blob: 4868adc26eaccf4d5ec83cbddb2771d130fdbfde [file] [log] [blame]
// Copyright 2012 the V8 project authors. All rights reserved.
// Use of this source code is governed by a BSD-style license that can be
// found in the LICENSE file.
#include "src/heap/incremental-marking.h"
#include "src/code-stubs.h"
#include "src/compilation-cache.h"
#include "src/conversions.h"
#include "src/heap/concurrent-marking.h"
#include "src/heap/gc-idle-time-handler.h"
#include "src/heap/gc-tracer.h"
#include "src/heap/heap-inl.h"
#include "src/heap/mark-compact-inl.h"
#include "src/heap/object-stats.h"
#include "src/heap/objects-visiting-inl.h"
#include "src/heap/objects-visiting.h"
#include "src/heap/sweeper.h"
#include "src/tracing/trace-event.h"
#include "src/v8.h"
#include "src/visitors.h"
#include "src/vm-state-inl.h"
namespace v8 {
namespace internal {
using IncrementalMarkingMarkingVisitor =
MarkingVisitor<FixedArrayVisitationMode::kIncremental,
TraceRetainingPathMode::kDisabled,
IncrementalMarking::MarkingState>;
void IncrementalMarking::Observer::Step(int bytes_allocated, Address addr,
size_t size) {
Heap* heap = incremental_marking_.heap();
VMState<GC> state(heap->isolate());
RuntimeCallTimerScope runtime_timer(
heap->isolate(),
RuntimeCallCounterId::kGC_Custom_IncrementalMarkingObserver);
incremental_marking_.AdvanceIncrementalMarkingOnAllocation();
if (incremental_marking_.black_allocation() && addr != nullptr) {
// AdvanceIncrementalMarkingOnAllocation can start black allocation.
// Ensure that the new object is marked black.
HeapObject* object = HeapObject::FromAddress(addr);
if (incremental_marking_.marking_state()->IsWhite(object) &&
!heap->InNewSpace(object)) {
if (heap->lo_space()->Contains(object)) {
incremental_marking_.marking_state()->WhiteToBlack(object);
} else {
Page::FromAddress(addr)->CreateBlackArea(addr, addr + size);
}
}
}
}
IncrementalMarking::IncrementalMarking(
Heap* heap, MarkCompactCollector::MarkingWorklist* marking_worklist)
: heap_(heap),
marking_worklist_(marking_worklist),
initial_old_generation_size_(0),
bytes_marked_ahead_of_schedule_(0),
bytes_marked_concurrently_(0),
unscanned_bytes_of_large_object_(0),
is_compacting_(false),
should_hurry_(false),
was_activated_(false),
black_allocation_(false),
finalize_marking_completed_(false),
trace_wrappers_toggle_(false),
request_type_(NONE),
new_generation_observer_(*this, kYoungGenerationAllocatedThreshold),
old_generation_observer_(*this, kOldGenerationAllocatedThreshold) {
DCHECK_NOT_NULL(marking_worklist_);
SetState(STOPPED);
}
bool IncrementalMarking::BaseRecordWrite(HeapObject* obj, Object* value) {
HeapObject* value_heap_obj = HeapObject::cast(value);
DCHECK(!marking_state()->IsImpossible(value_heap_obj));
DCHECK(!marking_state()->IsImpossible(obj));
#ifdef V8_CONCURRENT_MARKING
// The write barrier stub generated with V8_CONCURRENT_MARKING does not
// check the color of the source object.
const bool need_recording = true;
#else
const bool need_recording = marking_state()->IsBlack(obj);
#endif
if (need_recording && WhiteToGreyAndPush(value_heap_obj)) {
RestartIfNotMarking();
}
return is_compacting_ && need_recording;
}
void IncrementalMarking::RecordWriteSlow(HeapObject* obj, Object** slot,
Object* value) {
if (BaseRecordWrite(obj, value) && slot != nullptr) {
// Object is not going to be rescanned we need to record the slot.
heap_->mark_compact_collector()->RecordSlot(obj, slot, value);
}
}
int IncrementalMarking::RecordWriteFromCode(HeapObject* obj, Object** slot,
Isolate* isolate) {
DCHECK(obj->IsHeapObject());
isolate->heap()->incremental_marking()->RecordWrite(obj, slot, *slot);
// Called by RecordWriteCodeStubAssembler, which doesnt accept void type
return 0;
}
void IncrementalMarking::RecordWriteIntoCodeSlow(Code* host, RelocInfo* rinfo,
Object* value) {
if (BaseRecordWrite(host, value)) {
// Object is not going to be rescanned. We need to record the slot.
heap_->mark_compact_collector()->RecordRelocSlot(host, rinfo, value);
}
}
bool IncrementalMarking::WhiteToGreyAndPush(HeapObject* obj) {
if (marking_state()->WhiteToGrey(obj)) {
marking_worklist()->Push(obj);
return true;
}
return false;
}
void IncrementalMarking::MarkBlackAndPush(HeapObject* obj) {
// Color the object black and push it into the bailout deque.
