 #include "mono/sgen/sgen-pointer-queue.h"
190 #include "mono/sgen/gc-internal-agnostic.h"
191 #include "mono/utils/mono-proclib.h"
192 #include "mono/utils/mono-memory-model.h"
193 #include "mono/utils/hazard-pointer.h"
195 #include <mono/utils/memcheck.h>
196 #include <mono/utils/mono-mmap-internals.h>
198 #undef pthread_create
200 #undef pthread_detach
203 * ######################################################################
204 * ######## Types and constants used by the GC.
205 * ######################################################################
208 /* 0 means not initialized, 1 is initialized, -1 means in progress */
209 static int gc_initialized = 0;
210 /* If set, check if we need to do something every X allocations */
211 gboolean has_per_allocation_action;
212 /* If set, do a heap check every X allocation */
213 guint32 verify_before_allocs = 0;
214 /* If set, do a minor collection before every X allocation */
215 guint32 collect_before_allocs = 0;
216 /* If set, do a whole heap check before each collection */
217 static gboolean whole_heap_check_before_collection = FALSE;
218 /* If set, do a remset consistency check at various opportunities */
219 static gboolean remset_consistency_checks = FALSE;
220 /* If set, do a mod union consistency check before each finishing collection pause */
221 static gboolean mod_union_consistency_check = FALSE;
222 /* If set, check whether mark bits are consistent after major collections */
223 static gboolean check_mark_bits_after_major_collection = FALSE;
224 /* If set, check that all nursery objects are pinned/not pinned, depending on context */
225 static gboolean check_nursery_objects_pinned = FALSE;
226 /* If set, do a few checks when the concurrent collector is used */
227 static gboolean do_concurrent_checks = FALSE;
228 /* If set, do a plausibility check on the scan_starts before and after
230 static gboolean do_scan_starts_check = FALSE;
232 static gboolean disable_minor_collections = FALSE;
233 static gboolean disable_major_collections = FALSE;
234 static gboolean do_verify_nursery = FALSE;
235 static gboolean do_dump_nursery_content = FALSE;
236 static gboolean enable_nursery_canaries = FALSE;
238 static gboolean precleaning_enabled = TRUE;
239 static gboolean dynamic_nursery = FALSE;
240 static size_t min_nursery_size = 0;
241 static size_t max_nursery_size = 0;
243 #ifdef HEAVY_STATISTICS
244 guint64 stat_objects_alloced_degraded = 0;
245 guint64 stat_bytes_alloced_degraded = 0;
247 guint64 stat_copy_object_called_nursery = 0;
248 guint64 stat_objects_copied_nursery = 0;
249 guint64 stat_copy_object_called_major = 0;
250 guint64 stat_objects_copied_major = 0;
252 guint64 stat_scan_object_called_nursery = 0;
253 guint64 stat_scan_object_called_major = 0;
255 guint64 stat_slots_allocated_in_vain;
257 guint64 stat_nursery_copy_object_failed_from_space = 0;
258 guint64 stat_nursery_copy_object_failed_forwarded = 0;
259 guint64 stat_nursery_copy_object_failed_pinned = 0;
260 guint64 stat_nursery_copy_object_failed_to_space = 0;
262 static guint64 stat_wbarrier_add_to_global_remset = 0;
263 static guint64 stat_wbarrier_arrayref_copy = 0;
264 static guint64 stat_wbarrier_generic_store = 0;
265 static guint64 stat_wbarrier_generic_store_atomic = 0;
266 static guint64 stat_wbarrier_set_root = 0;
269 static guint64 stat_pinned_objects = 0;
271 static guint64 time_minor_pre_collection_fragment_clear = 0;
272 static guint64 time_minor_pinning = 0;
273 static guint64 time_minor_scan_remsets = 0;
274 static guint64 time_minor_scan_major_blocks = 0;
275 static guint64 time_minor_scan_los = 0;
276 static guint64 time_minor_scan_pinned = 0;
277 static guint64 time_minor_scan_roots = 0;
278 static guint64 time_minor_finish_gray_stack = 0;
279 static guint64 time_minor_fragment_creation = 0;
281 static guint64 time_major_pre_collection_fragment_clear = 0;
282 static guint64 time_major_pinning = 0;
283 static guint64 time_major_scan_pinned = 0;
284 static guint64 time_major_scan_roots = 0;
285 static guint64 time_major_scan_mod_union_blocks = 0;
286 static guint64 time_major_scan_mod_union_los = 0;
287 static guint64 time_major_finish_gray_stack = 0;
288 static guint64 time_major_free_bigobjs = 0;
289 static guint64 time_major_los_sweep = 0;
290 static guint64 time_major_sweep = 0;
291 static guint64 time_major_fragment_creation = 0;
293 static guint64 time_max = 0;
295 static int sgen_max_pause_time = SGEN_DEFAULT_MAX_PAUSE_TIME;
296 static float sgen_max_pause_margin = SGEN_DEFAULT_MAX_PAUSE_MARGIN;
298 static SGEN_TV_DECLARE (time_major_conc_collection_start);
299 static SGEN_TV_DECLARE (time_major_conc_collection_end);
301 int gc_debug_level = 0;
303 static char* gc_params_options;
304 static char* gc_debug_options;
308 mono_gc_flush_info (void)
310 fflush (gc_debug_file);
314 #define TV_DECLARE SGEN_TV_DECLARE
315 #define TV_GETTIME SGEN_TV_GETTIME
316 #define TV_ELAPSED SGEN_TV_ELAPSED
318 static SGEN_TV_DECLARE (sgen_init_timestamp);
320 NurseryClearPolicy nursery_clear_policy = CLEAR_AT_TLAB_CREATION;
322 #define object_is_forwarded SGEN_OBJECT_IS_FORWARDED
323 #define object_is_pinned SGEN_OBJECT_IS_PINNED
324 #define pin_object SGEN_PIN_OBJECT
326 #define ptr_in_nursery sgen_ptr_in_nursery
328 #define LOAD_VTABLE SGEN_LOAD_VTABLE
331 nursery_canaries_enabled (void)
333 return enable_nursery_canaries;
336 #define safe_object_get_size sgen_safe_object_get_size
338 #if defined(HAVE_CONC_GC_AS_DEFAULT)
339 /* Use concurrent major on deskstop platforms */
340 #define DEFAULT_MAJOR SGEN_MAJOR_CONCURRENT
342 #define DEFAULT_MAJOR SGEN_MAJOR_SERIAL
348 SGEN_MAJOR_CONCURRENT,
349 SGEN_MAJOR_CONCURRENT_PARALLEL
355 SGEN_MINOR_SIMPLE_PARALLEL,
362 SGEN_MODE_THROUGHPUT,
367 * ######################################################################
368 * ######## Global data.
369 * ######################################################################
371 MonoCoopMutex gc_mutex;
373 #define SCAN_START_SIZE SGEN_SCAN_START_SIZE
375 size_t degraded_mode = 0;
377 static mword bytes_pinned_from_failed_allocation = 0;
379 GCMemSection *nursery_section = NULL;
380 static volatile mword lowest_heap_address = ~(mword)0;
381 static volatile mword highest_heap_address = 0;
383 MonoCoopMutex sgen_interruption_mutex;
385 int current_collection_generation = -1;
386 volatile gboolean concurrent_collection_in_progress = FALSE;
388 /* objects that are ready to be finalized */
389 static SgenPointerQueue fin_ready_queue = SGEN_POINTER_QUEUE_INIT (INTERNAL_MEM_FINALIZE_READY);
390 static SgenPointerQueue critical_fin_queue = SGEN_POINTER_QUEUE_INIT (INTERNAL_MEM_FINALIZE_READY);
392 /* registered roots: the key to the hash is the root start address */
394 * Different kinds of roots are kept separate to speed up pin_from_roots () for example.
396 SgenHashTable roots_hash [ROOT_TYPE_NUM] = {
397 SGEN_HASH_TABLE_INIT (INTERNAL_MEM_ROOTS_TABLE, INTERNAL_MEM_ROOT_RECORD, sizeof (RootRecord), sgen_aligned_addr_hash, NULL),
398 SGEN_HASH_TABLE_INIT (INTERNAL_MEM_ROOTS_TABLE, INTERNAL_MEM_ROOT_RECORD, sizeof (RootRecord), sgen_aligned_addr_hash, NULL),
399 SGEN_HASH_TABLE_INIT (INTERNAL_MEM_ROOTS_TABLE, INTERNAL_MEM_ROOT_RECORD, sizeof (RootRecord), sgen_aligned_addr_hash, NULL)
401 static mword roots_size = 0; /* amount of memory in the root set */
403 /* The size of a TLAB */
404 /* The bigger the value, the less often we have to go to the slow path to allocate a new
405 * one, but the more space is wasted by threads not allocating much memory.
407 * FIXME: Make this self-tuning for each thread.
409 guint32 tlab_size = (1024 * 4);
411 #define MAX_SMALL_OBJ_SIZE SGEN_MAX_SMALL_OBJ_SIZE
413 #define ALLOC_ALIGN SGEN_ALLOC_ALIGN
415 #define ALIGN_UP SGEN_ALIGN_UP
417 #ifdef SGEN_DEBUG_INTERNAL_ALLOC
418 MonoNativeThreadId main_gc_thread = NULL;
421 /*Object was pinned during the current collection*/
422 static mword objects_pinned;
425 * ######################################################################
426 * ######## Macros and function declarations.
427 * ######################################################################
430 /* forward declarations */
431 static void scan_from_registered_roots (char *addr_start, char *addr_end, int root_type, ScanCopyContext ctx);
433 static void pin_from_roots (void *start_nursery, void *end_nursery, ScanCopyContext ctx);
434 static void finish_gray_stack (int generation, ScanCopyContext ctx);
437 SgenMajorCollector major_collector;
438 SgenMinorCollector sgen_minor_collector;
440 static SgenRememberedSet remset;
443 * The gray queue a worker job must use. If we're not parallel or
444 * concurrent, we use the main gray queue.
446 static SgenGrayQueue*
447 sgen_workers_get_job_gray_queue (WorkerData *worker_data, SgenGrayQueue *default_gray_queue)
450 return &worker_data->private_gray_queue;
451 SGEN_ASSERT (0, default_gray_queue, "Why don't we have a default gray queue when we're not running in a worker thread?");
452 return default_gray_queue;
456 gray_queue_redirect (SgenGrayQueue *queue)
458 sgen_workers_take_from_queue (current_collection_generation, queue);
462 sgen_scan_area_with_callback (char *start, char *end, IterateObjectCallbackFunc callback, void *data, gboolean allow_flags, gboolean fail_on_canaries)
464 while (start < end) {
468 if (!*(void**)start) {
469 start += sizeof (void*); /* should be ALLOC_ALIGN, really */
474 if (!(obj = (char *)SGEN_OBJECT_IS_FORWARDED (start)))
480 if (!sgen_client_object_is_array_fill ((GCObject*)obj)) {
481 CHECK_CANARY_FOR_OBJECT ((GCObject*)obj, fail_on_canaries);
482 size = ALIGN_UP (safe_object_get_size ((GCObject*)obj));
483 callback ((GCObject*)obj, size, data);
484 CANARIFY_SIZE (size);
486 size = ALIGN_UP (safe_object_get_size ((GCObject*)obj));
494 * sgen_add_to_global_remset:
496 * The global remset contains locations which point into newspace after
497 * a minor collection. This can happen if the objects they point to are pinned.
499 * LOCKING: If called from a parallel collector, the global remset
500 * lock must be held. For serial collectors that is not necessary.
503 sgen_add_to_global_remset (gpointer ptr, GCObject *obj)
505 SGEN_ASSERT (5, sgen_ptr_in_nursery (obj), "Target pointer of global remset must be in the nursery");
507 HEAVY_STAT (++stat_wbarrier_add_to_global_remset);
509 if (!major_collector.is_concurrent) {
510 SGEN_ASSERT (5, current_collection_generation != -1, "Global remsets can only be added during collections");
512 if (current_collection_generation == -1)
513 SGEN_ASSERT (5, sgen_concurrent_collection_in_progress (), "Global remsets outside of collection pauses can only be added by the concurrent collector");
516 if (!object_is_pinned (obj))
517 SGEN_ASSERT (5, sgen_minor_collector.is_split || sgen_concurrent_collection_in_progress (), "Non-pinned objects can only remain in nursery if it is a split nursery");
518 else if (sgen_cement_lookup_or_register (obj))
521 remset.record_pointer (ptr);
523 sgen_pin_stats_register_global_remset (obj);
525 SGEN_LOG (8, "Adding global remset for %p", ptr);
526 binary_protocol_global_remset (ptr, obj, (gpointer)SGEN_LOAD_VTABLE (obj));
530 * sgen_drain_gray_stack:
532 * Scan objects in the gray stack until the stack is empty. This should be called
533 * frequently after each object is copied, to achieve better locality and cache
538 sgen_drain_gray_stack (ScanCopyContext ctx)
540 SGEN_ASSERT (0, ctx.ops->drain_gray_stack, "Why do we have a scan/copy context with a missing drain gray stack function?");
542 return ctx.ops->drain_gray_stack (ctx.queue);
546 * Addresses in the pin queue are already sorted. This function finds
547 * the object header for each address and pins the object. The
548 * addresses must be inside the nursery section. The (start of the)
549 * address array is overwritten with the addresses of the actually
550 * pinned objects. Return the number of pinned objects.
553 pin_objects_from_nursery_pin_queue (gboolean do_scan_objects, ScanCopyContext ctx)
555 GCMemSection *section = nursery_section;
556 void **start = sgen_pinning_get_entry (section->pin_queue_first_entry);
557 void **end = sgen_pinning_get_entry (section->pin_queue_last_entry);
558 void *start_nursery = section->data;
559 void *end_nursery = section->end_data;
564 void *pinning_front = start_nursery;
566 void **definitely_pinned = start;
567 ScanObjectFunc scan_func = ctx.ops->scan_object;
568 SgenGrayQueue *queue = ctx.queue;
570 sgen_nursery_allocator_prepare_for_pinning ();
572 while (start < end) {
573 GCObject *obj_to_pin = NULL;
574 size_t obj_to_pin_size = 0;
579 SGEN_ASSERT (0, addr >= start_nursery && addr < end_nursery, "Potential pinning address out of range");
580 SGEN_ASSERT (0, addr >= last, "Pin queue not sorted");
587 SGEN_LOG (5, "Considering pinning addr %p", addr);
588 /* We've already processed everything up to pinning_front. */
589 if (addr < pinning_front) {
595 * Find the closest scan start <= addr. We might search backward in the
596 * scan_starts array because entries might be NULL. In the worst case we
597 * start at start_nursery.
599 idx = ((char*)addr - (char*)section->data) / SCAN_START_SIZE;
600 SGEN_ASSERT (0, idx < section->num_scan_start, "Scan start index out of range");
601 search_start = (void*)section->scan_starts [idx];
602 if (!search_start || search_start > addr) {
605 search_start = section->scan_starts [idx];
606 if (search_start && search_start <= addr)
609 if (!search_start || search_start > addr)
610 search_start = start_nursery;
614 * If the pinning front is closer than the scan start we found, start
615 * searching at the front.
617 if (search_start < pinning_front)
618 search_start = pinning_front;
621 * Now addr should be in an object a short distance from search_start.
623 * search_start must point to zeroed mem or point to an object.
626 size_t obj_size, canarified_obj_size;
629 if (!*(void**)search_start) {
630 search_start = (void*)ALIGN_UP ((mword)search_start + sizeof (gpointer));
631 /* The loop condition makes sure we don't overrun addr. */
635 canarified_obj_size = obj_size = ALIGN_UP (safe_object_get_size ((GCObject*)search_start));
638 * Filler arrays are marked by an invalid sync word. We don't
639 * consider them for pinning. They are not delimited by canaries,
642 if (!sgen_client_object_is_array_fill ((GCObject*)search_start)) {
643 CHECK_CANARY_FOR_OBJECT (search_start, TRUE);
644 CANARIFY_SIZE (canarified_obj_size);
646 if (addr >= search_start && (char*)addr < (char*)search_start + obj_size) {
647 /* This is the object we're looking for. */
648 obj_to_pin = (GCObject*)search_start;
649 obj_to_pin_size = canarified_obj_size;
654 /* Skip to the next object */
655 search_start = (void*)((char*)search_start + canarified_obj_size);
656 } while (search_start <= addr);
658 /* We've searched past the address we were looking for. */
660 pinning_front = search_start;
661 goto next_pin_queue_entry;
665 * We've found an object to pin. It might still be a dummy array, but we
666 * can advance the pinning front in any case.
668 pinning_front = (char*)obj_to_pin + obj_to_pin_size;
671 * If this is a dummy array marking the beginning of a nursery
672 * fragment, we don't pin it.
674 if (sgen_client_object_is_array_fill (obj_to_pin))
675 goto next_pin_queue_entry;
678 * Finally - pin the object!
