summaryrefslogtreecommitdiffstats
path: root/js/src/gc/Allocator.cpp
diff options
context:
space:
mode:
authorDaniel Baumann <daniel.baumann@progress-linux.org>2024-04-07 19:33:14 +0000
committerDaniel Baumann <daniel.baumann@progress-linux.org>2024-04-07 19:33:14 +0000
commit36d22d82aa202bb199967e9512281e9a53db42c9 (patch)
tree105e8c98ddea1c1e4784a60a5a6410fa416be2de /js/src/gc/Allocator.cpp
parentInitial commit. (diff)
downloadfirefox-esr-36d22d82aa202bb199967e9512281e9a53db42c9.tar.xz
firefox-esr-36d22d82aa202bb199967e9512281e9a53db42c9.zip
Adding upstream version 115.7.0esr.upstream/115.7.0esr
Signed-off-by: Daniel Baumann <daniel.baumann@progress-linux.org>
Diffstat (limited to 'js/src/gc/Allocator.cpp')
-rw-r--r--js/src/gc/Allocator.cpp672
1 files changed, 672 insertions, 0 deletions
diff --git a/js/src/gc/Allocator.cpp b/js/src/gc/Allocator.cpp
new file mode 100644
index 0000000000..92d4653961
--- /dev/null
+++ b/js/src/gc/Allocator.cpp
@@ -0,0 +1,672 @@
+/* -*- Mode: C++; tab-width: 8; indent-tabs-mode: nil; c-basic-offset: 2 -*-
+ * vim: set ts=8 sts=2 et sw=2 tw=80:
+ * This Source Code Form is subject to the terms of the Mozilla Public
+ * License, v. 2.0. If a copy of the MPL was not distributed with this
+ * file, You can obtain one at http://mozilla.org/MPL/2.0/. */
+
+#include "gc/Allocator.h"
+
+#include "mozilla/DebugOnly.h"
+#include "mozilla/OperatorNewExtensions.h"
+#include "mozilla/TimeStamp.h"
+
+#include "gc/GCInternals.h"
+#include "gc/GCLock.h"
+#include "gc/GCProbes.h"
+#include "gc/Nursery.h"
+#include "threading/CpuCount.h"
+#include "util/Poison.h"
+#include "vm/BigIntType.h"
+#include "vm/JSContext.h"
+#include "vm/Runtime.h"
+#include "vm/StringType.h"
+
+#include "gc/ArenaList-inl.h"
+#include "gc/Heap-inl.h"
+#include "gc/PrivateIterators-inl.h"
+#include "vm/JSContext-inl.h"
+
+using mozilla::TimeDuration;
+using mozilla::TimeStamp;
+
+using namespace js;
+using namespace js::gc;
+
+// Return a Heap value that can be compared numerically with an
+// allocation's requested heap to determine whether to allocate in the nursery
+// or the tenured heap.
+//
+// If nursery allocation is allowed this returns Heap::Tenured, meaning only
+// Heap::Tenured allocations will be tenured. If nursery allocation is not
+// allowed this returns Heap::Default, meaning all allocations are tenured.
+static Heap MinHeapToTenure(bool allowNurseryAlloc) {
+ static_assert(Heap::Tenured > Heap::Default);
+ return allowNurseryAlloc ? Heap::Tenured : Heap::Default;
+}
+
+void Zone::setNurseryAllocFlags(bool allocObjects, bool allocStrings,
+ bool allocBigInts) {
+ allocNurseryObjects_ = allocObjects;
+ allocNurseryStrings_ = allocStrings;
+ allocNurseryBigInts_ = allocBigInts;
+
+ minObjectHeapToTenure_ = MinHeapToTenure(allocNurseryObjects());
+ minStringHeapToTenure_ = MinHeapToTenure(allocNurseryStrings());
+ minBigintHeapToTenure_ = MinHeapToTenure(allocNurseryBigInts());
+}
+
+template <JS::TraceKind traceKind, AllowGC allowGC /* = CanGC */>
+void* gc::CellAllocator::AllocNurseryOrTenuredCell(JSContext* cx,
+ AllocKind allocKind,
+ gc::Heap heap,
+ AllocSite* site) {
+ MOZ_ASSERT(!cx->isHelperThreadContext());
+ MOZ_ASSERT_IF(heap != gc::Heap::Tenured, IsNurseryAllocable(allocKind));
+ MOZ_ASSERT(MapAllocToTraceKind(allocKind) == traceKind);
+ MOZ_ASSERT_IF(site && site->initialHeap() == Heap::Tenured,
+ heap == Heap::Tenured);
+
+ size_t thingSize = Arena::thingSize(allocKind);
+
+ JSRuntime* rt = cx->runtime();
+ if (!rt->gc.checkAllocatorState<allowGC>(cx, allocKind)) {
+ return nullptr;
+ }
+
+ if (heap < cx->zone()->minHeapToTenure(traceKind)) {
+ if (!site) {
+ site = cx->zone()->unknownAllocSite(traceKind);
+ }
+
+ void* obj =
+ rt->gc.tryNewNurseryCell<traceKind, allowGC>(cx, thingSize, site);
+ if (obj) {
+ return obj;
+ }
+
+ // Our most common non-jit allocation path is NoGC; thus, if we fail the
+ // alloc and cannot GC, we *must* return nullptr here so that the caller
+ // will do a CanGC allocation to clear the nursery. Failing to do so will
+ // cause all allocations on this path to land in Tenured, and we will not
+ // get the benefit of the nursery.