marking_state()->WhiteToGrey(obj);
if (marking_state()->GreyToBlack(obj)) {
if (FLAG_concurrent_marking) {
marking_worklist()->PushBailout(obj);
} else {
marking_worklist()->Push(obj);
}
}
}
void IncrementalMarking::NotifyLeftTrimming(HeapObject* from, HeapObject* to) {
DCHECK(IsMarking());
DCHECK(MemoryChunk::FromAddress(from->address())->SweepingDone());
DCHECK_EQ(MemoryChunk::FromAddress(from->address()),
MemoryChunk::FromAddress(to->address()));
DCHECK_NE(from, to);
MarkBit old_mark_bit = marking_state()->MarkBitFrom(from);
MarkBit new_mark_bit = marking_state()->MarkBitFrom(to);
if (black_allocation() && Marking::IsBlack<kAtomicity>(new_mark_bit)) {
// Nothing to do if the object is in black area.
return;
}
bool marked_black_due_to_left_trimming = false;
if (FLAG_concurrent_marking) {
// We need to mark the array black before overwriting its map and length
// so that the concurrent marker does not observe inconsistent state.
Marking::WhiteToGrey<kAtomicity>(old_mark_bit);
if (Marking::GreyToBlack<kAtomicity>(old_mark_bit)) {
// The concurrent marker will not mark the array. We need to push the
// new array start in marking deque to ensure that it will be marked.
marked_black_due_to_left_trimming = true;
}
DCHECK(Marking::IsBlack<kAtomicity>(old_mark_bit));
}
if (Marking::IsBlack<kAtomicity>(old_mark_bit) &&
!marked_black_due_to_left_trimming) {
// The array was black before left trimming or was marked black by the
// concurrent marker. Simply transfer the color.
if (from->address() + kPointerSize == to->address()) {
// The old and the new markbits overlap. The |to| object has the
// grey color. To make it black, we need to set the second bit.
DCHECK(new_mark_bit.Get<kAtomicity>());
new_mark_bit.Next().Set<kAtomicity>();
} else {
bool success = Marking::WhiteToBlack<kAtomicity>(new_mark_bit);
DCHECK(success);
USE(success);
}
} else if (Marking::IsGrey<kAtomicity>(old_mark_bit) ||
marked_black_due_to_left_trimming) {
// The array was already grey or was marked black by this function.
// Mark the new array grey and push it to marking deque.
if (from->address() + kPointerSize == to->address()) {
// The old and the new markbits overlap. The |to| object is either white
// or grey. Set the first bit to make sure that it is grey.
new_mark_bit.Set<kAtomicity>();
DCHECK(!new_mark_bit.Next().Get<kAtomicity>());
} else {
bool success = Marking::WhiteToGrey<kAtomicity>(new_mark_bit);
DCHECK(success);
USE(success);
}
marking_worklist()->Push(to);
RestartIfNotMarking();
}
}
class IncrementalMarkingRootMarkingVisitor : public RootVisitor {
public:
explicit IncrementalMarkingRootMarkingVisitor(
IncrementalMarking* incremental_marking)
: heap_(incremental_marking->heap()) {}
void VisitRootPointer(Root root, Object** p) override {
MarkObjectByPointer(p);
}
void VisitRootPointers(Root root, Object** start, Object** end) override {
for (Object** p = start; p < end; p++) MarkObjectByPointer(p);
}
private:
void MarkObjectByPointer(Object** p) {
Object* obj = *p;
if (!obj->IsHeapObject()) return;
heap_->incremental_marking()->WhiteToGreyAndPush(HeapObject::cast(obj));
}
Heap* heap_;
};
void IncrementalMarking::SetOldSpacePageFlags(MemoryChunk* chunk,
bool is_marking,
bool is_compacting) {
if (is_marking) {
chunk->SetFlag(MemoryChunk::POINTERS_TO_HERE_ARE_INTERESTING);
chunk->SetFlag(MemoryChunk::POINTERS_FROM_HERE_ARE_INTERESTING);
} else {
chunk->ClearFlag(MemoryChunk::POINTERS_TO_HERE_ARE_INTERESTING);
chunk->SetFlag(MemoryChunk::POINTERS_FROM_HERE_ARE_INTERESTING);
}
}
void IncrementalMarking::SetNewSpacePageFlags(MemoryChunk* chunk,
bool is_marking) {
chunk->SetFlag(MemoryChunk::POINTERS_TO_HERE_ARE_INTERESTING);
if (is_marking) {
chunk->SetFlag(MemoryChunk::POINTERS_FROM_HERE_ARE_INTERESTING);
} else {
chunk->ClearFlag(MemoryChunk::POINTERS_FROM_HERE_ARE_INTERESTING);
}
}
void IncrementalMarking::DeactivateIncrementalWriteBarrierForSpace(
PagedSpace* space) {
for (Page* p : *space) {
SetOldSpacePageFlags(p, false, false);
}
}
void IncrementalMarking::DeactivateIncrementalWriteBarrierForSpace(
NewSpace* space) {
for (Page* p : *space) {
SetNewSpacePageFlags(p, false);
}
}
void IncrementalMarking::DeactivateIncrementalWriteBarrier() {
DeactivateIncrementalWriteBarrierForSpace(heap_->old_space());