680 desc = sgen_obj_get_descriptor_safe (obj_to_pin);
681 if (do_scan_objects) {
682 scan_func (obj_to_pin, desc, queue);
684 SGEN_LOG (4, "Pinned object %p, vtable %p (%s), count %d\n",
685 obj_to_pin, *(void**)obj_to_pin, sgen_client_vtable_get_name (SGEN_LOAD_VTABLE (obj_to_pin)), count);
686 binary_protocol_pin (obj_to_pin,
687 (gpointer)LOAD_VTABLE (obj_to_pin),
688 safe_object_get_size (obj_to_pin));
690 pin_object (obj_to_pin);
691 GRAY_OBJECT_ENQUEUE_SERIAL (queue, obj_to_pin, desc);
692 sgen_pin_stats_register_object (obj_to_pin, GENERATION_NURSERY);
693 definitely_pinned [count] = obj_to_pin;
696 if (concurrent_collection_in_progress)
697 sgen_pinning_register_pinned_in_nursery (obj_to_pin);
699 next_pin_queue_entry:
703 sgen_client_nursery_objects_pinned (definitely_pinned, count);
704 stat_pinned_objects += count;
709 pin_objects_in_nursery (gboolean do_scan_objects, ScanCopyContext ctx)
713 if (nursery_section->pin_queue_first_entry == nursery_section->pin_queue_last_entry)
716 reduced_to = pin_objects_from_nursery_pin_queue (do_scan_objects, ctx);
717 nursery_section->pin_queue_last_entry = nursery_section->pin_queue_first_entry + reduced_to;
721 * This function is only ever called (via `collector_pin_object()` in `sgen-copy-object.h`)
722 * when we can't promote an object because we're out of memory.
725 sgen_pin_object (GCObject *object, SgenGrayQueue *queue)
727 SGEN_ASSERT (0, sgen_ptr_in_nursery (object), "We're only supposed to use this for pinning nursery objects when out of memory.");
730 * All pinned objects are assumed to have been staged, so we need to stage as well.
731 * Also, the count of staged objects shows that "late pinning" happened.
733 sgen_pin_stage_ptr (object);
735 SGEN_PIN_OBJECT (object);
736 binary_protocol_pin (object, (gpointer)LOAD_VTABLE (object), safe_object_get_size (object));
739 sgen_pin_stats_register_object (object, GENERATION_NURSERY);
741 GRAY_OBJECT_ENQUEUE_SERIAL (queue, object, sgen_obj_get_descriptor_safe (object));
744 /* Sort the addresses in array in increasing order.
745 * Done using a by-the book heap sort. Which has decent and stable performance, is pretty cache efficient.
748 sgen_sort_addresses (void **array, size_t size)
753 for (i = 1; i < size; ++i) {
756 size_t parent = (child - 1) / 2;
758 if (array [parent] >= array [child])
761 tmp = array [parent];
762 array [parent] = array [child];
769 for (i = size - 1; i > 0; --i) {
772 array [i] = array [0];
778 while (root * 2 + 1 <= end) {
779 size_t child = root * 2 + 1;
781 if (child < end && array [child] < array [child + 1])
783 if (array [root] >= array [child])
787 array [root] = array [child];
796 * Scan the memory between start and end and queue values which could be pointers
797 * to the area between start_nursery and end_nursery for later consideration.
798 * Typically used for thread stacks.
801 sgen_conservatively_pin_objects_from (void **start, void **end, void *start_nursery, void *end_nursery, int pin_type)
805 SGEN_ASSERT (0, ((mword)start & (SIZEOF_VOID_P - 1)) == 0, "Why are we scanning for references in unaligned memory ?");
807 #if defined(VALGRIND_MAKE_MEM_DEFINED_IF_ADDRESSABLE) && !defined(_WIN64)
808 VALGRIND_MAKE_MEM_DEFINED_IF_ADDRESSABLE (start, (char*)end - (char*)start);
811 while (start < end) {
813 * *start can point to the middle of an object
814 * note: should we handle pointing at the end of an object?
815 * pinning in C# code disallows pointing at the end of an object
816 * but there is some small chance that an optimizing C compiler
817 * may keep the only reference to an object by pointing
818 * at the end of it. We ignore this small chance for now.
819 * Pointers to the end of an object are indistinguishable
820 * from pointers to the start of the next object in memory
821 * so if we allow that we'd need to pin two objects...
822 * We queue the pointer in an array, the
823 * array will then be sorted and uniqued. This way
824 * we can coalesce several pinning pointers and it should
825 * be faster since we'd do a memory scan with increasing
826 * addresses. Note: we can align the address to the allocation
827 * alignment, so the unique process is more effective.
829 mword addr = (mword)*start;
830 addr &= ~(ALLOC_ALIGN - 1);
831 if (addr >= (mword)start_nursery && addr < (mword)end_nursery) {
832 SGEN_LOG (6, "Pinning address %p from %p", (void*)addr, start);
833 sgen_pin_stage_ptr ((void*)addr);
834 binary_protocol_pin_stage (start, (void*)addr);
835 sgen_pin_stats_register_address ((char*)addr, pin_type);
841 SGEN_LOG (7, "found %d potential pinned heap pointers", count);
845 * The first thing we do in a collection is to identify pinned objects.
846 * This function considers all the areas of memory that need to be
847 * conservatively scanned.
850 pin_from_roots (void *start_nursery, void *end_nursery, ScanCopyContext ctx)
854 SGEN_LOG (2, "Scanning pinned roots (%d bytes, %d/%d entries)", (int)roots_size, roots_hash [ROOT_TYPE_NORMAL].num_entries, roots_hash [ROOT_TYPE_PINNED].num_entries);
855 /* objects pinned from the API are inside these roots */
856 SGEN_HASH_TABLE_FOREACH (&roots_hash [ROOT_TYPE_PINNED], void **, start_root, RootRecord *, root) {
857 SGEN_LOG (6, "Pinned roots %p-%p", start_root, root->end_root);
858 sgen_conservatively_pin_objects_from (start_root, (void**)root->end_root, start_nursery, end_nursery, PIN_TYPE_OTHER);
859 } SGEN_HASH_TABLE_FOREACH_END;
860 /* now deal with the thread stacks
861 * in the future we should be able to conservatively scan only:
862 * *) the cpu registers
863 * *) the unmanaged stack frames
864 * *) the _last_ managed stack frame
865 * *) pointers slots in managed frames
867 sgen_client_scan_thread_data (start_nursery, end_nursery, FALSE, ctx);
871 single_arg_user_copy_or_mark (GCObject **obj, void *gc_data)
873 ScanCopyContext *ctx = (ScanCopyContext *)gc_data;
874 ctx->ops->copy_or_mark_object (obj, ctx->queue);
878 * The memory area from start_root to end_root contains pointers to objects.
879 * Their position is precisely described by @desc (this means that the pointer
880 * can be either NULL or the pointer to the start of an object).
881 * This functions copies them to to_space updates them.
883 * This function is not thread-safe!
886 precisely_scan_objects_from (void** start_root, void** end_root, char* n_start, char *n_end, SgenDescriptor desc, ScanCopyContext ctx)
888 CopyOrMarkObjectFunc copy_func = ctx.ops->copy_or_mark_object;
889 ScanPtrFieldFunc scan_field_func = ctx.ops->scan_ptr_field;
890 SgenGrayQueue *queue = ctx.queue;
892 switch (desc & ROOT_DESC_TYPE_MASK) {
893 case ROOT_DESC_BITMAP:
894 desc >>= ROOT_DESC_TYPE_SHIFT;
896 if ((desc & 1) && *start_root) {
897 copy_func ((GCObject**)start_root, queue);
898 SGEN_LOG (9, "Overwrote root at %p with %p", start_root, *start_root);
904 case ROOT_DESC_COMPLEX: {
905 gsize *bitmap_data = (gsize *)sgen_get_complex_descriptor_bitmap (desc);
906 gsize bwords = (*bitmap_data) - 1;
907 void **start_run = start_root;
909 while (bwords-- > 0) {
910 gsize bmap = *bitmap_data++;
911 void **objptr = start_run;
913 if ((bmap & 1) && *objptr) {
914 copy_func ((GCObject**)objptr, queue);
915 SGEN_LOG (9, "Overwrote root at %p with %p", objptr, *objptr);
920 start_run += GC_BITS_PER_WORD;
924 case ROOT_DESC_VECTOR: {
927 for (p = start_root; p < end_root; p++) {
929 scan_field_func (NULL, (GCObject**)p, queue);
933 case ROOT_DESC_USER: {
934 SgenUserRootMarkFunc marker = sgen_get_user_descriptor_func (desc);
935 marker (start_root, single_arg_user_copy_or_mark, &ctx);
938 case ROOT_DESC_RUN_LEN:
939 g_assert_not_reached ();
941 g_assert_not_reached ();
946 reset_heap_boundaries (void)
948 lowest_heap_address = ~(mword)0;
949 highest_heap_address = 0;
953 sgen_update_heap_boundaries (mword low, mword high)
958 old = lowest_heap_address;
961 } while (SGEN_CAS_PTR ((gpointer*)&lowest_heap_address, (gpointer)low, (gpointer)old) != (gpointer)old);
964 old = highest_heap_address;
967 } while (SGEN_CAS_PTR ((gpointer*)&highest_heap_address, (gpointer)high, (gpointer)old) != (gpointer)old);
971 * Allocate and setup the data structures needed to be able to allocate objects
972 * in the nursery. The nursery is stored in nursery_section.
975 alloc_nursery (gboolean dynamic, size_t min_size, size_t max_size)
982 min_size = SGEN_DEFAULT_NURSERY_MIN_SIZE;
984 max_size = SGEN_DEFAULT_NURSERY_MAX_SIZE;
986 SGEN_ASSERT (0, min_size == max_size, "We can't have nursery ranges for static configuration.");
988 min_size = max_size = SGEN_DEFAULT_NURSERY_SIZE;
991 SGEN_ASSERT (0, !nursery_section, "Why are we allocating the nursery twice?");
992 SGEN_LOG (2, "Allocating nursery size: %zu, initial %zu", max_size, min_size);
994 /* FIXME: handle OOM */
995 nursery_section = (GCMemSection *)sgen_alloc_internal (INTERNAL_MEM_SECTION);
997 /* If there isn't enough space even for the nursery we should simply abort. */
998 g_assert (sgen_memgov_try_alloc_space (max_size, SPACE_NURSERY));
1001 * The nursery section range represents the memory section where objects
1002 * can be found. This is used when iterating for objects in the nursery,
1003 * pinning etc. sgen_nursery_max_size represents the total allocated space
1004 * for the nursery. sgen_nursery_size represents the current size of the
1005 * nursery and it is used for allocation limits, heuristics etc. The
1006 * nursery section is not always identical to the current nursery size
1007 * because it can contain pinned objects from when the nursery was larger.
1009 * sgen_nursery_size <= nursery_section size <= sgen_nursery_max_size
1011 data = (char *)major_collector.alloc_heap (max_size, max_size);
1012 sgen_update_heap_boundaries ((mword)data, (mword)(data + max_size));
1013 nursery_section->data = data;
1014 nursery_section->end_data = data + min_size;
1015 scan_starts = (max_size + SCAN_START_SIZE - 1) / SCAN_START_SIZE;
1016 nursery_section->scan_starts = (char **)sgen_alloc_internal_dynamic (sizeof (char*) * scan_starts, INTERNAL_MEM_SCAN_STARTS, TRUE);
1017 nursery_section->num_scan_start = scan_starts;
1019 sgen_nursery_allocator_set_nursery_bounds (data, min_size, max_size);
1023 mono_gc_get_logfile (void)
1025 return gc_debug_file;
1029 mono_gc_params_set (const char* options)
1031 if (gc_params_options)
1032 g_free (gc_params_options);
1034 gc_params_options = g_strdup (options);
1038 mono_gc_debug_set (const char* options)
1040 if (gc_debug_options)
1041 g_free (gc_debug_options);
1043 gc_debug_options = g_strdup (options);
1047 scan_finalizer_entries (SgenPointerQueue *fin_queue, ScanCopyContext ctx)
1049 CopyOrMarkObjectFunc copy_func = ctx.ops->copy_or_mark_object;
1050 SgenGrayQueue *queue = ctx.queue;
1053 for (i = 0; i < fin_queue->next_slot; ++i) {
1054 GCObject *obj = (GCObject *)fin_queue->data [i];
1057 SGEN_LOG (5, "Scan of fin ready object: %p (%s)\n", obj, sgen_client_vtable_get_name (SGEN_LOAD_VTABLE (obj)));
1058 copy_func ((GCObject**)&fin_queue->data [i], queue);
1063 generation_name (int generation)
1065 switch (generation) {
1066 case GENERATION_NURSERY: return "nursery";
1067 case GENERATION_OLD: return "old";
1068 default: g_assert_not_reached ();
1073 sgen_generation_name (int generation)
1075 return generation_name (generation);
1079 finish_gray_stack (int generation, ScanCopyContext ctx)
1083 int done_with_ephemerons, ephemeron_rounds = 0;
1084 char *start_addr = generation == GENERATION_NURSERY ? sgen_get_nursery_start () : NULL;
1085 char *end_addr = generation == GENERATION_NURSERY ? sgen_get_nursery_end () : (char*)-1;
1086 SgenGrayQueue *queue = ctx.queue;
1088 binary_protocol_finish_gray_stack_start (sgen_timestamp (), generation);
1090 * We copied all the reachable objects. Now it's the time to copy
1091 * the objects that were not referenced by the roots, but by the copied objects.
1092 * we built a stack of objects pointed to by gray_start: they are
1093 * additional roots and we may add more items as we go.
1094 * We loop until gray_start == gray_objects which means no more objects have
1095 * been added. Note this is iterative: no recursion is involved.
1096 * We need to walk the LO list as well in search of marked big objects
1097 * (use a flag since this is needed only on major collections). We need to loop
1098 * here as well, so keep a counter of marked LO (increasing it in copy_object).
1099 * To achieve better cache locality and cache usage, we drain the gray stack
1100 * frequently, after each object is copied, and just finish the work here.
1102 sgen_drain_gray_stack (ctx);
1104 SGEN_LOG (2, "%s generation done", generation_name (generation));
1107 Reset bridge data, we might have lingering data from a previous collection if this is a major
1108 collection trigged by minor overflow.
1110 We must reset the gathered bridges since their original block might be evacuated due to major
1111 fragmentation in the meanwhile and the bridge code should not have to deal with that.
1113 if (sgen_client_bridge_need_processing ())
1114 sgen_client_bridge_reset_data ();
1117 * Mark all strong toggleref objects. This must be done before we walk ephemerons or finalizers
1118 * to ensure they see the full set of live objects.
1120 sgen_client_mark_togglerefs (start_addr, end_addr, ctx);
1123 * Walk the ephemeron tables marking all values with reachable keys. This must be completely done
1124 * before processing finalizable objects and non-tracking weak links to avoid finalizing/clearing
1125 * objects that are in fact reachable.
1127 done_with_ephemerons = 0;
1129 done_with_ephemerons = sgen_client_mark_ephemerons (ctx);
1130 sgen_drain_gray_stack (ctx);
1132 } while (!done_with_ephemerons);
1134 if (sgen_client_bridge_need_processing ()) {
1135 /*Make sure the gray stack is empty before we process bridge objects so we get liveness right*/
1136 sgen_drain_gray_stack (ctx);
1137 sgen_collect_bridge_objects (generation, ctx);
1138 if (generation == GENERATION_OLD)
1139 sgen_collect_bridge_objects (GENERATION_NURSERY, ctx);
1142 Do the first bridge step here, as the collector liveness state will become useless after that.
1144 An important optimization is to only proccess the possibly dead part of the object graph and skip
1145 over all live objects as we transitively know everything they point must be alive too.
1147 The above invariant is completely wrong if we let the gray queue be drained and mark/copy everything.
1149 This has the unfortunate side effect of making overflow collections perform the first step twice, but
1150 given we now have heuristics that perform major GC in anticipation of minor overflows this should not
1153 sgen_client_bridge_processing_stw_step ();
1157 Make sure we drain the gray stack before processing disappearing links and finalizers.
1158 If we don't make sure it is empty we might wrongly see a live object as dead.
1160 sgen_drain_gray_stack (ctx);
1163 We must clear weak links that don't track resurrection before processing object ready for
1164 finalization so they can be cleared before that.
1166 sgen_null_link_in_range (generation, ctx, FALSE);
1167 if (generation == GENERATION_OLD)
1168 sgen_null_link_in_range (GENERATION_NURSERY, ctx, FALSE);
1171 /* walk the finalization queue and move also the objects that need to be
1172 * finalized: use the finalized objects as new roots so the objects they depend
1173 * on are also not reclaimed. As with the roots above, only objects in the nursery
1174 * are marked/copied.
1176 sgen_finalize_in_range (generation, ctx);
1177 if (generation == GENERATION_OLD)
1178 sgen_finalize_in_range (GENERATION_NURSERY, ctx);
1179 /* drain the new stack that might have been created */
1180 SGEN_LOG (6, "Precise scan of gray area post fin");
1181 sgen_drain_gray_stack (ctx);
1184 * This must be done again after processing finalizable objects since CWL slots are cleared only after the key is finalized.