+ if (!allowGC) {
+ return nullptr;
+ }
+ }
+
+ return GCRuntime::tryNewTenuredThing<allowGC>(cx, allocKind, thingSize);
+}
+
+#define INSTANTIATE_ALLOC_NURSERY_CELL(traceKind, allowGc) \
+ template void* \
+ gc::CellAllocator::AllocNurseryOrTenuredCell<traceKind, allowGc>( \
+ JSContext*, AllocKind, gc::Heap, AllocSite*);
+INSTANTIATE_ALLOC_NURSERY_CELL(JS::TraceKind::Object, NoGC)
+INSTANTIATE_ALLOC_NURSERY_CELL(JS::TraceKind::Object, CanGC)
+INSTANTIATE_ALLOC_NURSERY_CELL(JS::TraceKind::String, NoGC)
+INSTANTIATE_ALLOC_NURSERY_CELL(JS::TraceKind::String, CanGC)
+INSTANTIATE_ALLOC_NURSERY_CELL(JS::TraceKind::BigInt, NoGC)
+INSTANTIATE_ALLOC_NURSERY_CELL(JS::TraceKind::BigInt, CanGC)
+#undef INSTANTIATE_ALLOC_NURSERY_CELL
+
+// Attempt to allocate a new cell in the nursery. If there is not enough room in
+// the nursery or there is an OOM, this method will return nullptr.
+template <JS::TraceKind kind, AllowGC allowGC>
+void* GCRuntime::tryNewNurseryCell(JSContext* cx, size_t thingSize,
+ AllocSite* site) {
+ MOZ_ASSERT(cx->isNurseryAllocAllowed());
+ MOZ_ASSERT(cx->zone() == site->zone());
+ MOZ_ASSERT(!cx->zone()->isAtomsZone());
+ MOZ_ASSERT(cx->zone()->allocKindInNursery(kind));
+
+ void* ptr = cx->nursery().allocateCell(site, thingSize, kind);
+ if (ptr) {
+ return ptr;
+ }
+
+ if constexpr (allowGC) {
+ if (!cx->suppressGC) {
+ cx->runtime()->gc.minorGC(JS::GCReason::OUT_OF_NURSERY);
+
+ // Exceeding gcMaxBytes while tenuring can disable the Nursery.
+ if (cx->zone()->allocKindInNursery(kind)) {
+ return cx->nursery().allocateCell(site, thingSize, kind);
+ }
+ }
+ }
+
+ return nullptr;
+}
+
+template <AllowGC allowGC /* = CanGC */>
+void* gc::CellAllocator::AllocTenuredCell(JSContext* cx, gc::AllocKind kind,
+ size_t size) {
+ MOZ_ASSERT(!cx->isHelperThreadContext());
+ MOZ_ASSERT(!IsNurseryAllocable(kind));
+ MOZ_ASSERT(size == Arena::thingSize(kind));
+ MOZ_ASSERT(
+ size >= gc::MinCellSize,
+ "All allocations must be at least the allocator-imposed minimum size.");
+
+ if (!cx->runtime()->gc.checkAllocatorState<allowGC>(cx, kind)) {
+ return nullptr;
+ }
+
+ return GCRuntime::tryNewTenuredThing<allowGC>(cx, kind, size);
+}
+template void* gc::CellAllocator::AllocTenuredCell<NoGC>(JSContext*, AllocKind,
+ size_t);
+template void* gc::CellAllocator::AllocTenuredCell<CanGC>(JSContext*, AllocKind,
+ size_t);
+
+template <AllowGC allowGC>
+/* static */
+void* GCRuntime::tryNewTenuredThing(JSContext* cx, AllocKind kind,
+ size_t thingSize) {
+ // Bump allocate in the arena's current free-list span.
+ Zone* zone = cx->zone();
+ void* ptr = zone->arenas.freeLists().allocate(kind);
+ if (MOZ_UNLIKELY(!ptr)) {
+ // Get the next available free list and allocate out of it. This may
+ // acquire a new arena, which will lock the chunk list. If there are no
+ // chunks available it may also allocate new memory directly.
+ ptr = refillFreeList(cx, kind);
+
+ if (MOZ_UNLIKELY(!ptr)) {
+ if constexpr (allowGC) {
+ cx->runtime()->gc.attemptLastDitchGC(cx);
+ ptr = tryNewTenuredThing<NoGC>(cx, kind, thingSize);
+ if (ptr) {
+ return ptr;
+ }
+ ReportOutOfMemory(cx);
+ }
+
+ return nullptr;
+ }
+ }
+
+#ifdef DEBUG
+ checkIncrementalZoneState(cx, ptr);
+#endif
+
+ gcprobes::TenuredAlloc(ptr, kind);
+
+ // We count this regardless of the profiler's state, assuming that it costs
+ // just as much to count it, as to check the profiler's state and decide not
+ // to count it.