DeactivateIncrementalWriteBarrierForSpace(heap_->map_space());
DeactivateIncrementalWriteBarrierForSpace(heap_->code_space());
DeactivateIncrementalWriteBarrierForSpace(heap_->new_space());
for (LargePage* lop : *heap_->lo_space()) {
SetOldSpacePageFlags(lop, false, false);
}
}
void IncrementalMarking::ActivateIncrementalWriteBarrier(PagedSpace* space) {
for (Page* p : *space) {
SetOldSpacePageFlags(p, true, is_compacting_);
}
}
void IncrementalMarking::ActivateIncrementalWriteBarrier(NewSpace* space) {
for (Page* p : *space) {
SetNewSpacePageFlags(p, true);
}
}
void IncrementalMarking::ActivateIncrementalWriteBarrier() {
ActivateIncrementalWriteBarrier(heap_->old_space());
ActivateIncrementalWriteBarrier(heap_->map_space());
ActivateIncrementalWriteBarrier(heap_->code_space());
ActivateIncrementalWriteBarrier(heap_->new_space());
for (LargePage* lop : *heap_->lo_space()) {
SetOldSpacePageFlags(lop, true, is_compacting_);
}
}
bool IncrementalMarking::WasActivated() { return was_activated_; }
bool IncrementalMarking::CanBeActivated() {
// Only start incremental marking in a safe state: 1) when incremental
// marking is turned on, 2) when we are currently not in a GC, and
// 3) when we are currently not serializing or deserializing the heap.
return FLAG_incremental_marking && heap_->gc_state() == Heap::NOT_IN_GC &&
heap_->deserialization_complete() &&
!heap_->isolate()->serializer_enabled();
}
void IncrementalMarking::Deactivate() {
DeactivateIncrementalWriteBarrier();
}
void IncrementalMarking::Start(GarbageCollectionReason gc_reason) {
if (FLAG_trace_incremental_marking) {
int old_generation_size_mb =
static_cast<int>(heap()->PromotedSpaceSizeOfObjects() / MB);
int old_generation_limit_mb =
static_cast<int>(heap()->old_generation_allocation_limit() / MB);
heap()->isolate()->PrintWithTimestamp(
"[IncrementalMarking] Start (%s): old generation %dMB, limit %dMB, "
"slack %dMB\n",
Heap::GarbageCollectionReasonToString(gc_reason),
old_generation_size_mb, old_generation_limit_mb,
Max(0, old_generation_limit_mb - old_generation_size_mb));
}
DCHECK(FLAG_incremental_marking);
DCHECK(state_ == STOPPED);
DCHECK(heap_->gc_state() == Heap::NOT_IN_GC);
DCHECK(!heap_->isolate()->serializer_enabled());
Counters* counters = heap_->isolate()->counters();
counters->incremental_marking_reason()->AddSample(
static_cast<int>(gc_reason));
HistogramTimerScope incremental_marking_scope(
counters->gc_incremental_marking_start());
TRACE_EVENT0("v8", "V8.GCIncrementalMarkingStart");
TRACE_GC(heap()->tracer(), GCTracer::Scope::MC_INCREMENTAL_START);
heap_->tracer()->NotifyIncrementalMarkingStart();
start_time_ms_ = heap()->MonotonicallyIncreasingTimeInMs();
initial_old_generation_size_ = heap_->PromotedSpaceSizeOfObjects();
old_generation_allocation_counter_ = heap_->OldGenerationAllocationCounter();
bytes_allocated_ = 0;
bytes_marked_ahead_of_schedule_ = 0;
bytes_marked_concurrently_ = 0;
should_hurry_ = false;
was_activated_ = true;
if (!heap_->mark_compact_collector()->sweeping_in_progress()) {
StartMarking();
} else {
if (FLAG_trace_incremental_marking) {
heap()->isolate()->PrintWithTimestamp(
"[IncrementalMarking] Start sweeping.\n");
}
SetState(SWEEPING);
}
heap_->AddAllocationObserversToAllSpaces(&old_generation_observer_,
&new_generation_observer_);
incremental_marking_job()->Start(heap_);
}
void IncrementalMarking::StartMarking() {
if (heap_->isolate()->serializer_enabled()) {
// Black allocation currently starts when we start incremental marking,
// but we cannot enable black allocation while deserializing. Hence, we
// have to delay the start of incremental marking in that case.
if (FLAG_trace_incremental_marking) {
heap()->isolate()->PrintWithTimestamp(
"[IncrementalMarking] Start delayed - serializer\n");
}
return;
}
if (FLAG_trace_incremental_marking) {
heap()->isolate()->PrintWithTimestamp(
"[IncrementalMarking] Start marking\n");
}
is_compacting_ =
!FLAG_never_compact && heap_->mark_compact_collector()->StartCompaction();
SetState(MARKING);
{
TRACE_GC(heap()->tracer(),
GCTracer::Scope::MC_INCREMENTAL_WRAPPER_PROLOGUE);
heap_->local_embedder_heap_tracer()->TracePrologue();
}
ActivateIncrementalWriteBarrier();
// Marking bits are cleared by the sweeper.