1186 done_with_ephemerons = 0;
1188 done_with_ephemerons = sgen_client_mark_ephemerons (ctx);
1189 sgen_drain_gray_stack (ctx);
1191 } while (!done_with_ephemerons);
1193 sgen_client_clear_unreachable_ephemerons (ctx);
1196 * We clear togglerefs only after all possible chances of revival are done.
1197 * This is semantically more inline with what users expect and it allows for
1198 * user finalizers to correctly interact with TR objects.
1200 sgen_client_clear_togglerefs (start_addr, end_addr, ctx);
1203 SGEN_LOG (2, "Finalize queue handling scan for %s generation: %lld usecs %d ephemeron rounds", generation_name (generation), (long long)TV_ELAPSED (atv, btv), ephemeron_rounds);
1206 * handle disappearing links
1207 * Note we do this after checking the finalization queue because if an object
1208 * survives (at least long enough to be finalized) we don't clear the link.
1209 * This also deals with a possible issue with the monitor reclamation: with the Boehm
1210 * GC a finalized object my lose the monitor because it is cleared before the finalizer is
1213 g_assert (sgen_gray_object_queue_is_empty (queue));
1215 sgen_null_link_in_range (generation, ctx, TRUE);
1216 if (generation == GENERATION_OLD)
1217 sgen_null_link_in_range (GENERATION_NURSERY, ctx, TRUE);
1218 if (sgen_gray_object_queue_is_empty (queue))
1220 sgen_drain_gray_stack (ctx);
1223 g_assert (sgen_gray_object_queue_is_empty (queue));
1225 binary_protocol_finish_gray_stack_end (sgen_timestamp (), generation);
1229 sgen_check_section_scan_starts (GCMemSection *section)
1232 for (i = 0; i < section->num_scan_start; ++i) {
1233 if (section->scan_starts [i]) {
1234 mword size = safe_object_get_size ((GCObject*) section->scan_starts [i]);
1235 SGEN_ASSERT (0, size >= SGEN_CLIENT_MINIMUM_OBJECT_SIZE && size <= MAX_SMALL_OBJ_SIZE, "Weird object size at scan starts.");
1241 check_scan_starts (void)
1243 if (!do_scan_starts_check)
1245 sgen_check_section_scan_starts (nursery_section);
1246 major_collector.check_scan_starts ();
1250 scan_from_registered_roots (char *addr_start, char *addr_end, int root_type, ScanCopyContext ctx)
1254 SGEN_HASH_TABLE_FOREACH (&roots_hash [root_type], void **, start_root, RootRecord *, root) {
1255 SGEN_LOG (6, "Precise root scan %p-%p (desc: %p)", start_root, root->end_root, (void*)root->root_desc);
1256 precisely_scan_objects_from (start_root, (void**)root->end_root, addr_start, addr_end, root->root_desc, ctx);
1257 } SGEN_HASH_TABLE_FOREACH_END;
1263 static gboolean inited = FALSE;
1268 mono_counters_register ("Collection max time", MONO_COUNTER_GC | MONO_COUNTER_ULONG | MONO_COUNTER_TIME | MONO_COUNTER_MONOTONIC, &time_max);
1270 mono_counters_register ("Minor fragment clear", MONO_COUNTER_GC | MONO_COUNTER_ULONG | MONO_COUNTER_TIME, &time_minor_pre_collection_fragment_clear);
1271 mono_counters_register ("Minor pinning", MONO_COUNTER_GC | MONO_COUNTER_ULONG | MONO_COUNTER_TIME, &time_minor_pinning);
1272 mono_counters_register ("Minor scan remembered set", MONO_COUNTER_GC | MONO_COUNTER_ULONG | MONO_COUNTER_TIME, &time_minor_scan_remsets);
1273 mono_counters_register ("Minor scan major blocks", MONO_COUNTER_GC | MONO_COUNTER_ULONG | MONO_COUNTER_TIME, &time_minor_scan_major_blocks);
1274 mono_counters_register ("Minor scan los", MONO_COUNTER_GC | MONO_COUNTER_ULONG | MONO_COUNTER_TIME, &time_minor_scan_los);
1275 mono_counters_register ("Minor scan pinned", MONO_COUNTER_GC | MONO_COUNTER_ULONG | MONO_COUNTER_TIME, &time_minor_scan_pinned);
1276 mono_counters_register ("Minor scan roots", MONO_COUNTER_GC | MONO_COUNTER_ULONG | MONO_COUNTER_TIME, &time_minor_scan_roots);
1277 mono_counters_register ("Minor fragment creation", MONO_COUNTER_GC | MONO_COUNTER_ULONG | MONO_COUNTER_TIME, &time_minor_fragment_creation);
1279 mono_counters_register ("Major fragment clear", MONO_COUNTER_GC | MONO_COUNTER_ULONG | MONO_COUNTER_TIME, &time_major_pre_collection_fragment_clear);
1280 mono_counters_register ("Major pinning", MONO_COUNTER_GC | MONO_COUNTER_ULONG | MONO_COUNTER_TIME, &time_major_pinning);
1281 mono_counters_register ("Major scan pinned", MONO_COUNTER_GC | MONO_COUNTER_ULONG | MONO_COUNTER_TIME, &time_major_scan_pinned);
1282 mono_counters_register ("Major scan roots", MONO_COUNTER_GC | MONO_COUNTER_ULONG | MONO_COUNTER_TIME, &time_major_scan_roots);
1283 mono_counters_register ("Major scan mod union blocks", MONO_COUNTER_GC | MONO_COUNTER_ULONG | MONO_COUNTER_TIME, &time_major_scan_mod_union_blocks);
1284 mono_counters_register ("Major scan mod union los", MONO_COUNTER_GC | MONO_COUNTER_ULONG | MONO_COUNTER_TIME, &time_major_scan_mod_union_los);
1285 mono_counters_register ("Major finish gray stack", MONO_COUNTER_GC | MONO_COUNTER_ULONG | MONO_COUNTER_TIME, &time_major_finish_gray_stack);
1286 mono_counters_register ("Major free big objects", MONO_COUNTER_GC | MONO_COUNTER_ULONG | MONO_COUNTER_TIME, &time_major_free_bigobjs);
1287 mono_counters_register ("Major LOS sweep", MONO_COUNTER_GC | MONO_COUNTER_ULONG | MONO_COUNTER_TIME, &time_major_los_sweep);
1288 mono_counters_register ("Major sweep", MONO_COUNTER_GC | MONO_COUNTER_ULONG | MONO_COUNTER_TIME, &time_major_sweep);
1289 mono_counters_register ("Major fragment creation", MONO_COUNTER_GC | MONO_COUNTER_ULONG | MONO_COUNTER_TIME, &time_major_fragment_creation);
1291 mono_counters_register ("Number of pinned objects", MONO_COUNTER_GC | MONO_COUNTER_ULONG, &stat_pinned_objects);
1293 #ifdef HEAVY_STATISTICS
1294 mono_counters_register ("WBarrier remember pointer", MONO_COUNTER_GC | MONO_COUNTER_ULONG, &stat_wbarrier_add_to_global_remset);
1295 mono_counters_register ("WBarrier arrayref copy", MONO_COUNTER_GC | MONO_COUNTER_ULONG, &stat_wbarrier_arrayref_copy);
1296 mono_counters_register ("WBarrier generic store called", MONO_COUNTER_GC | MONO_COUNTER_ULONG, &stat_wbarrier_generic_store);
1297 mono_counters_register ("WBarrier generic atomic store called", MONO_COUNTER_GC | MONO_COUNTER_ULONG, &stat_wbarrier_generic_store_atomic);
1298 mono_counters_register ("WBarrier set root", MONO_COUNTER_GC | MONO_COUNTER_ULONG, &stat_wbarrier_set_root);
1300 mono_counters_register ("# objects allocated degraded", MONO_COUNTER_GC | MONO_COUNTER_ULONG, &stat_objects_alloced_degraded);
1301 mono_counters_register ("bytes allocated degraded", MONO_COUNTER_GC | MONO_COUNTER_ULONG, &stat_bytes_alloced_degraded);
1303 mono_counters_register ("# copy_object() called (nursery)", MONO_COUNTER_GC | MONO_COUNTER_ULONG, &stat_copy_object_called_nursery);
1304 mono_counters_register ("# objects copied (nursery)", MONO_COUNTER_GC | MONO_COUNTER_ULONG, &stat_objects_copied_nursery);
1305 mono_counters_register ("# copy_object() called (major)", MONO_COUNTER_GC | MONO_COUNTER_ULONG, &stat_copy_object_called_major);
1306 mono_counters_register ("# objects copied (major)", MONO_COUNTER_GC | MONO_COUNTER_ULONG, &stat_objects_copied_major);
1308 mono_counters_register ("# scan_object() called (nursery)", MONO_COUNTER_GC | MONO_COUNTER_ULONG, &stat_scan_object_called_nursery);
1309 mono_counters_register ("# scan_object() called (major)", MONO_COUNTER_GC | MONO_COUNTER_ULONG, &stat_scan_object_called_major);
1311 mono_counters_register ("Slots allocated in vain", MONO_COUNTER_GC | MONO_COUNTER_ULONG, &stat_slots_allocated_in_vain);
1313 mono_counters_register ("# nursery copy_object() failed from space", MONO_COUNTER_GC | MONO_COUNTER_ULONG, &stat_nursery_copy_object_failed_from_space);
1314 mono_counters_register ("# nursery copy_object() failed forwarded", MONO_COUNTER_GC | MONO_COUNTER_ULONG, &stat_nursery_copy_object_failed_forwarded);
1315 mono_counters_register ("# nursery copy_object() failed pinned", MONO_COUNTER_GC | MONO_COUNTER_ULONG, &stat_nursery_copy_object_failed_pinned);
1316 mono_counters_register ("# nursery copy_object() failed to space", MONO_COUNTER_GC | MONO_COUNTER_ULONG, &stat_nursery_copy_object_failed_to_space);
1318 sgen_nursery_allocator_init_heavy_stats ();
1326 reset_pinned_from_failed_allocation (void)
1328 bytes_pinned_from_failed_allocation = 0;
1332 sgen_set_pinned_from_failed_allocation (mword objsize)
1334 bytes_pinned_from_failed_allocation += objsize;
1338 sgen_collection_is_concurrent (void)
1340 switch (current_collection_generation) {
1341 case GENERATION_NURSERY:
1343 case GENERATION_OLD:
1344 return concurrent_collection_in_progress;
1346 g_error ("Invalid current generation %d", current_collection_generation);
1352 sgen_concurrent_collection_in_progress (void)
1354 return concurrent_collection_in_progress;
1358 SgenThreadPoolJob job;
1359 SgenObjectOperations *ops;
1360 SgenGrayQueue *gc_thread_gray_queue;
1365 int job_index, job_split_count;
1368 static ScanCopyContext
1369 scan_copy_context_for_scan_job (void *worker_data_untyped, ScanJob *job)
1371 WorkerData *worker_data = (WorkerData *)worker_data_untyped;
1375 * For jobs enqueued on workers we set the ops at job runtime in order
1376 * to be able to profit from on the fly optimized object ops or other
1377 * object ops changes, like forced concurrent finish.
1379 SGEN_ASSERT (0, sgen_workers_is_worker_thread (mono_native_thread_id_get ()), "We need a context for the scan job");
1380 job->ops = sgen_workers_get_idle_func_object_ops (worker_data);
1383 return CONTEXT_FROM_OBJECT_OPERATIONS (job->ops, sgen_workers_get_job_gray_queue (worker_data, job->gc_thread_gray_queue));
1391 } ScanFromRegisteredRootsJob;
1394 job_scan_from_registered_roots (void *worker_data_untyped, SgenThreadPoolJob *job)
1396 ScanFromRegisteredRootsJob *job_data = (ScanFromRegisteredRootsJob*)job;
1397 ScanCopyContext ctx = scan_copy_context_for_scan_job (worker_data_untyped, &job_data->scan_job);
1399 scan_from_registered_roots (job_data->heap_start, job_data->heap_end, job_data->root_type, ctx);
1406 } ScanThreadDataJob;
1409 job_scan_thread_data (void *worker_data_untyped, SgenThreadPoolJob *job)
1411 ScanThreadDataJob *job_data = (ScanThreadDataJob*)job;
1412 ScanCopyContext ctx = scan_copy_context_for_scan_job (worker_data_untyped, &job_data->scan_job);
1414 sgen_client_scan_thread_data (job_data->heap_start, job_data->heap_end, TRUE, ctx);
1419 SgenPointerQueue *queue;
1420 } ScanFinalizerEntriesJob;
1423 job_scan_finalizer_entries (void *worker_data_untyped, SgenThreadPoolJob *job)
1425 ScanFinalizerEntriesJob *job_data = (ScanFinalizerEntriesJob*)job;
1426 ScanCopyContext ctx = scan_copy_context_for_scan_job (worker_data_untyped, &job_data->scan_job);
1428 scan_finalizer_entries (job_data->queue, ctx);
1432 job_scan_wbroots (void *worker_data_untyped, SgenThreadPoolJob *job)
1434 ScanJob *job_data = (ScanJob*)job;
1435 ScanCopyContext ctx = scan_copy_context_for_scan_job (worker_data_untyped, job_data);
1437 sgen_wbroots_scan_card_table (ctx);
1441 job_scan_major_card_table (void *worker_data_untyped, SgenThreadPoolJob *job)
1443 SGEN_TV_DECLARE (atv);
1444 SGEN_TV_DECLARE (btv);
1445 ParallelScanJob *job_data = (ParallelScanJob*)job;
1446 ScanCopyContext ctx = scan_copy_context_for_scan_job (worker_data_untyped, (ScanJob*)job_data);
1448 SGEN_TV_GETTIME (atv);
1449 major_collector.scan_card_table (CARDTABLE_SCAN_GLOBAL, ctx, job_data->job_index, job_data->job_split_count);
1450 SGEN_TV_GETTIME (btv);
1451 time_minor_scan_major_blocks += SGEN_TV_ELAPSED (atv, btv);
1455 job_scan_los_card_table (void *worker_data_untyped, SgenThreadPoolJob *job)
1457 SGEN_TV_DECLARE (atv);
1458 SGEN_TV_DECLARE (btv);
1459 ParallelScanJob *job_data = (ParallelScanJob*)job;
1460 ScanCopyContext ctx = scan_copy_context_for_scan_job (worker_data_untyped, (ScanJob*)job_data);
1462 SGEN_TV_GETTIME (atv);
1463 sgen_los_scan_card_table (CARDTABLE_SCAN_GLOBAL, ctx, job_data->job_index, job_data->job_split_count);
1464 SGEN_TV_GETTIME (btv);
1465 time_minor_scan_los += SGEN_TV_ELAPSED (atv, btv);
1469 job_scan_major_mod_union_card_table (void *worker_data_untyped, SgenThreadPoolJob *job)
1471 SGEN_TV_DECLARE (atv);
1472 SGEN_TV_DECLARE (btv);
1473 ParallelScanJob *job_data = (ParallelScanJob*)job;
1474 ScanCopyContext ctx = scan_copy_context_for_scan_job (worker_data_untyped, (ScanJob*)job_data);
1476 g_assert (concurrent_collection_in_progress);
1477 SGEN_TV_GETTIME (atv);
1478 major_collector.scan_card_table (CARDTABLE_SCAN_MOD_UNION, ctx, job_data->job_index, job_data->job_split_count);
1479 SGEN_TV_GETTIME (btv);
1480 time_major_scan_mod_union_blocks += SGEN_TV_ELAPSED (atv, btv);
1484 job_scan_los_mod_union_card_table (void *worker_data_untyped, SgenThreadPoolJob *job)
1486 SGEN_TV_DECLARE (atv);
1487 SGEN_TV_DECLARE (btv);
1488 ParallelScanJob *job_data = (ParallelScanJob*)job;
1489 ScanCopyContext ctx = scan_copy_context_for_scan_job (worker_data_untyped, (ScanJob*)job_data);
1491 g_assert (concurrent_collection_in_progress);
1492 SGEN_TV_GETTIME (atv);
1493 sgen_los_scan_card_table (CARDTABLE_SCAN_MOD_UNION, ctx, job_data->job_index, job_data->job_split_count);
1494 SGEN_TV_GETTIME (btv);
1495 time_major_scan_mod_union_los += SGEN_TV_ELAPSED (atv, btv);
1499 job_major_mod_union_preclean (void *worker_data_untyped, SgenThreadPoolJob *job)
1501 ParallelScanJob *job_data = (ParallelScanJob*)job;
1502 ScanCopyContext ctx = scan_copy_context_for_scan_job (worker_data_untyped, (ScanJob*)job_data);
1504 g_assert (concurrent_collection_in_progress);
1506 major_collector.scan_card_table (CARDTABLE_SCAN_MOD_UNION_PRECLEAN, ctx, job_data->job_index, job_data->job_split_count);