#ifdef VERIFY_HEAP
if (FLAG_verify_heap) {
heap_->mark_compact_collector()->VerifyMarkbitsAreClean();
}
#endif
heap_->isolate()->compilation_cache()->MarkCompactPrologue();
#ifdef V8_CONCURRENT_MARKING
// The write-barrier does not check the color of the source object.
// Start black allocation earlier to ensure faster marking progress.
if (!black_allocation_) {
StartBlackAllocation();
}
#endif
// Mark strong roots grey.
IncrementalMarkingRootMarkingVisitor visitor(this);
heap_->IterateStrongRoots(&visitor, VISIT_ONLY_STRONG);
if (FLAG_concurrent_marking && heap_->use_tasks()) {
heap_->concurrent_marking()->ScheduleTasks();
}
// Ready to start incremental marking.
if (FLAG_trace_incremental_marking) {
heap()->isolate()->PrintWithTimestamp("[IncrementalMarking] Running\n");
}
}
void IncrementalMarking::StartBlackAllocation() {
DCHECK(FLAG_black_allocation);
DCHECK(!black_allocation_);
DCHECK(IsMarking());
black_allocation_ = true;
heap()->old_space()->MarkLinearAllocationAreaBlack();
heap()->map_space()->MarkLinearAllocationAreaBlack();
heap()->code_space()->MarkLinearAllocationAreaBlack();
if (FLAG_trace_incremental_marking) {
heap()->isolate()->PrintWithTimestamp(
"[IncrementalMarking] Black allocation started\n");
}
}
void IncrementalMarking::PauseBlackAllocation() {
DCHECK(FLAG_black_allocation);
DCHECK(IsMarking());
heap()->old_space()->UnmarkLinearAllocationArea();
heap()->map_space()->UnmarkLinearAllocationArea();
heap()->code_space()->UnmarkLinearAllocationArea();
if (FLAG_trace_incremental_marking) {
heap()->isolate()->PrintWithTimestamp(
"[IncrementalMarking] Black allocation paused\n");
}
black_allocation_ = false;
}
void IncrementalMarking::FinishBlackAllocation() {
if (black_allocation_) {
black_allocation_ = false;
if (FLAG_trace_incremental_marking) {
heap()->isolate()->PrintWithTimestamp(
"[IncrementalMarking] Black allocation finished\n");
}
}
}
void IncrementalMarking::AbortBlackAllocation() {
if (FLAG_trace_incremental_marking) {
heap()->isolate()->PrintWithTimestamp(
"[IncrementalMarking] Black allocation aborted\n");
}
}
void IncrementalMarking::MarkRoots() {
DCHECK(!finalize_marking_completed_);
DCHECK(IsMarking());
IncrementalMarkingRootMarkingVisitor visitor(this);
heap_->IterateStrongRoots(&visitor, VISIT_ONLY_STRONG);
}
bool ShouldRetainMap(Map* map, int age) {
if (age == 0) {
// The map has aged. Do not retain this map.
return false;
}
Object* constructor = map->GetConstructor();
Heap* heap = map->GetHeap();
if (!constructor->IsHeapObject() ||
heap->incremental_marking()->marking_state()->IsWhite(
HeapObject::cast(constructor))) {
// The constructor is dead, no new objects with this map can
// be created. Do not retain this map.
return false;
}
return true;
}
void IncrementalMarking::RetainMaps() {
// Do not retain dead maps if flag disables it or there is
// - memory pressure (reduce_memory_footprint_),
// - GC is requested by tests or dev-tools (abort_incremental_marking_).
bool map_retaining_is_disabled = heap()->ShouldReduceMemory() ||
heap()->ShouldAbortIncrementalMarking() ||
FLAG_retain_maps_for_n_gc == 0;
ArrayList* retained_maps = heap()->retained_maps();
int length = retained_maps->Length();
// The number_of_disposed_maps separates maps in the retained_maps
// array that were created before and after context disposal.
// We do not age and retain disposed maps to avoid memory leaks.
int number_of_disposed_maps = heap()->number_of_disposed_maps_;
for (int i = 0; i < length; i += 2) {
DCHECK(retained_maps->Get(i)->IsWeakCell());
WeakCell* cell = WeakCell::cast(retained_maps->Get(i));
if (cell->cleared()) continue;
int age = Smi::ToInt(retained_maps->Get(i + 1));
int new_age;
Map* map = Map::cast(cell->value());
if (i >= number_of_disposed_maps && !map_retaining_is_disabled &&
marking_state()->IsWhite(map)) {
if (ShouldRetainMap(map, age)) {
WhiteToGreyAndPush(map);
}
Object* prototype = map->prototype();
if (age > 0 && prototype->IsHeapObject() &&
marking_state()->IsWhite(HeapObject::cast(prototype))) {
// The prototype is not marked, age the map.
new_age = age - 1;
} else {
// The prototype and the constructor are marked, this map keeps only
// transition tree alive, not JSObjects. Do not age the map.
new_age = age;
}
} else {
new_age = FLAG_retain_maps_for_n_gc;
}
// Compact the array and update the age.
if (new_age != age) {
retained_maps->Set(i + 1, Smi::FromInt(new_age));
}
}
}
void IncrementalMarking::FinalizeIncrementally() {
TRACE_GC(heap()->tracer(), GCTracer::Scope::MC_INCREMENTAL_FINALIZE_BODY);
DCHECK(!finalize_marking_completed_);
DCHECK(IsMarking());
double start = heap_->MonotonicallyIncreasingTimeInMs();
// After finishing incremental marking, we try to discover all unmarked
// objects to reduce the marking load in the final pause.
// 1) We scan and mark the roots again to find all changes to the root set.
// 2) Age and retain maps embedded in optimized code.
// 3) Remove weak cell with live values from the list of weak cells, they
// do not need processing during GC.
MarkRoots();
// Map retaining is needed for perfromance, not correctness,
// so we can do it only once at the beginning of the finalization.
RetainMaps();
finalize_marking_completed_ = true;
if (FLAG_black_allocation && !heap()->ShouldReduceMemory() &&
!black_allocation_) {
// TODO(hpayer): Move to an earlier point as soon as we make faster marking
// progress.
StartBlackAllocation();
}
if (FLAG_trace_incremental_marking) {
double end = heap_->MonotonicallyIncreasingTimeInMs();
double delta = end - start;
heap()->isolate()->PrintWithTimestamp(
"[IncrementalMarking] Finalize incrementally spent %.1f ms.\n", delta);
}
}
void IncrementalMarking::UpdateMarkingWorklistAfterScavenge() {
if (!IsMarking()) return;
Map* filler_map = heap_->one_pointer_filler_map();
MinorMarkCompactCollector::MarkingState* minor_marking_state =
heap()->minor_mark_compact_collector()->marking_state();
marking_worklist()->Update([this, filler_map, minor_marking_state](
HeapObject* obj, HeapObject** out) -> bool {
DCHECK(obj->IsHeapObject());
// Only pointers to from space have to be updated.
if (heap_->InFromSpace(obj)) {
MapWord map_word = obj->map_word();
if (!map_word.IsForwardingAddress()) {
// There may be objects on the marking deque that do not exist anymore,
// e.g. left trimmed objects or objects from the root set (frames).
// If these object are dead at scavenging time, their marking deque
// entries will not point to forwarding addresses. Hence, we can discard
// them.
return false;
}
HeapObject* dest = map_word.ToForwardingAddress();
DCHECK_IMPLIES(marking_state()->IsWhite(obj), obj->IsFiller());
*out = dest;
return true;
} else if (heap_->InToSpace(obj)) {
// The object may be on a page that was moved in new space.
DCHECK(
Page::FromAddress(obj->address())->IsFlagSet(Page::SWEEP_TO_ITERATE));
if (minor_marking_state->IsGrey(obj)) {
*out = obj;
return true;
}
return false;
} else {
// The object may be on a page that was moved from new to old space.
if (Page::FromAddress(obj->address())
->IsFlagSet(Page::SWEEP_TO_ITERATE)) {
if (minor_marking_state->IsGrey(obj)) {
*out = obj;
return true;
}
return false;
}
DCHECK_IMPLIES(marking_state()->IsWhite(obj), obj->IsFiller());
// Skip one word filler objects that appear on the
// stack when we perform in place array shift.
if (obj->map() != filler_map) {
*out = obj;
return true;
}
return false;
}
});
}
void IncrementalMarking::UpdateMarkedBytesAfterScavenge(
size_t dead_bytes_in_new_space) {
if (!IsMarking()) return;
bytes_marked_ahead_of_schedule_ -=
Min(bytes_marked_ahead_of_schedule_, dead_bytes_in_new_space);
}
bool IncrementalMarking::IsFixedArrayWithProgressBar(HeapObject* obj) {
if (!obj->IsFixedArray()) return false;
MemoryChunk* chunk = MemoryChunk::FromAddress(obj->address());
return chunk->IsFlagSet(MemoryChunk::HAS_PROGRESS_BAR);
}
int IncrementalMarking::VisitObject(Map* map, HeapObject* obj) {
DCHECK(marking_state()->IsGrey(obj) || marking_state()->IsBlack(obj));
// The object can already be black in two cases:
// 1. The object is a fixed array with the progress bar.
// 2. The object is a JSObject that was colored black before
// unsafe layout change.