1510 job_los_mod_union_preclean (void *worker_data_untyped, SgenThreadPoolJob *job)
1512 ParallelScanJob *job_data = (ParallelScanJob*)job;
1513 ScanCopyContext ctx = scan_copy_context_for_scan_job (worker_data_untyped, (ScanJob*)job_data);
1515 g_assert (concurrent_collection_in_progress);
1517 sgen_los_scan_card_table (CARDTABLE_SCAN_MOD_UNION_PRECLEAN, ctx, job_data->job_index, job_data->job_split_count);
1521 job_scan_last_pinned (void *worker_data_untyped, SgenThreadPoolJob *job)
1523 ScanJob *job_data = (ScanJob*)job;
1524 ScanCopyContext ctx = scan_copy_context_for_scan_job (worker_data_untyped, job_data);
1526 g_assert (concurrent_collection_in_progress);
1528 sgen_scan_pin_queue_objects (ctx);
1532 workers_finish_callback (void)
1534 ParallelScanJob *psj;
1536 int split_count = sgen_workers_get_job_split_count (GENERATION_OLD);
1538 /* Mod union preclean jobs */
1539 for (i = 0; i < split_count; i++) {
1540 psj = (ParallelScanJob*)sgen_thread_pool_job_alloc ("preclean major mod union cardtable", job_major_mod_union_preclean, sizeof (ParallelScanJob));
1541 psj->scan_job.gc_thread_gray_queue = NULL;
1543 psj->job_split_count = split_count;
1544 sgen_workers_enqueue_job (GENERATION_OLD, &psj->scan_job.job, TRUE);
1547 for (i = 0; i < split_count; i++) {
1548 psj = (ParallelScanJob*)sgen_thread_pool_job_alloc ("preclean los mod union cardtable", job_los_mod_union_preclean, sizeof (ParallelScanJob));
1549 psj->scan_job.gc_thread_gray_queue = NULL;
1551 psj->job_split_count = split_count;
1552 sgen_workers_enqueue_job (GENERATION_OLD, &psj->scan_job.job, TRUE);
1555 sj = (ScanJob*)sgen_thread_pool_job_alloc ("scan last pinned", job_scan_last_pinned, sizeof (ScanJob));
1556 sj->gc_thread_gray_queue = NULL;
1557 sgen_workers_enqueue_job (GENERATION_OLD, &sj->job, TRUE);
1561 init_gray_queue (SgenGrayQueue *gc_thread_gray_queue)
1563 sgen_gray_object_queue_init (gc_thread_gray_queue, NULL, TRUE);
1567 enqueue_scan_remembered_set_jobs (SgenGrayQueue *gc_thread_gray_queue, SgenObjectOperations *ops, gboolean enqueue)
1569 int i, split_count = sgen_workers_get_job_split_count (GENERATION_NURSERY);
1572 sj = (ScanJob*)sgen_thread_pool_job_alloc ("scan wbroots", job_scan_wbroots, sizeof (ScanJob));
1574 sj->gc_thread_gray_queue = gc_thread_gray_queue;
1575 sgen_workers_enqueue_job (GENERATION_NURSERY, &sj->job, enqueue);
1577 for (i = 0; i < split_count; i++) {
1578 ParallelScanJob *psj;
1580 psj = (ParallelScanJob*)sgen_thread_pool_job_alloc ("scan major remsets", job_scan_major_card_table, sizeof (ParallelScanJob));
1581 psj->scan_job.ops = ops;
1582 psj->scan_job.gc_thread_gray_queue = gc_thread_gray_queue;
1584 psj->job_split_count = split_count;
1585 sgen_workers_enqueue_job (GENERATION_NURSERY, &psj->scan_job.job, enqueue);
1587 psj = (ParallelScanJob*)sgen_thread_pool_job_alloc ("scan LOS remsets", job_scan_los_card_table, sizeof (ParallelScanJob));
1588 psj->scan_job.ops = ops;
1589 psj->scan_job.gc_thread_gray_queue = gc_thread_gray_queue;
1591 psj->job_split_count = split_count;
1592 sgen_workers_enqueue_job (GENERATION_NURSERY, &psj->scan_job.job, enqueue);
1597 enqueue_scan_from_roots_jobs (SgenGrayQueue *gc_thread_gray_queue, char *heap_start, char *heap_end, SgenObjectOperations *ops, gboolean enqueue)
1599 ScanFromRegisteredRootsJob *scrrj;
1600 ScanThreadDataJob *stdj;
1601 ScanFinalizerEntriesJob *sfej;
1603 /* registered roots, this includes static fields */
1605 scrrj = (ScanFromRegisteredRootsJob*)sgen_thread_pool_job_alloc ("scan from registered roots normal", job_scan_from_registered_roots, sizeof (ScanFromRegisteredRootsJob));
1606 scrrj->scan_job.ops = ops;
1607 scrrj->scan_job.gc_thread_gray_queue = gc_thread_gray_queue;
1608 scrrj->heap_start = heap_start;
1609 scrrj->heap_end = heap_end;
1610 scrrj->root_type = ROOT_TYPE_NORMAL;
1611 sgen_workers_enqueue_job (current_collection_generation, &scrrj->scan_job.job, enqueue);
1613 if (current_collection_generation == GENERATION_OLD) {
1614 /* During minors we scan the cardtable for these roots instead */
1615 scrrj = (ScanFromRegisteredRootsJob*)sgen_thread_pool_job_alloc ("scan from registered roots wbarrier", job_scan_from_registered_roots, sizeof (ScanFromRegisteredRootsJob));
1616 scrrj->scan_job.ops = ops;
1617 scrrj->scan_job.gc_thread_gray_queue = gc_thread_gray_queue;
1618 scrrj->heap_start = heap_start;
1619 scrrj->heap_end = heap_end;
1620 scrrj->root_type = ROOT_TYPE_WBARRIER;
1621 sgen_workers_enqueue_job (current_collection_generation, &scrrj->scan_job.job, enqueue);
1626 stdj = (ScanThreadDataJob*)sgen_thread_pool_job_alloc ("scan thread data", job_scan_thread_data, sizeof (ScanThreadDataJob));
1627 stdj->scan_job.ops = ops;
1628 stdj->scan_job.gc_thread_gray_queue = gc_thread_gray_queue;
1629 stdj->heap_start = heap_start;
1630 stdj->heap_end = heap_end;
1631 sgen_workers_enqueue_job (current_collection_generation, &stdj->scan_job.job, enqueue);
1633 /* Scan the list of objects ready for finalization. */
1635 sfej = (ScanFinalizerEntriesJob*)sgen_thread_pool_job_alloc ("scan finalizer entries", job_scan_finalizer_entries, sizeof (ScanFinalizerEntriesJob));
1636 sfej->scan_job.ops = ops;
1637 sfej->scan_job.gc_thread_gray_queue = gc_thread_gray_queue;
1638 sfej->queue = &fin_ready_queue;
1639 sgen_workers_enqueue_job (current_collection_generation, &sfej->scan_job.job, enqueue);
1641 sfej = (ScanFinalizerEntriesJob*)sgen_thread_pool_job_alloc ("scan critical finalizer entries", job_scan_finalizer_entries, sizeof (ScanFinalizerEntriesJob));
1642 sfej->scan_job.ops = ops;
1643 sfej->scan_job.gc_thread_gray_queue = gc_thread_gray_queue;
1644 sfej->queue = &critical_fin_queue;
1645 sgen_workers_enqueue_job (current_collection_generation, &sfej->scan_job.job, enqueue);
1649 * Perform a nursery collection.
1651 * Return whether any objects were late-pinned due to being out of memory.
1654 collect_nursery (const char *reason, gboolean is_overflow, SgenGrayQueue *unpin_queue)
1656 gboolean needs_major, is_parallel = FALSE;
1657 mword fragment_total;
1658 SgenGrayQueue gc_thread_gray_queue;
1659 SgenObjectOperations *object_ops_nopar, *object_ops_par = NULL;
1660 ScanCopyContext ctx;
1663 SGEN_TV_DECLARE (last_minor_collection_start_tv);
1664 SGEN_TV_DECLARE (last_minor_collection_end_tv);
1666 if (disable_minor_collections)
1669 TV_GETTIME (last_minor_collection_start_tv);
1670 atv = last_minor_collection_start_tv;
1672 binary_protocol_collection_begin (gc_stats.minor_gc_count, GENERATION_NURSERY);
1674 object_ops_nopar = sgen_concurrent_collection_in_progress ()
1675 ? &sgen_minor_collector.serial_ops_with_concurrent_major
1676 : &sgen_minor_collector.serial_ops;
1677 if (sgen_minor_collector.is_parallel && sgen_nursery_size >= SGEN_PARALLEL_MINOR_MIN_NURSERY_SIZE) {
1678 object_ops_par = sgen_concurrent_collection_in_progress ()
1679 ? &sgen_minor_collector.parallel_ops_with_concurrent_major
1680 : &sgen_minor_collector.parallel_ops;
1684 if (do_verify_nursery || do_dump_nursery_content)
1685 sgen_debug_verify_nursery (do_dump_nursery_content);
1687 current_collection_generation = GENERATION_NURSERY;
1689 SGEN_ASSERT (0, !sgen_collection_is_concurrent (), "Why is the nursery collection concurrent?");
1691 reset_pinned_from_failed_allocation ();
1693 check_scan_starts ();
1695 sgen_nursery_alloc_prepare_for_minor ();
1700 SGEN_LOG (1, "Start nursery collection %d %p-%p, size: %d", gc_stats.minor_gc_count, nursery_section->data, nursery_section->end_data, (int)(nursery_section->end_data - nursery_section->data));
1702 /* world must be stopped already */
1704 time_minor_pre_collection_fragment_clear += TV_ELAPSED (atv, btv);
1706 sgen_client_pre_collection_checks ();
1708 major_collector.start_nursery_collection ();
1710 sgen_memgov_minor_collection_start ();
1712 init_gray_queue (&gc_thread_gray_queue);
1713 ctx = CONTEXT_FROM_OBJECT_OPERATIONS (object_ops_nopar, &gc_thread_gray_queue);
1715 gc_stats.minor_gc_count ++;
1717 sgen_process_fin_stage_entries ();
1719 /* pin from pinned handles */
1720 sgen_init_pinning ();
1721 if (concurrent_collection_in_progress)
1722 sgen_init_pinning_for_conc ();
1723 sgen_client_binary_protocol_mark_start (GENERATION_NURSERY);
1724 pin_from_roots (nursery_section->data, nursery_section->end_data, ctx);
1725 /* pin cemented objects */
1726 sgen_pin_cemented_objects ();
1727 /* identify pinned objects */
1728 sgen_optimize_pin_queue ();
1729 sgen_pinning_setup_section (nursery_section);
1731 pin_objects_in_nursery (FALSE, ctx);
1732 sgen_pinning_trim_queue_to_section (nursery_section);
1733 if (concurrent_collection_in_progress)
1734 sgen_finish_pinning_for_conc ();
1736 if (remset_consistency_checks)
1737 sgen_check_remset_consistency ();
1739 if (whole_heap_check_before_collection) {
1740 sgen_clear_nursery_fragments ();
1741 sgen_check_whole_heap (FALSE);
1745 time_minor_pinning += TV_ELAPSED (btv, atv);
1746 SGEN_LOG (2, "Finding pinned pointers: %zd in %lld usecs", sgen_get_pinned_count (), (long long)TV_ELAPSED (btv, atv));
1747 SGEN_LOG (4, "Start scan with %zd pinned objects", sgen_get_pinned_count ());
1749 remset.start_scan_remsets ();
1751 enqueue_scan_remembered_set_jobs (&gc_thread_gray_queue, is_parallel ? NULL : object_ops_nopar, is_parallel);
1753 /* we don't have complete write barrier yet, so we scan all the old generation sections */
1755 time_minor_scan_remsets += TV_ELAPSED (atv, btv);
1756 SGEN_LOG (2, "Old generation scan: %lld usecs", (long long)TV_ELAPSED (atv, btv));
1758 sgen_pin_stats_report ();
1760 /* FIXME: Why do we do this at this specific, seemingly random, point? */
1761 sgen_client_collecting_minor (&fin_ready_queue, &critical_fin_queue);
1764 time_minor_scan_pinned += TV_ELAPSED (btv, atv);
1766 enqueue_scan_from_roots_jobs (&gc_thread_gray_queue, nursery_section->data, nursery_section->end_data, is_parallel ? NULL : object_ops_nopar, is_parallel);
1769 gray_queue_redirect (&gc_thread_gray_queue);
1770 sgen_workers_start_all_workers (GENERATION_NURSERY, object_ops_nopar, object_ops_par, NULL);
1771 sgen_workers_join (GENERATION_NURSERY);
1775 time_minor_scan_roots += TV_ELAPSED (atv, btv);
1777 finish_gray_stack (GENERATION_NURSERY, ctx);
1780 time_minor_finish_gray_stack += TV_ELAPSED (btv, atv);
1781 sgen_client_binary_protocol_mark_end (GENERATION_NURSERY);
1783 if (objects_pinned) {
1784 sgen_optimize_pin_queue ();
1785 sgen_pinning_setup_section (nursery_section);
1789 * This is the latest point at which we can do this check, because
1790 * sgen_build_nursery_fragments() unpins nursery objects again.
1792 if (remset_consistency_checks)
1793 sgen_check_remset_consistency ();
1796 if (sgen_max_pause_time) {
1800 duration = (int)(TV_ELAPSED (last_minor_collection_start_tv, btv) / 10000);
1801 if (duration > (sgen_max_pause_time * sgen_max_pause_margin))
1802 sgen_resize_nursery (TRUE);
1804 sgen_resize_nursery (FALSE);
1806 sgen_resize_nursery (FALSE);
1809 /* walk the pin_queue, build up the fragment list of free memory, unmark
1810 * pinned objects as we go, memzero() the empty fragments so they are ready for the
1813 sgen_client_binary_protocol_reclaim_start (GENERATION_NURSERY);
1814 fragment_total = sgen_build_nursery_fragments (nursery_section, unpin_queue);
1815 if (!fragment_total)
1818 /* Clear TLABs for all threads */
1819 sgen_clear_tlabs ();
1821 sgen_client_binary_protocol_reclaim_end (GENERATION_NURSERY);
1823 time_minor_fragment_creation += TV_ELAPSED (atv, btv);
1824 SGEN_LOG (2, "Fragment creation: %lld usecs, %lu bytes available", (long long)TV_ELAPSED (atv, btv), (unsigned long)fragment_total);
1826 if (remset_consistency_checks)
1827 sgen_check_major_refs ();
1829 major_collector.finish_nursery_collection ();
1831 TV_GETTIME (last_minor_collection_end_tv);
1832 gc_stats.minor_gc_time += TV_ELAPSED (last_minor_collection_start_tv, last_minor_collection_end_tv);
1834 sgen_debug_dump_heap ("minor", gc_stats.minor_gc_count - 1, NULL);
1836 /* prepare the pin queue for the next collection */
1837 sgen_finish_pinning ();
1838 if (sgen_have_pending_finalizers ()) {
1839 SGEN_LOG (4, "Finalizer-thread wakeup");
1840 sgen_client_finalize_notify ();
1842 sgen_pin_stats_reset ();
1843 /* clear cemented hash */
1844 sgen_cement_clear_below_threshold ();
1846 sgen_gray_object_queue_dispose (&gc_thread_gray_queue);
1848 check_scan_starts ();
1850 binary_protocol_flush_buffers (FALSE);
1852 sgen_memgov_minor_collection_end (reason, is_overflow);
1854 /*objects are late pinned because of lack of memory, so a major is a good call*/
1855 needs_major = objects_pinned > 0;
1856 current_collection_generation = -1;
1859 binary_protocol_collection_end (gc_stats.minor_gc_count - 1, GENERATION_NURSERY, 0, 0);
1861 if (check_nursery_objects_pinned && !sgen_minor_collector.is_split)
1862 sgen_check_nursery_objects_pinned (unpin_queue != NULL);
1868 COPY_OR_MARK_FROM_ROOTS_SERIAL,
1869 COPY_OR_MARK_FROM_ROOTS_START_CONCURRENT,
1870 COPY_OR_MARK_FROM_ROOTS_FINISH_CONCURRENT
1871 } CopyOrMarkFromRootsMode;
1874 major_copy_or_mark_from_roots (SgenGrayQueue *gc_thread_gray_queue, size_t *old_next_pin_slot, CopyOrMarkFromRootsMode mode, SgenObjectOperations *object_ops_nopar, SgenObjectOperations *object_ops_par)
1879 /* FIXME: only use these values for the precise scan
1880 * note that to_space pointers should be excluded anyway...