// 3. The object is a string that was colored black before
// unsafe layout change.
if (!marking_state()->GreyToBlack(obj)) {
DCHECK(IsFixedArrayWithProgressBar(obj) || obj->IsJSObject() ||
obj->IsString());
}
DCHECK(marking_state()->IsBlack(obj));
WhiteToGreyAndPush(map);
IncrementalMarkingMarkingVisitor visitor(heap()->mark_compact_collector(),
marking_state());
return visitor.Visit(map, obj);
}
void IncrementalMarking::ProcessBlackAllocatedObject(HeapObject* obj) {
if (IsMarking() && marking_state()->IsBlack(obj)) {
RevisitObject(obj);
}
}
void IncrementalMarking::RevisitObject(HeapObject* obj) {
DCHECK(IsMarking());
DCHECK(FLAG_concurrent_marking || marking_state()->IsBlack(obj));
Page* page = Page::FromAddress(obj->address());
if (page->owner()->identity() == LO_SPACE) {
page->ResetProgressBar();
}
Map* map = obj->map();
WhiteToGreyAndPush(map);
IncrementalMarkingMarkingVisitor visitor(heap()->mark_compact_collector(),
marking_state());
visitor.Visit(map, obj);
}
template <WorklistToProcess worklist_to_process>
intptr_t IncrementalMarking::ProcessMarkingWorklist(
intptr_t bytes_to_process, ForceCompletionAction completion) {
intptr_t bytes_processed = 0;
while (bytes_processed < bytes_to_process || completion == FORCE_COMPLETION) {
HeapObject* obj;
if (worklist_to_process == WorklistToProcess::kBailout) {
obj = marking_worklist()->PopBailout();
} else {
obj = marking_worklist()->Pop();
}
if (obj == nullptr) break;
// Left trimming may result in white, grey, or black filler objects on the
// marking deque. Ignore these objects.
if (obj->IsFiller()) {
DCHECK(!marking_state()->IsImpossible(obj));
continue;
}
unscanned_bytes_of_large_object_ = 0;
int size = VisitObject(obj->map(), obj);
bytes_processed += size - unscanned_bytes_of_large_object_;
}
// Report all found wrappers to the embedder. This is necessary as the
// embedder could potentially invalidate wrappers as soon as V8 is done
// with its incremental marking processing. Any cached wrappers could
// result in broken pointers at this point.
heap_->local_embedder_heap_tracer()->RegisterWrappersWithRemoteTracer();
return bytes_processed;
}
void IncrementalMarking::Hurry() {
// A scavenge may have pushed new objects on the marking deque (due to black
// allocation) even in COMPLETE state. This may happen if scavenges are
// forced e.g. in tests. It should not happen when COMPLETE was set when
// incremental marking finished and a regular GC was triggered after that
// because should_hurry_ will force a full GC.
if (!marking_worklist()->IsEmpty()) {
double start = 0.0;
if (FLAG_trace_incremental_marking) {
start = heap_->MonotonicallyIncreasingTimeInMs();
if (FLAG_trace_incremental_marking) {
heap()->isolate()->PrintWithTimestamp("[IncrementalMarking] Hurry\n");
}
}
// TODO(gc) hurry can mark objects it encounters black as mutator
// was stopped.
ProcessMarkingWorklist(0, FORCE_COMPLETION);
SetState(COMPLETE);
if (FLAG_trace_incremental_marking) {
double end = heap_->MonotonicallyIncreasingTimeInMs();
double delta = end - start;
if (FLAG_trace_incremental_marking) {
heap()->isolate()->PrintWithTimestamp(
"[IncrementalMarking] Complete (hurry), spent %d ms.\n",
static_cast<int>(delta));
}
}
}
}
void IncrementalMarking::Stop() {
if (IsStopped()) return;
if (FLAG_trace_incremental_marking) {
int old_generation_size_mb =
static_cast<int>(heap()->PromotedSpaceSizeOfObjects() / MB);
int old_generation_limit_mb =
static_cast<int>(heap()->old_generation_allocation_limit() / MB);
heap()->isolate()->PrintWithTimestamp(
"[IncrementalMarking] Stopping: old generation %dMB, limit %dMB, "
"overshoot %dMB\n",
old_generation_size_mb, old_generation_limit_mb,
Max(0, old_generation_size_mb - old_generation_limit_mb));
}
SpaceIterator it(heap_);
while (it.has_next()) {
Space* space = it.next();
if (space == heap_->new_space()) {
space->RemoveAllocationObserver(&new_generation_observer_);
} else {
space->RemoveAllocationObserver(&old_generation_observer_);
}
}
IncrementalMarking::set_should_hurry(false);
heap_->isolate()->stack_guard()->ClearGC();
SetState(STOPPED);
is_compacting_ = false;
FinishBlackAllocation();
}
void IncrementalMarking::Finalize() {
Hurry();
Stop();
}
void IncrementalMarking::FinalizeMarking(CompletionAction action) {
DCHECK(!finalize_marking_completed_);
if (FLAG_trace_incremental_marking) {
heap()->isolate()->PrintWithTimestamp(
"[IncrementalMarking] requesting finalization of incremental "
"marking.\n");
}
request_type_ = FINALIZATION;
if (action == GC_VIA_STACK_GUARD) {
heap_->isolate()->stack_guard()->RequestGC();
}
}
void IncrementalMarking::MarkingComplete(CompletionAction action) {
SetState(COMPLETE);
// We will set the stack guard to request a GC now. This will mean the rest
// of the GC gets performed as soon as possible (we can't do a GC here in a
// record-write context). If a few things get allocated between now and then
// that shouldn't make us do a scavenge and keep being incremental, so we set
// the should-hurry flag to indicate that there can't be much work left to do.
set_should_hurry(true);
if (FLAG_trace_incremental_marking) {
heap()->isolate()->PrintWithTimestamp(
"[IncrementalMarking] Complete (normal).\n");
}
request_type_ = COMPLETE_MARKING;
if (action == GC_VIA_STACK_GUARD) {
heap_->isolate()->stack_guard()->RequestGC();
}
}
void IncrementalMarking::Epilogue() {
was_activated_ = false;
finalize_marking_completed_ = false;
}
double IncrementalMarking::AdvanceIncrementalMarking(
double deadline_in_ms, CompletionAction completion_action,
StepOrigin step_origin) {
HistogramTimerScope incremental_marking_scope(
heap_->isolate()->counters()->gc_incremental_marking());
TRACE_EVENT0("v8", "V8.GCIncrementalMarking");
TRACE_GC(heap_->tracer(), GCTracer::Scope::MC_INCREMENTAL);
DCHECK(!IsStopped());
DCHECK_EQ(
0, heap_->local_embedder_heap_tracer()->NumberOfCachedWrappersToTrace());
double remaining_time_in_ms = 0.0;
intptr_t step_size_in_bytes = GCIdleTimeHandler::EstimateMarkingStepSize(
kStepSizeInMs,
heap()->tracer()->IncrementalMarkingSpeedInBytesPerMillisecond());
const bool incremental_wrapper_tracing =
state_ == MARKING && FLAG_incremental_marking_wrappers &&
heap_->local_embedder_heap_tracer()->InUse();
do {
if (incremental_wrapper_tracing && trace_wrappers_toggle_) {
TRACE_GC(heap()->tracer(),
GCTracer::Scope::MC_INCREMENTAL_WRAPPER_TRACING);
const double wrapper_deadline =
heap_->MonotonicallyIncreasingTimeInMs() + kStepSizeInMs;
if (!heap_->local_embedder_heap_tracer()
->ShouldFinalizeIncrementalMarking()) {
heap_->local_embedder_heap_tracer()->Trace(
wrapper_deadline, EmbedderHeapTracer::AdvanceTracingActions(
EmbedderHeapTracer::ForceCompletionAction::
DO_NOT_FORCE_COMPLETION));
}
} else {
Step(step_size_in_bytes, completion_action, step_origin);
}
trace_wrappers_toggle_ = !trace_wrappers_toggle_;
remaining_time_in_ms =
deadline_in_ms - heap()->MonotonicallyIncreasingTimeInMs();
} while (remaining_time_in_ms >= kStepSizeInMs && !IsComplete() &&
!marking_worklist()->IsEmpty());
return remaining_time_in_ms;
}
void IncrementalMarking::FinalizeSweeping() {
DCHECK(state_ == SWEEPING);
if (heap_->mark_compact_collector()->sweeping_in_progress() &&
(!FLAG_concurrent_sweeping ||
!heap_->mark_compact_collector()->sweeper()->AreSweeperTasksRunning())) {
heap_->mark_compact_collector()->EnsureSweepingCompleted();
}
if (!heap_->mark_compact_collector()->sweeping_in_progress()) {
#ifdef DEBUG
heap_->VerifyCountersAfterSweeping();
#endif
StartMarking();
}
}
size_t IncrementalMarking::StepSizeToKeepUpWithAllocations() {
// Update bytes_allocated_ based on the allocation counter.
size_t current_counter = heap_->OldGenerationAllocationCounter();
bytes_allocated_ += current_counter - old_generation_allocation_counter_;
old_generation_allocation_counter_ = current_counter;
return bytes_allocated_;
}
size_t IncrementalMarking::StepSizeToMakeProgress() {
// We increase step size gradually based on the time passed in order to
// leave marking work to standalone tasks. The ramp up duration and the
// target step count are chosen based on benchmarks.
const int kRampUpIntervalMs = 300;
const size_t kTargetStepCount = 256;
const size_t kTargetStepCountAtOOM = 32;
size_t oom_slack = heap()->new_space()->Capacity() + 64 * MB;
if (heap()->IsCloseToOutOfMemory(oom_slack)) {
return heap()->PromotedSpaceSizeOfObjects() / kTargetStepCountAtOOM;
}
size_t step_size = Max(initial_old_generation_size_ / kTargetStepCount,
IncrementalMarking::kMinStepSizeInBytes);
double time_passed_ms =
heap_->MonotonicallyIncreasingTimeInMs() - start_time_ms_;
double factor = Min(time_passed_ms / kRampUpIntervalMs, 1.0);
return static_cast<size_t>(factor * step_size);
}
void IncrementalMarking::AdvanceIncrementalMarkingOnAllocation() {
// Code using an AlwaysAllocateScope assumes that the GC state does not
// change; that implies that no marking steps must be performed.
if (heap_->gc_state() != Heap::NOT_IN_GC || !FLAG_incremental_marking ||
(state_ != SWEEPING && state_ != MARKING) || heap_->always_allocate()) {
return;
}
size_t bytes_to_process =
StepSizeToKeepUpWithAllocations() + StepSizeToMakeProgress();
if (bytes_to_process >= IncrementalMarking::kMinStepSizeInBytes) {
HistogramTimerScope incremental_marking_scope(
heap_->isolate()->counters()->gc_incremental_marking());
TRACE_EVENT0("v8", "V8.GCIncrementalMarking");
TRACE_GC(heap_->tracer(), GCTracer::Scope::MC_INCREMENTAL);
// The first step after Scavenge will see many allocated bytes.