1882 char *heap_start = NULL;
1883 char *heap_end = (char*)-1;
1884 ScanCopyContext ctx = CONTEXT_FROM_OBJECT_OPERATIONS (object_ops_nopar, gc_thread_gray_queue);
1885 gboolean concurrent = mode != COPY_OR_MARK_FROM_ROOTS_SERIAL;
1887 SGEN_ASSERT (0, !!concurrent == !!concurrent_collection_in_progress, "We've been called with the wrong mode.");
1889 if (mode == COPY_OR_MARK_FROM_ROOTS_START_CONCURRENT) {
1890 /*This cleans up unused fragments */
1891 sgen_nursery_allocator_prepare_for_pinning ();
1893 if (do_concurrent_checks)
1894 sgen_debug_check_nursery_is_clean ();
1896 /* The concurrent collector doesn't touch the nursery. */
1897 sgen_nursery_alloc_prepare_for_major ();
1902 /* Pinning depends on this */
1903 sgen_clear_nursery_fragments ();
1905 if (whole_heap_check_before_collection)
1906 sgen_check_whole_heap (TRUE);
1909 time_major_pre_collection_fragment_clear += TV_ELAPSED (atv, btv);
1913 sgen_client_pre_collection_checks ();
1915 if (mode != COPY_OR_MARK_FROM_ROOTS_START_CONCURRENT) {
1916 /* Remsets are not useful for a major collection */
1917 remset.clear_cards ();
1920 sgen_process_fin_stage_entries ();
1923 sgen_init_pinning ();
1924 if (mode == COPY_OR_MARK_FROM_ROOTS_START_CONCURRENT)
1925 sgen_init_pinning_for_conc ();
1926 SGEN_LOG (6, "Collecting pinned addresses");
1927 pin_from_roots ((void*)lowest_heap_address, (void*)highest_heap_address, ctx);
1928 if (mode == COPY_OR_MARK_FROM_ROOTS_FINISH_CONCURRENT) {
1929 /* Pin cemented objects that were forced */
1930 sgen_pin_cemented_objects ();
1932 sgen_optimize_pin_queue ();
1933 if (mode == COPY_OR_MARK_FROM_ROOTS_START_CONCURRENT) {
1935 * Cemented objects that are in the pinned list will be marked. When
1936 * marking concurrently we won't mark mod-union cards for these objects.
1937 * Instead they will remain cemented until the next major collection,
1938 * when we will recheck if they are still pinned in the roots.
1940 sgen_cement_force_pinned ();
1943 sgen_client_collecting_major_1 ();
1946 * pin_queue now contains all candidate pointers, sorted and
1947 * uniqued. We must do two passes now to figure out which
1948 * objects are pinned.
1950 * The first is to find within the pin_queue the area for each
1951 * section. This requires that the pin_queue be sorted. We
1952 * also process the LOS objects and pinned chunks here.
1954 * The second, destructive, pass is to reduce the section
1955 * areas to pointers to the actually pinned objects.
1957 SGEN_LOG (6, "Pinning from sections");
1958 /* first pass for the sections */
1959 sgen_find_section_pin_queue_start_end (nursery_section);
1960 /* identify possible pointers to the insize of large objects */
1961 SGEN_LOG (6, "Pinning from large objects");
1962 for (bigobj = los_object_list; bigobj; bigobj = bigobj->next) {
1964 if (sgen_find_optimized_pin_queue_area ((char*)bigobj->data, (char*)bigobj->data + sgen_los_object_size (bigobj), &dummy, &dummy)) {
1965 binary_protocol_pin (bigobj->data, (gpointer)LOAD_VTABLE (bigobj->data), safe_object_get_size (bigobj->data));
1967 if (sgen_los_object_is_pinned (bigobj->data)) {
1968 SGEN_ASSERT (0, mode == COPY_OR_MARK_FROM_ROOTS_FINISH_CONCURRENT, "LOS objects can only be pinned here after concurrent marking.");
1971 sgen_los_pin_object (bigobj->data);
1972 if (SGEN_OBJECT_HAS_REFERENCES (bigobj->data))
1973 GRAY_OBJECT_ENQUEUE_SERIAL (gc_thread_gray_queue, bigobj->data, sgen_obj_get_descriptor ((GCObject*)bigobj->data));
1974 sgen_pin_stats_register_object (bigobj->data, GENERATION_OLD);
1975 SGEN_LOG (6, "Marked large object %p (%s) size: %lu from roots", bigobj->data,
1976 sgen_client_vtable_get_name (SGEN_LOAD_VTABLE (bigobj->data)),
1977 (unsigned long)sgen_los_object_size (bigobj));
1979 sgen_client_pinned_los_object (bigobj->data);
1983 pin_objects_in_nursery (mode == COPY_OR_MARK_FROM_ROOTS_START_CONCURRENT, ctx);
1984 if (check_nursery_objects_pinned && !sgen_minor_collector.is_split)
1985 sgen_check_nursery_objects_pinned (mode != COPY_OR_MARK_FROM_ROOTS_START_CONCURRENT);
1987 major_collector.pin_objects (gc_thread_gray_queue);
1988 if (old_next_pin_slot)
1989 *old_next_pin_slot = sgen_get_pinned_count ();
1992 time_major_pinning += TV_ELAPSED (atv, btv);
1993 SGEN_LOG (2, "Finding pinned pointers: %zd in %lld usecs", sgen_get_pinned_count (), (long long)TV_ELAPSED (atv, btv));
1994 SGEN_LOG (4, "Start scan with %zd pinned objects", sgen_get_pinned_count ());
1996 if (mode == COPY_OR_MARK_FROM_ROOTS_START_CONCURRENT)
1997 sgen_finish_pinning_for_conc ();
1999 major_collector.init_to_space ();
2001 SGEN_ASSERT (0, sgen_workers_all_done (), "Why are the workers not done when we start or finish a major collection?");
2002 if (mode == COPY_OR_MARK_FROM_ROOTS_FINISH_CONCURRENT) {
2003 if (object_ops_par != NULL)
2004 sgen_workers_set_num_active_workers (GENERATION_OLD, 0);
2005 if (sgen_workers_have_idle_work (GENERATION_OLD)) {
2007 * We force the finish of the worker with the new object ops context
2008 * which can also do copying. We need to have finished pinning.
2010 sgen_workers_start_all_workers (GENERATION_OLD, object_ops_nopar, object_ops_par, NULL);
2012 sgen_workers_join (GENERATION_OLD);
2016 #ifdef SGEN_DEBUG_INTERNAL_ALLOC
2017 main_gc_thread = mono_native_thread_self ();
2020 sgen_client_collecting_major_2 ();
2023 time_major_scan_pinned += TV_ELAPSED (btv, atv);
2025 sgen_client_collecting_major_3 (&fin_ready_queue, &critical_fin_queue);
2027 enqueue_scan_from_roots_jobs (gc_thread_gray_queue, heap_start, heap_end, object_ops_nopar, FALSE);
2030 time_major_scan_roots += TV_ELAPSED (atv, btv);
2033 * We start the concurrent worker after pinning and after we scanned the roots
2034 * in order to make sure that the worker does not finish before handling all
2037 if (mode == COPY_OR_MARK_FROM_ROOTS_START_CONCURRENT) {
2038 sgen_workers_set_num_active_workers (GENERATION_OLD, 1);
2039 gray_queue_redirect (gc_thread_gray_queue);
2040 if (precleaning_enabled) {
2041 sgen_workers_start_all_workers (GENERATION_OLD, object_ops_nopar, object_ops_par, workers_finish_callback);
2043 sgen_workers_start_all_workers (GENERATION_OLD, object_ops_nopar, object_ops_par, NULL);
2047 if (mode == COPY_OR_MARK_FROM_ROOTS_FINISH_CONCURRENT) {
2048 int i, split_count = sgen_workers_get_job_split_count (GENERATION_OLD);
2049 gboolean parallel = object_ops_par != NULL;
2051 /* If we're not parallel we finish the collection on the gc thread */
2053 gray_queue_redirect (gc_thread_gray_queue);
2055 /* Mod union card table */
2056 for (i = 0; i < split_count; i++) {
2057 ParallelScanJob *psj;
2059 psj = (ParallelScanJob*)sgen_thread_pool_job_alloc ("scan mod union cardtable", job_scan_major_mod_union_card_table, sizeof (ParallelScanJob));
2060 psj->scan_job.ops = parallel ? NULL : object_ops_nopar;
2061 psj->scan_job.gc_thread_gray_queue = gc_thread_gray_queue;
2063 psj->job_split_count = split_count;
2064 sgen_workers_enqueue_job (GENERATION_OLD, &psj->scan_job.job, parallel);
2066 psj = (ParallelScanJob*)sgen_thread_pool_job_alloc ("scan LOS mod union cardtable", job_scan_los_mod_union_card_table, sizeof (ParallelScanJob));
2067 psj->scan_job.ops = parallel ? NULL : object_ops_nopar;
2068 psj->scan_job.gc_thread_gray_queue = gc_thread_gray_queue;
2070 psj->job_split_count = split_count;
2071 sgen_workers_enqueue_job (GENERATION_OLD, &psj->scan_job.job, parallel);
2076 * If we enqueue a job while workers are running we need to sgen_workers_ensure_awake
2077 * in order to make sure that we are running the idle func and draining all worker
2078 * gray queues. The operation of starting workers implies this, so we start them after
2079 * in order to avoid doing this operation twice. The workers will drain the main gray
2080 * stack that contained roots and pinned objects and also scan the mod union card
2083 sgen_workers_start_all_workers (GENERATION_OLD, object_ops_nopar, object_ops_par, NULL);
2084 sgen_workers_join (GENERATION_OLD);
2088 sgen_pin_stats_report ();
2090 if (mode == COPY_OR_MARK_FROM_ROOTS_START_CONCURRENT) {
2091 sgen_finish_pinning ();
2093 sgen_pin_stats_reset ();
2095 if (do_concurrent_checks)
2096 sgen_debug_check_nursery_is_clean ();
2101 major_start_collection (SgenGrayQueue *gc_thread_gray_queue, const char *reason, gboolean concurrent, size_t *old_next_pin_slot)
2103 SgenObjectOperations *object_ops_nopar, *object_ops_par = NULL;
2105 binary_protocol_collection_begin (gc_stats.major_gc_count, GENERATION_OLD);
2107 current_collection_generation = GENERATION_OLD;
2109 sgen_workers_assert_gray_queue_is_empty (GENERATION_OLD);
2112 sgen_cement_reset ();
2115 g_assert (major_collector.is_concurrent);
2116 concurrent_collection_in_progress = TRUE;
2118 object_ops_nopar = &major_collector.major_ops_concurrent_start;
2119 if (major_collector.is_parallel)
2120 object_ops_par = &major_collector.major_ops_conc_par_start;
2123 object_ops_nopar = &major_collector.major_ops_serial;
2126 reset_pinned_from_failed_allocation ();
2128 sgen_memgov_major_collection_start (concurrent, reason);
2130 //count_ref_nonref_objs ();
2131 //consistency_check ();
2133 check_scan_starts ();
2136 SGEN_LOG (1, "Start major collection %d", gc_stats.major_gc_count);
2137 gc_stats.major_gc_count ++;
2139 if (major_collector.start_major_collection)
2140 major_collector.start_major_collection ();
2142 major_copy_or_mark_from_roots (gc_thread_gray_queue, old_next_pin_slot, concurrent ? COPY_OR_MARK_FROM_ROOTS_START_CONCURRENT : COPY_OR_MARK_FROM_ROOTS_SERIAL, object_ops_nopar, object_ops_par);
2146 major_finish_collection (SgenGrayQueue *gc_thread_gray_queue, const char *reason, gboolean is_overflow, size_t old_next_pin_slot, gboolean forced)
2148 ScannedObjectCounts counts;
2149 SgenObjectOperations *object_ops_nopar;
2150 mword fragment_total;
2156 if (concurrent_collection_in_progress) {
2157 SgenObjectOperations *object_ops_par = NULL;
2159 object_ops_nopar = &major_collector.major_ops_concurrent_finish;
2160 if (major_collector.is_parallel)
2161 object_ops_par = &major_collector.major_ops_conc_par_finish;
2163 major_copy_or_mark_from_roots (gc_thread_gray_queue, NULL, COPY_OR_MARK_FROM_ROOTS_FINISH_CONCURRENT, object_ops_nopar, object_ops_par);
2165 #ifdef SGEN_DEBUG_INTERNAL_ALLOC
2166 main_gc_thread = NULL;
2169 object_ops_nopar = &major_collector.major_ops_serial;
2172 sgen_workers_assert_gray_queue_is_empty (GENERATION_OLD);
2174 finish_gray_stack (GENERATION_OLD, CONTEXT_FROM_OBJECT_OPERATIONS (object_ops_nopar, gc_thread_gray_queue));
2176 time_major_finish_gray_stack += TV_ELAPSED (btv, atv);
2178 SGEN_ASSERT (0, sgen_workers_all_done (), "Can't have workers working after joining");
2180 if (objects_pinned) {
2181 g_assert (!concurrent_collection_in_progress);
2184 * This is slow, but we just OOM'd.
2186 * See comment at `sgen_pin_queue_clear_discarded_entries` for how the pin
2187 * queue is laid out at this point.
2189 sgen_pin_queue_clear_discarded_entries (nursery_section, old_next_pin_slot);
2191 * We need to reestablish all pinned nursery objects in the pin queue
2192 * because they're needed for fragment creation. Unpinning happens by
2193 * walking the whole queue, so it's not necessary to reestablish where major
2194 * heap block pins are - all we care is that they're still in there
2197 sgen_optimize_pin_queue ();
2198 sgen_find_section_pin_queue_start_end (nursery_section);
2202 reset_heap_boundaries ();
2203 sgen_update_heap_boundaries ((mword)sgen_get_nursery_start (), (mword)sgen_get_nursery_end ());
2205 /* walk the pin_queue, build up the fragment list of free memory, unmark
2206 * pinned objects as we go, memzero() the empty fragments so they are ready for the
2209 fragment_total = sgen_build_nursery_fragments (nursery_section, NULL);
2210 if (!fragment_total)
2212 SGEN_LOG (4, "Free space in nursery after major %ld", (long)fragment_total);
2214 if (do_concurrent_checks && concurrent_collection_in_progress)
2215 sgen_debug_check_nursery_is_clean ();
2217 /* prepare the pin queue for the next collection */
2218 sgen_finish_pinning ();
2220 /* Clear TLABs for all threads */
2221 sgen_clear_tlabs ();
2223 sgen_pin_stats_reset ();
2225 sgen_cement_clear_below_threshold ();
2227 if (check_mark_bits_after_major_collection)
2228 sgen_check_heap_marked (concurrent_collection_in_progress);
2231 time_major_fragment_creation += TV_ELAPSED (atv, btv);
2233 binary_protocol_sweep_begin (GENERATION_OLD, !major_collector.sweeps_lazily);
2234 sgen_memgov_major_pre_sweep ();
2237 time_major_free_bigobjs += TV_ELAPSED (btv, atv);
2242 time_major_los_sweep += TV_ELAPSED (atv, btv);
2244 major_collector.sweep ();
2246 binary_protocol_sweep_end (GENERATION_OLD, !major_collector.sweeps_lazily);
2249 time_major_sweep += TV_ELAPSED (btv, atv);
2251 sgen_debug_dump_heap ("major", gc_stats.major_gc_count - 1, reason);
2253 if (sgen_have_pending_finalizers ()) {
2254 SGEN_LOG (4, "Finalizer-thread wakeup");
2255 sgen_client_finalize_notify ();
2258 sgen_memgov_major_collection_end (forced, concurrent_collection_in_progress, reason, is_overflow);
2259 current_collection_generation = -1;
2261 memset (&counts, 0, sizeof (ScannedObjectCounts));
2262 major_collector.finish_major_collection (&counts);
2264 sgen_workers_assert_gray_queue_is_empty (GENERATION_OLD);
2266 SGEN_ASSERT (0, sgen_workers_all_done (), "Can't have workers working after major collection has finished");
2267 if (concurrent_collection_in_progress)
2268 concurrent_collection_in_progress = FALSE;
2270 check_scan_starts ();
2272 binary_protocol_flush_buffers (FALSE);
2274 //consistency_check ();
2276 binary_protocol_collection_end (gc_stats.major_gc_count - 1, GENERATION_OLD, counts.num_scanned_objects, counts.num_unique_scanned_objects);
2280 major_do_collection (const char *reason, gboolean is_overflow, gboolean forced)
2282 TV_DECLARE (time_start);
2283 TV_DECLARE (time_end);
2284 size_t old_next_pin_slot;
2285 SgenGrayQueue gc_thread_gray_queue;
2287 if (disable_major_collections)
2290 if (major_collector.get_and_reset_num_major_objects_marked) {
2291 long long num_marked = major_collector.get_and_reset_num_major_objects_marked ();
2292 g_assert (!num_marked);
2295 /* world must be stopped already */
2296 TV_GETTIME (time_start);
2298 init_gray_queue (&gc_thread_gray_queue);
2299 major_start_collection (&gc_thread_gray_queue, reason, FALSE, &old_next_pin_slot);
2300 major_finish_collection (&gc_thread_gray_queue, reason, is_overflow, old_next_pin_slot, forced);
2301 sgen_gray_object_queue_dispose (&gc_thread_gray_queue);
2303 TV_GETTIME (time_end);
2304 gc_stats.major_gc_time += TV_ELAPSED (time_start, time_end);
2306 /* FIXME: also report this to the user, preferably in gc-end. */
2307 if (major_collector.get_and_reset_num_major_objects_marked)
2308 major_collector.get_and_reset_num_major_objects_marked ();
2310 return bytes_pinned_from_failed_allocation > 0;
2314 major_start_concurrent_collection (const char *reason)
2316 TV_DECLARE (time_start);
2317 TV_DECLARE (time_end);
2318 long long num_objects_marked;
2319 SgenGrayQueue gc_thread_gray_queue;
2321 if (disable_major_collections)
2324 TV_GETTIME (time_start);
2325 SGEN_TV_GETTIME (time_major_conc_collection_start);
2327 num_objects_marked = major_collector.get_and_reset_num_major_objects_marked ();
2328 g_assert (num_objects_marked == 0);
2330 binary_protocol_concurrent_start ();
2332 init_gray_queue (&gc_thread_gray_queue);
2333 // FIXME: store reason and pass it when finishing
2334 major_start_collection (&gc_thread_gray_queue, reason, TRUE, NULL);
2335 sgen_gray_object_queue_dispose (&gc_thread_gray_queue);
2337 num_objects_marked = major_collector.get_and_reset_num_major_objects_marked ();
2339 TV_GETTIME (time_end);
2340 gc_stats.major_gc_time += TV_ELAPSED (time_start, time_end);
2342 current_collection_generation = -1;
2346 * Returns whether the major collection has finished.
2349 major_should_finish_concurrent_collection (void)
2351 return sgen_workers_all_done ();
2355 major_update_concurrent_collection (void)
2357 TV_DECLARE (total_start);
2358 TV_DECLARE (total_end);
2360 TV_GETTIME (total_start);
2362 binary_protocol_concurrent_update ();
2364 major_collector.update_cardtable_mod_union ();
2365 sgen_los_update_cardtable_mod_union ();
2367 TV_GETTIME (total_end);
2368 gc_stats.major_gc_time += TV_ELAPSED (total_start, total_end);
2372 major_finish_concurrent_collection (gboolean forced)
2374 SgenGrayQueue gc_thread_gray_queue;
2375 TV_DECLARE (total_start);
2376 TV_DECLARE (total_end);
2378 TV_GETTIME (total_start);
2380 binary_protocol_concurrent_finish ();
2383 * We need to stop all workers since we're updating the cardtable below.
2384 * The workers will be resumed with a finishing pause context to avoid
2385 * additional cardtable and object scanning.
2387 sgen_workers_stop_all_workers (GENERATION_OLD);
2389 SGEN_TV_GETTIME (time_major_conc_collection_end);
2390 gc_stats.major_gc_time_concurrent += SGEN_TV_ELAPSED (time_major_conc_collection_start, time_major_conc_collection_end);
2392 major_collector.update_cardtable_mod_union ();
2393 sgen_los_update_cardtable_mod_union ();
2395 if (mod_union_consistency_check)
2396 sgen_check_mod_union_consistency ();
2398 current_collection_generation = GENERATION_OLD;
2399 sgen_cement_reset ();
2400 init_gray_queue (&gc_thread_gray_queue);
2401 major_finish_collection (&gc_thread_gray_queue, "finishing", FALSE, -1, forced);
2402 sgen_gray_object_queue_dispose (&gc_thread_gray_queue);
2404 TV_GETTIME (total_end);
2405 gc_stats.major_gc_time += TV_ELAPSED (total_start, total_end);
2407 current_collection_generation = -1;
2411 * Ensure an allocation request for @size will succeed by freeing enough memory.
2413 * LOCKING: The GC lock MUST be held.
2416 sgen_ensure_free_space (size_t size, int generation)
2418 int generation_to_collect = -1;
2419 const char *reason = NULL;
2421 if (generation == GENERATION_OLD) {
2422 if (sgen_need_major_collection (size)) {
2423 reason = "LOS overflow";
2424 generation_to_collect = GENERATION_OLD;
2427 if (degraded_mode) {
2428 if (sgen_need_major_collection (size)) {
2429 reason = "Degraded mode overflow";
2430 generation_to_collect = GENERATION_OLD;
2432 } else if (sgen_need_major_collection (size)) {
2433 reason = concurrent_collection_in_progress ? "Forced finish concurrent collection" : "Minor allowance";
2434 generation_to_collect = GENERATION_OLD;
2436 generation_to_collect = GENERATION_NURSERY;
2437 reason = "Nursery full";
2441 if (generation_to_collect == -1) {
2442 if (concurrent_collection_in_progress && sgen_workers_all_done ()) {
2443 generation_to_collect = GENERATION_OLD;
2444 reason = "Finish concurrent collection";
2448 if (generation_to_collect == -1)
2450 sgen_perform_collection (size, generation_to_collect, reason, FALSE, TRUE);
2454 * LOCKING: Assumes the GC lock is held.
2457 sgen_perform_collection (size_t requested_size, int generation_to_collect, const char *reason, gboolean wait_to_finish, gboolean stw)
2459 TV_DECLARE (gc_total_start);
2460 TV_DECLARE (gc_total_end);
2461 int overflow_generation_to_collect = -1;
2462 int oldest_generation_collected = generation_to_collect;
2463 const char *overflow_reason = NULL;
2464 gboolean finish_concurrent = concurrent_collection_in_progress && (major_should_finish_concurrent_collection () || generation_to_collect == GENERATION_OLD);
2466 binary_protocol_collection_requested (generation_to_collect, requested_size, wait_to_finish ? 1 : 0);
2468 SGEN_ASSERT (0, generation_to_collect == GENERATION_NURSERY || generation_to_collect == GENERATION_OLD, "What generation is this?");
2471 sgen_stop_world (generation_to_collect);
2473 SGEN_ASSERT (0, sgen_is_world_stopped (), "We can only collect if the world is stopped");
2476 TV_GETTIME (gc_total_start);
2478 // FIXME: extract overflow reason
2479 // FIXME: minor overflow for concurrent case
2480 if (generation_to_collect == GENERATION_NURSERY && !finish_concurrent) {
2481 if (concurrent_collection_in_progress)
2482 major_update_concurrent_collection ();
2484 if (collect_nursery (reason, FALSE, NULL) && !concurrent_collection_in_progress) {
2485 overflow_generation_to_collect = GENERATION_OLD;
2486 overflow_reason = "Minor overflow";
2488 } else if (finish_concurrent) {
2489 major_finish_concurrent_collection (wait_to_finish);
2490 oldest_generation_collected = GENERATION_OLD;
2492 SGEN_ASSERT (0, generation_to_collect == GENERATION_OLD, "We should have handled nursery collections above");
2493 if (major_collector.is_concurrent && !wait_to_finish) {
2494 collect_nursery ("Concurrent start", FALSE, NULL);
2495 major_start_concurrent_collection (reason);
2496 oldest_generation_collected = GENERATION_NURSERY;
2497 } else if (major_do_collection (reason, FALSE, wait_to_finish)) {
2498 overflow_generation_to_collect = GENERATION_NURSERY;
2499 overflow_reason = "Excessive pinning";
2503 if (overflow_generation_to_collect != -1) {
2504 SGEN_ASSERT (0, !concurrent_collection_in_progress, "We don't yet support overflow collections with the concurrent collector");
2507 * We need to do an overflow collection, either because we ran out of memory
2508 * or the nursery is fully pinned.
2511 if (overflow_generation_to_collect == GENERATION_NURSERY)
2512 collect_nursery (overflow_reason, TRUE, NULL);
2514 major_do_collection (overflow_reason, TRUE, wait_to_finish);
2516 oldest_generation_collected = MAX (oldest_generation_collected, overflow_generation_to_collect);
2519 SGEN_LOG (2, "Heap size: %lu, LOS size: %lu", (unsigned long)sgen_gc_get_total_heap_allocation (), (unsigned long)los_memory_usage);
2521 /* this also sets the proper pointers for the next allocation */
2522 if (generation_to_collect == GENERATION_NURSERY && !sgen_can_alloc_size (requested_size)) {
2523 /* TypeBuilder and MonoMethod are killing mcs with fragmentation */
2524 SGEN_LOG (1, "nursery collection didn't find enough room for %zd alloc (%zd pinned)", requested_size, sgen_get_pinned_count ());
2525 sgen_dump_pin_queue ();
2529 TV_GETTIME (gc_total_end);
2530 time_max = MAX (time_max, TV_ELAPSED (gc_total_start, gc_total_end));
2533 sgen_restart_world (oldest_generation_collected);
2537 * ######################################################################
2538 * ######## Memory allocation from the OS
2539 * ######################################################################
2540 * This section of code deals with getting memory from the OS and
2541 * allocating memory for GC-internal data structures.
2542 * Internal memory can be handled with a freelist for small objects.
2548 G_GNUC_UNUSED static void
2549 report_internal_mem_usage (void)
2551 printf ("Internal memory usage:\n");
2552 sgen_report_internal_mem_usage ();
2553 printf ("Pinned memory usage:\n");
2554 major_collector.report_pinned_memory_usage ();
2558 * ######################################################################
2559 * ######## Finalization support
2560 * ######################################################################
2564 * If the object has been forwarded it means it's still referenced from a root.
2565 * If it is pinned it's still alive as well.
2566 * A LOS object is only alive if we have pinned it.
2567 * Return TRUE if @obj is ready to be finalized.
2569 static inline gboolean
2570 sgen_is_object_alive (GCObject *object)
2572 if (ptr_in_nursery (object))
2573 return sgen_nursery_is_object_alive (object);
2575 return sgen_major_is_object_alive (object);
2579 * This function returns true if @object is either alive and belongs to the
2580 * current collection - major collections are full heap, so old gen objects
2581 * are never alive during a minor collection.
2584 sgen_is_object_alive_and_on_current_collection (GCObject *object)
2586 if (ptr_in_nursery (object))
2587 return sgen_nursery_is_object_alive (object);
2589 if (current_collection_generation == GENERATION_NURSERY)
2592 return sgen_major_is_object_alive (object);
2597 sgen_gc_is_object_ready_for_finalization (GCObject *object)
2599 return !sgen_is_object_alive (object);
2603 sgen_queue_finalization_entry (GCObject *obj)
2605 gboolean critical = sgen_client_object_has_critical_finalizer (obj);
2607 sgen_pointer_queue_add (critical ? &critical_fin_queue : &fin_ready_queue, obj);
2609 sgen_client_object_queued_for_finalization (obj);
2613 sgen_object_is_live (GCObject *obj)
2615 return sgen_is_object_alive_and_on_current_collection (obj);
2619 * `System.GC.WaitForPendingFinalizers` first checks `sgen_have_pending_finalizers()` to
2620 * determine whether it can exit quickly. The latter must therefore only return FALSE if
2621 * all finalizers have really finished running.
2623 * `sgen_gc_invoke_finalizers()` first dequeues a finalizable object, and then finalizes it.
2624 * This means that just checking whether the queues are empty leaves the possibility that an
2625 * object might have been dequeued but not yet finalized. That's why we need the additional
2626 * flag `pending_unqueued_finalizer`.
2629 static volatile gboolean pending_unqueued_finalizer = FALSE;
2630 volatile gboolean sgen_suspend_finalizers = FALSE;
2633 sgen_set_suspend_finalizers (void)
2635 sgen_suspend_finalizers = TRUE;
2639 sgen_gc_invoke_finalizers (void)
2643 g_assert (!pending_unqueued_finalizer);
2645 /* FIXME: batch to reduce lock contention */
2646 while (sgen_have_pending_finalizers ()) {
2652 * We need to set `pending_unqueued_finalizer` before dequeing the
2653 * finalizable object.
2655 if (!sgen_pointer_queue_is_empty (&fin_ready_queue)) {
2656 pending_unqueued_finalizer = TRUE;
2657 mono_memory_write_barrier ();
2658 obj = (GCObject *)sgen_pointer_queue_pop (&fin_ready_queue);
2659 } else if (!sgen_pointer_queue_is_empty (&critical_fin_queue)) {
2660 pending_unqueued_finalizer = TRUE;
2661 mono_memory_write_barrier ();
2662 obj = (GCObject *)sgen_pointer_queue_pop (&critical_fin_queue);
2668 SGEN_LOG (7, "Finalizing object %p (%s)", obj, sgen_client_vtable_get_name (SGEN_LOAD_VTABLE (obj)));
2676 /* the object is on the stack so it is pinned */
2677 /*g_print ("Calling finalizer for object: %p (%s)\n", obj, sgen_client_object_safe_name (obj));*/
2678 sgen_client_run_finalize (obj);
2681 if (pending_unqueued_finalizer) {
2682 mono_memory_write_barrier ();
2683 pending_unqueued_finalizer = FALSE;
2690 sgen_have_pending_finalizers (void)
2692 if (sgen_suspend_finalizers)
2694 return pending_unqueued_finalizer || !sgen_pointer_queue_is_empty (&fin_ready_queue) || !sgen_pointer_queue_is_empty (&critical_fin_queue);
2698 * ######################################################################
2699 * ######## registered roots support
2700 * ######################################################################
2704 * We do not coalesce roots.
2707 sgen_register_root (char *start, size_t size, SgenDescriptor descr, int root_type, int source, const char *msg)
2709 RootRecord new_root;
2712 for (i = 0; i < ROOT_TYPE_NUM; ++i) {
2713 RootRecord *root = (RootRecord *)sgen_hash_table_lookup (&roots_hash [i], start);
2714 /* we allow changing the size and the descriptor (for thread statics etc) */
2716 size_t old_size = root->end_root - start;
2717 root->end_root = start + size;
2718 SGEN_ASSERT (0, !!root->root_desc == !!descr, "Can't change whether a root is precise or conservative.");
2719 SGEN_ASSERT (0, root->source == source, "Can't change a root's source identifier.");
2720 SGEN_ASSERT (0, !!root->msg == !!msg, "Can't change a root's message.");
2721 root->root_desc = descr;
2723 roots_size -= old_size;
2729 new_root.end_root = start + size;
2730 new_root.root_desc = descr;
2731 new_root.source = source;
2734 sgen_hash_table_replace (&roots_hash [root_type], start, &new_root, NULL);
2737 SGEN_LOG (3, "Added root for range: %p-%p, descr: %llx (%d/%d bytes)", start, new_root.end_root, (long long)descr, (int)size, (int)roots_size);
2744 sgen_deregister_root (char* addr)
2750 for (root_type = 0; root_type < ROOT_TYPE_NUM; ++root_type) {
2751 if (sgen_hash_table_remove (&roots_hash [root_type], addr, &root))
2752 roots_size -= (root.end_root - addr);
2758 sgen_wbroots_iterate_live_block_ranges (sgen_cardtable_block_callback cb)
2762 SGEN_HASH_TABLE_FOREACH (&roots_hash [ROOT_TYPE_WBARRIER], void **, start_root, RootRecord *, root) {
2763 cb ((mword)start_root, (mword)root->end_root - (mword)start_root);
2764 } SGEN_HASH_TABLE_FOREACH_END;
2767 /* Root equivalent of sgen_client_cardtable_scan_object */
2769 sgen_wbroot_scan_card_table (void** start_root, mword size, ScanCopyContext ctx)
2771 ScanPtrFieldFunc scan_field_func = ctx.ops->scan_ptr_field;
2772 guint8 *card_data = sgen_card_table_get_card_scan_address ((mword)start_root);
2773 guint8 *card_base = card_data;
2774 mword card_count = sgen_card_table_number_of_cards_in_range ((mword)start_root, size);
2775 guint8 *card_data_end = card_data + card_count;
2776 mword extra_idx = 0;
2777 char *obj_start = sgen_card_table_align_pointer (start_root);
2778 char *obj_end = (char*)start_root + size;
2779 #ifdef SGEN_HAVE_OVERLAPPING_CARDS
2780 guint8 *overflow_scan_end = NULL;
2783 #ifdef SGEN_HAVE_OVERLAPPING_CARDS
2784 /*Check for overflow and if so, setup to scan in two steps*/
2785 if (card_data_end >= SGEN_SHADOW_CARDTABLE_END) {
2786 overflow_scan_end = sgen_shadow_cardtable + (card_data_end - SGEN_SHADOW_CARDTABLE_END);
2787 card_data_end = SGEN_SHADOW_CARDTABLE_END;
2793 card_data = sgen_find_next_card (card_data, card_data_end);
2795 for (; card_data < card_data_end; card_data = sgen_find_next_card (card_data + 1, card_data_end)) {
2796 size_t idx = (card_data - card_base) + extra_idx;
2797 char *start = (char*)(obj_start + idx * CARD_SIZE_IN_BYTES);
2798 char *card_end = start + CARD_SIZE_IN_BYTES;
2799 char *elem = start, *first_elem = start;
2802 * Don't clean first and last card on 32bit systems since they
2803 * may also be part from other roots.