// Cap the step size to distribute the marking work more uniformly.
size_t max_step_size = GCIdleTimeHandler::EstimateMarkingStepSize(
kMaxStepSizeInMs,
heap()->tracer()->IncrementalMarkingSpeedInBytesPerMillisecond());
bytes_to_process = Min(bytes_to_process, max_step_size);
size_t bytes_processed = 0;
if (FLAG_concurrent_marking) {
bytes_processed = Step(bytes_to_process, GC_VIA_STACK_GUARD,
StepOrigin::kV8, WorklistToProcess::kBailout);
bytes_to_process = (bytes_processed >= bytes_to_process)
? 0
: bytes_to_process - bytes_processed;
size_t current_bytes_marked_concurrently =
heap()->concurrent_marking()->TotalMarkedBytes();
// The concurrent_marking()->TotalMarkedBytes() is not monothonic for a
// short period of time when a concurrent marking task is finishing.
if (current_bytes_marked_concurrently > bytes_marked_concurrently_) {
bytes_marked_ahead_of_schedule_ +=
current_bytes_marked_concurrently - bytes_marked_concurrently_;
bytes_marked_concurrently_ = current_bytes_marked_concurrently;
}
}
if (bytes_marked_ahead_of_schedule_ >= bytes_to_process) {
// Steps performed in tasks and concurrently have put us ahead of
// schedule. We skip processing of marking dequeue here and thus shift
// marking time from inside V8 to standalone tasks.
bytes_marked_ahead_of_schedule_ -= bytes_to_process;
bytes_processed += bytes_to_process;
bytes_to_process = IncrementalMarking::kMinStepSizeInBytes;
}
bytes_processed += Step(bytes_to_process, GC_VIA_STACK_GUARD,
StepOrigin::kV8, WorklistToProcess::kAll);
bytes_allocated_ -= Min(bytes_allocated_, bytes_processed);
}
}
size_t IncrementalMarking::Step(size_t bytes_to_process,
CompletionAction action, StepOrigin step_origin,
WorklistToProcess worklist_to_process) {
double start = heap_->MonotonicallyIncreasingTimeInMs();
if (state_ == SWEEPING) {
TRACE_GC(heap_->tracer(), GCTracer::Scope::MC_INCREMENTAL_SWEEPING);
FinalizeSweeping();
}
size_t bytes_processed = 0;
if (state_ == MARKING) {
if (FLAG_concurrent_marking) {
heap_->new_space()->ResetOriginalTop();
// It is safe to merge back all objects that were on hold to the shared
// work list at Step because we are at a safepoint where all objects
// are properly initialized.
marking_worklist()->shared()->MergeGlobalPool(
marking_worklist()->on_hold());
}
// Only print marking worklist in debug mode to save ~40KB of code size.
#ifdef DEBUG
if (FLAG_trace_incremental_marking && FLAG_trace_concurrent_marking &&
FLAG_trace_gc_verbose) {
marking_worklist()->Print();
}
#endif
if (worklist_to_process == WorklistToProcess::kBailout) {
bytes_processed =
ProcessMarkingWorklist<WorklistToProcess::kBailout>(bytes_to_process);
} else {
bytes_processed =
ProcessMarkingWorklist<WorklistToProcess::kAll>(bytes_to_process);
}
if (step_origin == StepOrigin::kTask) {
bytes_marked_ahead_of_schedule_ += bytes_processed;
}
if (marking_worklist()->IsEmpty()) {
if (heap_->local_embedder_heap_tracer()
->ShouldFinalizeIncrementalMarking()) {
if (!finalize_marking_completed_) {
FinalizeMarking(action);
} else {
MarkingComplete(action);
}
} else {
heap_->local_embedder_heap_tracer()->NotifyV8MarkingWorklistWasEmpty();
}
}
}
if (FLAG_concurrent_marking) {
heap_->concurrent_marking()->RescheduleTasksIfNeeded();
}
double end = heap_->MonotonicallyIncreasingTimeInMs();
double duration = (end - start);
// Note that we report zero bytes here when sweeping was in progress or
// when we just started incremental marking. In these cases we did not
// process the marking deque.
heap_->tracer()->AddIncrementalMarkingStep(duration, bytes_processed);
if (FLAG_trace_incremental_marking) {
heap_->isolate()->PrintWithTimestamp(
"[IncrementalMarking] Step %s %" PRIuS "KB (%" PRIuS "KB) in %.1f\n",
step_origin == StepOrigin::kV8 ? "in v8" : "in task",
bytes_processed / KB, bytes_to_process / KB, duration);
}
if (FLAG_trace_concurrent_marking) {
heap_->isolate()->PrintWithTimestamp(
"Concurrently marked %" PRIuS "KB\n",
heap_->concurrent_marking()->TotalMarkedBytes() / KB);
}
return bytes_processed;
}
} // namespace internal
} // namespace v8