2805 if (card_data != card_base && card_data != (card_data_end - 1))
2806 sgen_card_table_prepare_card_for_scanning (card_data);
2808 card_end = MIN (card_end, obj_end);
2810 if (elem < (char*)start_root)
2811 first_elem = elem = (char*)start_root;
2813 for (; elem < card_end; elem += SIZEOF_VOID_P) {
2814 if (*(GCObject**)elem)
2815 scan_field_func (NULL, (GCObject**)elem, ctx.queue);
2818 binary_protocol_card_scan (first_elem, elem - first_elem);
2821 #ifdef SGEN_HAVE_OVERLAPPING_CARDS
2822 if (overflow_scan_end) {
2823 extra_idx = card_data - card_base;
2824 card_base = card_data = sgen_shadow_cardtable;
2825 card_data_end = overflow_scan_end;
2826 overflow_scan_end = NULL;
2833 sgen_wbroots_scan_card_table (ScanCopyContext ctx)
2838 SGEN_HASH_TABLE_FOREACH (&roots_hash [ROOT_TYPE_WBARRIER], void **, start_root, RootRecord *, root) {
2839 SGEN_ASSERT (0, (root->root_desc & ROOT_DESC_TYPE_MASK) == ROOT_DESC_VECTOR, "Unsupported root type");
2841 sgen_wbroot_scan_card_table (start_root, (mword)root->end_root - (mword)start_root, ctx);
2842 } SGEN_HASH_TABLE_FOREACH_END;
2846 * ######################################################################
2847 * ######## Thread handling (stop/start code)
2848 * ######################################################################
2852 sgen_get_current_collection_generation (void)
2854 return current_collection_generation;
2858 sgen_thread_attach (SgenThreadInfo* info)
2860 info->tlab_start = info->tlab_next = info->tlab_temp_end = info->tlab_real_end = NULL;
2862 sgen_client_thread_attach (info);
2868 sgen_thread_detach_with_lock (SgenThreadInfo *p)
2870 sgen_client_thread_detach_with_lock (p);
2874 * ######################################################################
2875 * ######## Write barriers
2876 * ######################################################################
2880 * Note: the write barriers first do the needed GC work and then do the actual store:
2881 * this way the value is visible to the conservative GC scan after the write barrier
2882 * itself. If a GC interrupts the barrier in the middle, value will be kept alive by
2883 * the conservative scan, otherwise by the remembered set scan.
2887 * mono_gc_wbarrier_arrayref_copy:
2890 mono_gc_wbarrier_arrayref_copy (gpointer dest_ptr, gpointer src_ptr, int count)
2892 HEAVY_STAT (++stat_wbarrier_arrayref_copy);
2893 /*This check can be done without taking a lock since dest_ptr array is pinned*/
2894 if (ptr_in_nursery (dest_ptr) || count <= 0) {
2895 mono_gc_memmove_aligned (dest_ptr, src_ptr, count * sizeof (gpointer));
2899 #ifdef SGEN_HEAVY_BINARY_PROTOCOL
2900 if (binary_protocol_is_heavy_enabled ()) {
2902 for (i = 0; i < count; ++i) {
2903 gpointer dest = (gpointer*)dest_ptr + i;
2904 gpointer obj = *((gpointer*)src_ptr + i);
2906 binary_protocol_wbarrier (dest, obj, (gpointer)LOAD_VTABLE (obj));
2911 remset.wbarrier_arrayref_copy (dest_ptr, src_ptr, count);
2915 * mono_gc_wbarrier_generic_nostore:
2918 mono_gc_wbarrier_generic_nostore (gpointer ptr)
2922 HEAVY_STAT (++stat_wbarrier_generic_store);
2924 sgen_client_wbarrier_generic_nostore_check (ptr);
2926 obj = *(gpointer*)ptr;
2928 binary_protocol_wbarrier (ptr, obj, (gpointer)LOAD_VTABLE (obj));
2931 * We need to record old->old pointer locations for the
2932 * concurrent collector.
2934 if (!ptr_in_nursery (obj) && !concurrent_collection_in_progress) {
2935 SGEN_LOG (8, "Skipping remset at %p", ptr);
2939 SGEN_LOG (8, "Adding remset at %p", ptr);
2941 remset.wbarrier_generic_nostore (ptr);
2945 * mono_gc_wbarrier_generic_store:
2948 mono_gc_wbarrier_generic_store (gpointer ptr, GCObject* value)
2950 SGEN_LOG (8, "Wbarrier store at %p to %p (%s)", ptr, value, value ? sgen_client_vtable_get_name (SGEN_LOAD_VTABLE (value)) : "null");
2951 SGEN_UPDATE_REFERENCE_ALLOW_NULL (ptr, value);
2952 if (ptr_in_nursery (value) || concurrent_collection_in_progress)
2953 mono_gc_wbarrier_generic_nostore (ptr);
2954 sgen_dummy_use (value);
2958 * mono_gc_wbarrier_generic_store_atomic:
2959 * Same as \c mono_gc_wbarrier_generic_store but performs the store
2960 * as an atomic operation with release semantics.
2963 mono_gc_wbarrier_generic_store_atomic (gpointer ptr, GCObject *value)
2965 HEAVY_STAT (++stat_wbarrier_generic_store_atomic);
2967 SGEN_LOG (8, "Wbarrier atomic store at %p to %p (%s)", ptr, value, value ? sgen_client_vtable_get_name (SGEN_LOAD_VTABLE (value)) : "null");
2969 InterlockedWritePointer ((volatile gpointer *)ptr, value);
2971 if (ptr_in_nursery (value) || concurrent_collection_in_progress)
2972 mono_gc_wbarrier_generic_nostore (ptr);
2974 sgen_dummy_use (value);
2978 sgen_wbarrier_range_copy (gpointer _dest, gpointer _src, int size)
2980 remset.wbarrier_range_copy (_dest,_src, size);
2984 * ######################################################################
2985 * ######## Other mono public interface functions.
2986 * ######################################################################
2990 sgen_gc_collect (int generation)
2995 sgen_perform_collection (0, generation, "user request", TRUE, TRUE);
3000 sgen_gc_collection_count (int generation)
3002 if (generation == 0)
3003 return gc_stats.minor_gc_count;
3004 return gc_stats.major_gc_count;
3008 sgen_gc_get_used_size (void)
3012 tot = los_memory_usage;
3013 tot += nursery_section->end_data - nursery_section->data;
3014 tot += major_collector.get_used_size ();
3015 /* FIXME: account for pinned objects */
3021 sgen_env_var_error (const char *env_var, const char *fallback, const char *description_format, ...)
3025 va_start (ap, description_format);
3027 fprintf (stderr, "Warning: In environment variable `%s': ", env_var);
3028 vfprintf (stderr, description_format, ap);
3030 fprintf (stderr, " - %s", fallback);
3031 fprintf (stderr, "\n");
3037 parse_double_in_interval (const char *env_var, const char *opt_name, const char *opt, double min, double max, double *result)
3040 double val = strtod (opt, &endptr);
3041 if (endptr == opt) {
3042 sgen_env_var_error (env_var, "Using default value.", "`%s` must be a number.", opt_name);
3045 else if (val < min || val > max) {
3046 sgen_env_var_error (env_var, "Using default value.", "`%s` must be between %.2f - %.2f.", opt_name, min, max);
3054 parse_sgen_minor (const char *opt)
3057 return SGEN_MINOR_DEFAULT;
3059 if (!strcmp (opt, "simple")) {
3060 return SGEN_MINOR_SIMPLE;
3061 } else if (!strcmp (opt, "simple-par")) {
3062 return SGEN_MINOR_SIMPLE_PARALLEL;
3063 } else if (!strcmp (opt, "split")) {
3064 return SGEN_MINOR_SPLIT;
3066 sgen_env_var_error (MONO_GC_PARAMS_NAME, "Using default instead.", "Unknown minor collector `%s'.", opt);
3067 return SGEN_MINOR_DEFAULT;
3072 parse_sgen_major (const char *opt)
3075 return SGEN_MAJOR_DEFAULT;
3077 if (!strcmp (opt, "marksweep")) {
3078 return SGEN_MAJOR_SERIAL;
3079 } else if (!strcmp (opt, "marksweep-conc")) {
3080 return SGEN_MAJOR_CONCURRENT;
3081 } else if (!strcmp (opt, "marksweep-conc-par")) {
3082 return SGEN_MAJOR_CONCURRENT_PARALLEL;
3084 sgen_env_var_error (MONO_GC_PARAMS_NAME, "Using default instead.", "Unknown major collector `%s'.", opt);
3085 return SGEN_MAJOR_DEFAULT;
3091 parse_sgen_mode (const char *opt)
3094 return SGEN_MODE_NONE;
3096 if (!strcmp (opt, "balanced")) {
3097 return SGEN_MODE_BALANCED;
3098 } else if (!strcmp (opt, "throughput")) {
3099 return SGEN_MODE_THROUGHPUT;
3100 } else if (!strcmp (opt, "pause") || g_str_has_prefix (opt, "pause:")) {
3101 return SGEN_MODE_PAUSE;
3103 sgen_env_var_error (MONO_GC_PARAMS_NAME, "Using default configurations.", "Unknown mode `%s'.", opt);
3104 return SGEN_MODE_NONE;
3109 init_sgen_minor (SgenMinor minor)
3112 case SGEN_MINOR_DEFAULT:
3113 case SGEN_MINOR_SIMPLE:
3114 sgen_simple_nursery_init (&sgen_minor_collector, FALSE);
3116 case SGEN_MINOR_SIMPLE_PARALLEL:
3117 sgen_simple_nursery_init (&sgen_minor_collector, TRUE);
3119 case SGEN_MINOR_SPLIT:
3120 sgen_split_nursery_init (&sgen_minor_collector);
3123 g_assert_not_reached ();
3128 init_sgen_major (SgenMajor major)
3130 if (major == SGEN_MAJOR_DEFAULT)
3131 major = DEFAULT_MAJOR;
3134 case SGEN_MAJOR_SERIAL:
3135 sgen_marksweep_init (&major_collector);
3137 case SGEN_MAJOR_CONCURRENT:
3138 sgen_marksweep_conc_init (&major_collector);
3140 case SGEN_MAJOR_CONCURRENT_PARALLEL:
3141 sgen_marksweep_conc_par_init (&major_collector);
3144 g_assert_not_reached ();
3149 * If sgen mode is set, major/minor configuration is fixed. The other gc_params
3150 * are parsed and processed after major/minor initialization, so it can potentially
3151 * override some knobs set by the sgen mode. We can consider locking out additional
3152 * configurations when gc_modes are used.
3155 init_sgen_mode (SgenMode mode)
3157 SgenMinor minor = SGEN_MINOR_DEFAULT;
3158 SgenMajor major = SGEN_MAJOR_DEFAULT;
3161 case SGEN_MODE_BALANCED:
3163 * Use a dynamic parallel nursery with a major concurrent collector.
3164 * This uses the default values for max pause time and nursery size.
3166 minor = SGEN_MINOR_SIMPLE;
3167 major = SGEN_MAJOR_CONCURRENT;
3168 dynamic_nursery = TRUE;
3170 case SGEN_MODE_THROUGHPUT:
3172 * Use concurrent major to let the mutator do more work. Use a larger
3173 * nursery, without pause time constraints, in order to collect more
3174 * objects in parallel and avoid repetitive collection tasks (pinning,
3175 * root scanning etc)
3177 minor = SGEN_MINOR_SIMPLE_PARALLEL;
3178 major = SGEN_MAJOR_CONCURRENT;
3179 dynamic_nursery = TRUE;
3180 sgen_max_pause_time = 0;
3182 case SGEN_MODE_PAUSE:
3184 * Use concurrent major and dynamic nursery with a more
3185 * aggressive shrinking relative to pause times.
3187 minor = SGEN_MINOR_SIMPLE_PARALLEL;
3188 major = SGEN_MAJOR_CONCURRENT;
3189 dynamic_nursery = TRUE;
3190 sgen_max_pause_margin = SGEN_PAUSE_MODE_MAX_PAUSE_MARGIN;
3193 g_assert_not_reached ();
3196 init_sgen_minor (minor);
3197 init_sgen_major (major);
3205 SgenMajor sgen_major = SGEN_MAJOR_DEFAULT;
3206 SgenMinor sgen_minor = SGEN_MINOR_DEFAULT;
3207 SgenMode sgen_mode = SGEN_MODE_NONE;
3208 char *params_opts = NULL;
3209 char *debug_opts = NULL;
3210 size_t max_heap = 0;
3211 size_t soft_limit = 0;
3213 gboolean debug_print_allowance = FALSE;
3214 double allowance_ratio = 0, save_target = 0;
3215 gboolean cement_enabled = TRUE;
3218 result = InterlockedCompareExchange (&gc_initialized, -1, 0);
3221 /* already inited */
3224 /* being inited by another thread */
3225 mono_thread_info_usleep (1000);
3228 /* we will init it */
3231 g_assert_not_reached ();
3233 } while (result != 0);
3235 SGEN_TV_GETTIME (sgen_init_timestamp);
3237 #ifdef SGEN_WITHOUT_MONO
3238 mono_thread_smr_init ();
3241 mono_coop_mutex_init (&gc_mutex);
3243 gc_debug_file = stderr;
3245 mono_coop_mutex_init (&sgen_interruption_mutex);
3247 if ((env = g_getenv (MONO_GC_PARAMS_NAME)) || gc_params_options) {
3248 params_opts = g_strdup_printf ("%s,%s", gc_params_options ? gc_params_options : "", env ? env : "");
3253 opts = g_strsplit (params_opts, ",", -1);
3254 for (ptr = opts; *ptr; ++ptr) {
3256 if (g_str_has_prefix (opt, "major=")) {
3257 opt = strchr (opt, '=') + 1;
3258 sgen_major = parse_sgen_major (opt);
3259 } else if (g_str_has_prefix (opt, "minor=")) {
3260 opt = strchr (opt, '=') + 1;
3261 sgen_minor = parse_sgen_minor (opt);
3262 } else if (g_str_has_prefix (opt, "mode=")) {
3263 opt = strchr (opt, '=') + 1;
3264 sgen_mode = parse_sgen_mode (opt);
3272 sgen_init_internal_allocator ();
3273 sgen_init_nursery_allocator ();
3274 sgen_init_fin_weak_hash ();
3275 sgen_init_hash_table ();
3276 sgen_init_descriptors ();
3277 sgen_init_gray_queues ();
3278 sgen_init_allocator ();
3279 sgen_init_gchandles ();
3281 sgen_register_fixed_internal_mem_type (INTERNAL_MEM_SECTION, SGEN_SIZEOF_GC_MEM_SECTION);
3282 sgen_register_fixed_internal_mem_type (INTERNAL_MEM_GRAY_QUEUE, sizeof (GrayQueueSection));
3284 sgen_client_init ();
3286 if (sgen_mode != SGEN_MODE_NONE) {
3287 if (sgen_minor != SGEN_MINOR_DEFAULT || sgen_major != SGEN_MAJOR_DEFAULT)
3288 sgen_env_var_error (MONO_GC_PARAMS_NAME, "Ignoring major/minor configuration", "Major/minor configurations cannot be used with sgen modes");
3289 init_sgen_mode (sgen_mode);
3291 init_sgen_minor (sgen_minor);
3292 init_sgen_major (sgen_major);
3296 gboolean usage_printed = FALSE;
3298 for (ptr = opts; *ptr; ++ptr) {
3300 if (!strcmp (opt, ""))
3302 if (g_str_has_prefix (opt, "major="))
3304 if (g_str_has_prefix (opt, "minor="))
3306 if (g_str_has_prefix (opt, "mode=")) {
3307 if (g_str_has_prefix (opt, "mode=pause:")) {
3308 char *str_pause = strchr (opt, ':') + 1;
3309 int pause = atoi (str_pause);
3311 sgen_max_pause_time = pause;
3313 sgen_env_var_error (MONO_GC_PARAMS_NAME, "Using default", "Invalid maximum pause time for `pause` sgen mode");
3317 if (g_str_has_prefix (opt, "max-heap-size=")) {
3318 size_t page_size = mono_pagesize ();
3319 size_t max_heap_candidate = 0;
3320 opt = strchr (opt, '=') + 1;
3321 if (*opt && mono_gc_parse_environment_string_extract_number (opt, &max_heap_candidate)) {
3322 max_heap = (max_heap_candidate + page_size - 1) & ~(size_t)(page_size - 1);
3323 if (max_heap != max_heap_candidate)
3324 sgen_env_var_error (MONO_GC_PARAMS_NAME, "Rounding up.", "`max-heap-size` size must be a multiple of %d.", page_size);
3326 sgen_env_var_error (MONO_GC_PARAMS_NAME, NULL, "`max-heap-size` must be an integer.");
3330 if (g_str_has_prefix (opt, "soft-heap-limit=")) {
3331 opt = strchr (opt, '=') + 1;
3332 if (*opt && mono_gc_parse_environment_string_extract_number (opt, &soft_limit)) {
3333 if (soft_limit <= 0) {
3334 sgen_env_var_error (MONO_GC_PARAMS_NAME, NULL, "`soft-heap-limit` must be positive.");
3338 sgen_env_var_error (MONO_GC_PARAMS_NAME, NULL, "`soft-heap-limit` must be an integer.");
3342 if (g_str_has_prefix (opt, "nursery-size=")) {
3344 opt = strchr (opt, '=') + 1;
3345 if (*opt && mono_gc_parse_environment_string_extract_number (opt, &val)) {
3346 if ((val & (val - 1))) {
3347 sgen_env_var_error (MONO_GC_PARAMS_NAME, "Using default value.", "`nursery-size` must be a power of two.");
3351 if (val < SGEN_MAX_NURSERY_WASTE) {
3352 sgen_env_var_error (MONO_GC_PARAMS_NAME, "Using default value.",
3353 "`nursery-size` must be at least %d bytes.", SGEN_MAX_NURSERY_WASTE);
3357 min_nursery_size = max_nursery_size = val;
3358 dynamic_nursery = FALSE;
3360 sgen_env_var_error (MONO_GC_PARAMS_NAME, "Using default value.", "`nursery-size` must be an integer.");
3365 if (g_str_has_prefix (opt, "save-target-ratio=")) {
3367 opt = strchr (opt, '=') + 1;
3368 if (parse_double_in_interval (MONO_GC_PARAMS_NAME, "save-target-ratio", opt,
3369 SGEN_MIN_SAVE_TARGET_RATIO, SGEN_MAX_SAVE_TARGET_RATIO, &val)) {
3374 if (g_str_has_prefix (opt, "default-allowance-ratio=")) {
3376 opt = strchr (opt, '=') + 1;
3377 if (parse_double_in_interval (MONO_GC_PARAMS_NAME, "default-allowance-ratio", opt,
3378 SGEN_MIN_ALLOWANCE_NURSERY_SIZE_RATIO, SGEN_MAX_ALLOWANCE_NURSERY_SIZE_RATIO, &val)) {
3379 allowance_ratio = val;
3384 if (!strcmp (opt, "cementing")) {
3385 cement_enabled = TRUE;
3388 if (!strcmp (opt, "no-cementing")) {
3389 cement_enabled = FALSE;
3393 if (!strcmp (opt, "precleaning")) {
3394 precleaning_enabled = TRUE;
3397 if (!strcmp (opt, "no-precleaning")) {
3398 precleaning_enabled = FALSE;
3402 if (!strcmp (opt, "dynamic-nursery")) {
3403 if (sgen_minor_collector.is_split)
3404 sgen_env_var_error (MONO_GC_PARAMS_NAME, "Using default value.",
3405 "dynamic-nursery not supported with split-nursery.");
3407 dynamic_nursery = TRUE;
3410 if (!strcmp (opt, "no-dynamic-nursery")) {
3411 dynamic_nursery = FALSE;
3415 if (major_collector.handle_gc_param && major_collector.handle_gc_param (opt))
3418 if (sgen_minor_collector.handle_gc_param && sgen_minor_collector.handle_gc_param (opt))
3421 if (sgen_client_handle_gc_param (opt))
3424 sgen_env_var_error (MONO_GC_PARAMS_NAME, "Ignoring.", "Unknown option `%s`.", opt);
3429 fprintf (stderr, "\n%s must be a comma-delimited list of one or more of the following:\n", MONO_GC_PARAMS_NAME);
3430 fprintf (stderr, " max-heap-size=N (where N is an integer, possibly with a k, m or a g suffix)\n");
3431 fprintf (stderr, " soft-heap-limit=n (where N is an integer, possibly with a k, m or a g suffix)\n");
3432 fprintf (stderr, " mode=MODE (where MODE is 'balanced', 'throughput' or 'pause[:N]' and N is maximum pause in milliseconds)\n");
3433 fprintf (stderr, " nursery-size=N (where N is an integer, possibly with a k, m or a g suffix)\n");
3434 fprintf (stderr, " major=COLLECTOR (where COLLECTOR is `marksweep', `marksweep-conc', `marksweep-par')\n");
3435 fprintf (stderr, " minor=COLLECTOR (where COLLECTOR is `simple' or `split')\n");
3436 fprintf (stderr, " wbarrier=WBARRIER (where WBARRIER is `remset' or `cardtable')\n");
3437 fprintf (stderr, " [no-]cementing\n");
3438 fprintf (stderr, " [no-]dynamic-nursery\n");
3439 if (major_collector.print_gc_param_usage)
3440 major_collector.print_gc_param_usage ();
3441 if (sgen_minor_collector.print_gc_param_usage)
3442 sgen_minor_collector.print_gc_param_usage ();
3443 sgen_client_print_gc_params_usage ();
3444 fprintf (stderr, " Experimental options:\n");
3445 fprintf (stderr, " save-target-ratio=R (where R must be between %.2f - %.2f).\n", SGEN_MIN_SAVE_TARGET_RATIO, SGEN_MAX_SAVE_TARGET_RATIO);
3446 fprintf (stderr, " default-allowance-ratio=R (where R must be between %.2f - %.2f).\n", SGEN_MIN_ALLOWANCE_NURSERY_SIZE_RATIO, SGEN_MAX_ALLOWANCE_NURSERY_SIZE_RATIO);
3447 fprintf (stderr, "\n");
3449 usage_printed = TRUE;
3455 g_free (params_opts);
3457 alloc_nursery (dynamic_nursery, min_nursery_size, max_nursery_size);
3459 sgen_pinning_init ();
3460 sgen_cement_init (cement_enabled);
3462 if ((env = g_getenv (MONO_GC_DEBUG_NAME)) || gc_debug_options) {
3463 debug_opts = g_strdup_printf ("%s,%s", gc_debug_options ? gc_debug_options : "", env ? env : "");
3468 gboolean usage_printed = FALSE;
3470 opts = g_strsplit (debug_opts, ",", -1);
3471 for (ptr = opts; ptr && *ptr; ptr ++) {
3473 if (!strcmp (opt, ""))
3475 if (opt [0] >= '0' && opt [0] <= '9') {
3476 gc_debug_level = atoi (opt);
3481 char *rf = g_strdup_printf ("%s.%d", opt, mono_process_current_pid ());
3482 gc_debug_file = fopen (rf, "wb");
3484 gc_debug_file = stderr;
3487 } else if (!strcmp (opt, "print-allowance")) {
3488 debug_print_allowance = TRUE;
3489 } else if (!strcmp (opt, "print-pinning")) {
3490 sgen_pin_stats_enable ();
3491 } else if (!strcmp (opt, "verify-before-allocs")) {
3492 verify_before_allocs = 1;
3493 has_per_allocation_action = TRUE;
3494 } else if (g_str_has_prefix (opt, "max-valloc-size=")) {
3495 size_t max_valloc_size;
3496 char *arg = strchr (opt, '=') + 1;
3497 if (*opt && mono_gc_parse_environment_string_extract_number (arg, &max_valloc_size)) {
3498 mono_valloc_set_limit (max_valloc_size);
3500 sgen_env_var_error (MONO_GC_DEBUG_NAME, NULL, "`max-valloc-size` must be an integer.");
3503 } else if (g_str_has_prefix (opt, "verify-before-allocs=")) {
3504 char *arg = strchr (opt, '=') + 1;
3505 verify_before_allocs = atoi (arg);
3506 has_per_allocation_action = TRUE;
3507 } else if (!strcmp (opt, "collect-before-allocs")) {
3508 collect_before_allocs = 1;
3509 has_per_allocation_action = TRUE;
3510 } else if (g_str_has_prefix (opt, "collect-before-allocs=")) {
3511 char *arg = strchr (opt, '=') + 1;
3512 has_per_allocation_action = TRUE;
3513 collect_before_allocs = atoi (arg);
3514 } else if (!strcmp (opt, "verify-before-collections")) {
3515 whole_heap_check_before_collection = TRUE;
3516 } else if (!strcmp (opt, "check-remset-consistency")) {
3517 remset_consistency_checks = TRUE;
3518 nursery_clear_policy = CLEAR_AT_GC;
3519 } else if (!strcmp (opt, "mod-union-consistency-check")) {
3520 if (!major_collector.is_concurrent) {
3521 sgen_env_var_error (MONO_GC_DEBUG_NAME, "Ignoring.", "`mod-union-consistency-check` only works with concurrent major collector.");
3524 mod_union_consistency_check = TRUE;
3525 } else if (!strcmp (opt, "check-mark-bits")) {
3526 check_mark_bits_after_major_collection = TRUE;
3527 } else if (!strcmp (opt, "check-nursery-pinned")) {
3528 check_nursery_objects_pinned = TRUE;
3529 } else if (!strcmp (opt, "clear-at-gc")) {
3530 nursery_clear_policy = CLEAR_AT_GC;
3531 } else if (!strcmp (opt, "clear-nursery-at-gc")) {
3532 nursery_clear_policy = CLEAR_AT_GC;
3533 } else if (!strcmp (opt, "clear-at-tlab-creation")) {
3534 nursery_clear_policy = CLEAR_AT_TLAB_CREATION;
3535 } else if (!strcmp (opt, "debug-clear-at-tlab-creation")) {
3536 nursery_clear_policy = CLEAR_AT_TLAB_CREATION_DEBUG;
3537 } else if (!strcmp (opt, "check-scan-starts")) {
3538 do_scan_starts_check = TRUE;
3539 } else if (!strcmp (opt, "verify-nursery-at-minor-gc")) {
3540 do_verify_nursery = TRUE;
3541 } else if (!strcmp (opt, "check-concurrent")) {
3542 if (!major_collector.is_concurrent) {
3543 sgen_env_var_error (MONO_GC_DEBUG_NAME, "Ignoring.", "`check-concurrent` only works with concurrent major collectors.");
3546 nursery_clear_policy = CLEAR_AT_GC;
3547 do_concurrent_checks = TRUE;
3548 } else if (!strcmp (opt, "dump-nursery-at-minor-gc")) {
3549 do_dump_nursery_content = TRUE;
3550 } else if (!strcmp (opt, "disable-minor")) {
3551 disable_minor_collections = TRUE;
3552 } else if (!strcmp (opt, "disable-major")) {
3553 disable_major_collections = TRUE;
3554 } else if (g_str_has_prefix (opt, "heap-dump=")) {
3555 char *filename = strchr (opt, '=') + 1;
3556 nursery_clear_policy = CLEAR_AT_GC;
3557 sgen_debug_enable_heap_dump (filename);
3558 } else if (g_str_has_prefix (opt, "binary-protocol=")) {
3559 char *filename = strchr (opt, '=') + 1;
3560 char *colon = strrchr (filename, ':');
3563 if (!mono_gc_parse_environment_string_extract_number (colon + 1, &limit)) {
3564 sgen_env_var_error (MONO_GC_DEBUG_NAME, "Ignoring limit.", "Binary protocol file size limit must be an integer.");
3569 binary_protocol_init (filename, (long long)limit);
3570 } else if (!strcmp (opt, "nursery-canaries")) {
3571 do_verify_nursery = TRUE;
3572 enable_nursery_canaries = TRUE;
3573 } else if (!sgen_client_handle_gc_debug (opt)) {
3574 sgen_env_var_error (MONO_GC_DEBUG_NAME, "Ignoring.", "Unknown option `%s`.", opt);
3579 fprintf (stderr, "\n%s must be of the format [<l>[:<filename>]|<option>]+ where <l> is a debug level 0-9.\n", MONO_GC_DEBUG_NAME);
3580 fprintf (stderr, "Valid <option>s are:\n");
3581 fprintf (stderr, " collect-before-allocs[=<n>]\n");
3582 fprintf (stderr, " verify-before-allocs[=<n>]\n");
3583 fprintf (stderr, " max-valloc-size=N (where N is an integer, possibly with a k, m or a g suffix)\n");
3584 fprintf (stderr, " check-remset-consistency\n");
3585 fprintf (stderr, " check-mark-bits\n");
3586 fprintf (stderr, " check-nursery-pinned\n");
3587 fprintf (stderr, " verify-before-collections\n");
3588 fprintf (stderr, " verify-nursery-at-minor-gc\n");
3589 fprintf (stderr, " dump-nursery-at-minor-gc\n");
3590 fprintf (stderr, " disable-minor\n");
3591 fprintf (stderr, " disable-major\n");
3592 fprintf (stderr, " check-concurrent\n");
3593 fprintf (stderr, " clear-[nursery-]at-gc\n");
3594 fprintf (stderr, " clear-at-tlab-creation\n");
3595 fprintf (stderr, " debug-clear-at-tlab-creation\n");
3596 fprintf (stderr, " check-scan-starts\n");
3597 fprintf (stderr, " print-allowance\n");
3598 fprintf (stderr, " print-pinning\n");
3599 fprintf (stderr, " heap-dump=<filename>\n");
3600 fprintf (stderr, " binary-protocol=<filename>[:<file-size-limit>]\n");
3601 fprintf (stderr, " nursery-canaries\n");
3602 sgen_client_print_gc_debug_usage ();
3603 fprintf (stderr, "\n");
3605 usage_printed = TRUE;
3612 g_free (debug_opts);
3614 if (check_mark_bits_after_major_collection)
3615 nursery_clear_policy = CLEAR_AT_GC;
3617 if (major_collector.post_param_init)
3618 major_collector.post_param_init (&major_collector);
3620 sgen_thread_pool_start ();
3622 sgen_memgov_init (max_heap, soft_limit, debug_print_allowance, allowance_ratio, save_target);
3624 memset (&remset, 0, sizeof (remset));
3626 sgen_card_table_init (&remset);
3628 sgen_register_root (NULL, 0, sgen_make_user_root_descriptor (sgen_mark_normal_gc_handles), ROOT_TYPE_NORMAL, MONO_ROOT_SOURCE_GC_HANDLE, "normal gc handles");
3632 sgen_init_bridge ();
3636 sgen_gc_initialized ()
3638 return gc_initialized > 0;
3642 sgen_get_nursery_clear_policy (void)
3644 return nursery_clear_policy;
3650 mono_coop_mutex_lock (&gc_mutex);
3654 sgen_gc_unlock (void)
3656 mono_coop_mutex_unlock (&gc_mutex);
3660 sgen_major_collector_iterate_live_block_ranges (sgen_cardtable_block_callback callback)
3662 major_collector.iterate_live_block_ranges (callback);
3666 sgen_major_collector_iterate_block_ranges (sgen_cardtable_block_callback callback)
3668 major_collector.iterate_block_ranges (callback);
3672 sgen_get_major_collector (void)
3674 return &major_collector;
3678 sgen_get_minor_collector (void)
3680 return &sgen_minor_collector;
3684 sgen_get_remset (void)
3690 count_cards (long long *major_total, long long *major_marked, long long *los_total, long long *los_marked)
3692 sgen_get_major_collector ()->count_cards (major_total, major_marked);
3693 sgen_los_count_cards (los_total, los_marked);
3696 static gboolean world_is_stopped = FALSE;
3698 /* LOCKING: assumes the GC lock is held */
3700 sgen_stop_world (int generation)
3702 long long major_total = -1, major_marked = -1, los_total = -1, los_marked = -1;
3704 SGEN_ASSERT (0, !world_is_stopped, "Why are we stopping a stopped world?");
3706 binary_protocol_world_stopping (generation, sgen_timestamp (), (gpointer) (gsize) mono_native_thread_id_get ());
3708 sgen_client_stop_world (generation);
3710 world_is_stopped = TRUE;
3712 if (binary_protocol_is_heavy_enabled ())
3713 count_cards (&major_total, &major_marked, &los_total, &los_marked);
3714 binary_protocol_world_stopped (generation, sgen_timestamp (), major_total, major_marked, los_total, los_marked);
3717 /* LOCKING: assumes the GC lock is held */
3719 sgen_restart_world (int generation)
3721 long long major_total = -1, major_marked = -1, los_total = -1, los_marked = -1;
3724 SGEN_ASSERT (0, world_is_stopped, "Why are we restarting a running world?");
3726 if (binary_protocol_is_heavy_enabled ())
3727 count_cards (&major_total, &major_marked, &los_total, &los_marked);
3728 binary_protocol_world_restarting (generation, sgen_timestamp (), major_total, major_marked, los_total, los_marked);
3730 world_is_stopped = FALSE;
3732 sgen_client_restart_world (generation, &stw_time);
3734 binary_protocol_world_restarted (generation, sgen_timestamp ());
3736 if (sgen_client_bridge_need_processing ())
3737 sgen_client_bridge_processing_finish (generation);
3739 sgen_memgov_collection_end (generation, stw_time);
3743 sgen_is_world_stopped (void)
3745 return world_is_stopped;
3749 sgen_check_whole_heap_stw (void)
3751 sgen_stop_world (0);
3752 sgen_clear_nursery_fragments ();
3753 sgen_check_whole_heap (TRUE);
3754 sgen_restart_world (0);
3758 sgen_timestamp (void)
3760 SGEN_TV_DECLARE (timestamp);
3761 SGEN_TV_GETTIME (timestamp);
3762 return SGEN_TV_ELAPSED (sgen_init_timestamp, timestamp);
3765 #endif /* HAVE_SGEN_GC */