+ zone->noteTenuredAlloc();
+
+ return ptr;
+}
+
+void GCRuntime::attemptLastDitchGC(JSContext* cx) {
+ // Either there was no memory available for a new chunk or the heap hit its
+ // size limit. Try to perform an all-compartments, non-incremental, shrinking
+ // GC and wait for it to finish.
+
+ if (!lastLastDitchTime.IsNull() &&
+ TimeStamp::Now() - lastLastDitchTime <= tunables.minLastDitchGCPeriod()) {
+ return;
+ }
+
+ JS::PrepareForFullGC(cx);
+ gc(JS::GCOptions::Shrink, JS::GCReason::LAST_DITCH);
+ waitBackgroundAllocEnd();
+ waitBackgroundFreeEnd();
+
+ lastLastDitchTime = mozilla::TimeStamp::Now();
+}
+
+template <AllowGC allowGC>
+bool GCRuntime::checkAllocatorState(JSContext* cx, AllocKind kind) {
+ MOZ_ASSERT_IF(cx->zone()->isAtomsZone(),
+ kind == AllocKind::ATOM || kind == AllocKind::FAT_INLINE_ATOM ||
+ kind == AllocKind::SYMBOL || kind == AllocKind::JITCODE ||
+ kind == AllocKind::SCOPE);
+ MOZ_ASSERT_IF(!cx->zone()->isAtomsZone(),
+ kind != AllocKind::ATOM && kind != AllocKind::FAT_INLINE_ATOM);
+ MOZ_ASSERT(!JS::RuntimeHeapIsBusy());
+
+ if constexpr (allowGC) {
+ // Crash if we could perform a GC action when it is not safe.
+ if (!cx->suppressGC) {
+ cx->verifyIsSafeToGC();
+ }
+
+ gcIfNeededAtAllocation(cx);
+ }
+
+ // For testing out of memory conditions.
+ if (js::oom::ShouldFailWithOOM()) {
+ // If we are doing a fallible allocation, percolate up the OOM instead of
+ // reporting it.
+ if constexpr (allowGC) {
+ ReportOutOfMemory(cx);
+ }
+ return false;
+ }
+
+ return true;
+}
+
+inline void GCRuntime::gcIfNeededAtAllocation(JSContext* cx) {
+#ifdef JS_GC_ZEAL
+ if (needZealousGC()) {
+ runDebugGC();
+ }
+#endif
+
+ // Invoking the interrupt callback can fail and we can't usefully
+ // handle that here. Just check in case we need to collect instead.
+ if (cx->hasAnyPendingInterrupt()) {
+ gcIfRequested();
+ }
+}
+
+#ifdef DEBUG
+void GCRuntime::checkIncrementalZoneState(JSContext* cx, void* ptr) {
+ MOZ_ASSERT(ptr);
+ TenuredCell* cell = reinterpret_cast<TenuredCell*>(ptr);
+ TenuredChunkBase* chunk = detail::GetCellChunkBase(cell);
+ if (cx->zone()->isGCMarkingOrSweeping()) {
+ MOZ_ASSERT(chunk->markBits.isMarkedBlack(cell));
+ } else {
+ MOZ_ASSERT(!chunk->markBits.isMarkedAny(cell));
+ }
+}
+#endif
+
+void* js::gc::AllocateCellInGC(Zone* zone, AllocKind thingKind) {
+ void* ptr = zone->arenas.allocateFromFreeList(thingKind);
+ if (!ptr) {
+ AutoEnterOOMUnsafeRegion oomUnsafe;
+ ptr = GCRuntime::refillFreeListInGC(zone, thingKind);
+ if (!ptr) {
+ oomUnsafe.crash(ChunkSize, "Failed to allocate new chunk during GC");
+ }
+ }
+ return ptr;
+}
+
+// /////////// Arena -> Thing Allocator //////////////////////////////////////
+
+void GCRuntime::startBackgroundAllocTaskIfIdle() {
+ AutoLockHelperThreadState lock;
+ if (!allocTask.wasStarted(lock)) {
+ // Join the previous invocation of the task. This will return immediately
+ // if the thread has never been started.
+ allocTask.joinWithLockHeld(lock);
+ allocTask.startWithLockHeld(lock);
+ }
+}
+
+/* static */
+void* GCRuntime::refillFreeList(JSContext* cx, AllocKind thingKind) {
+ MOZ_ASSERT(cx->zone()->arenas.freeLists().isEmpty(thingKind));
+ MOZ_ASSERT(!cx->isHelperThreadContext());
+
+ // It should not be possible to allocate on the main thread while we are
+ // inside a GC.
+ MOZ_ASSERT(!JS::RuntimeHeapIsBusy(), "allocating while under GC");
+
+ return cx->zone()->arenas.refillFreeListAndAllocate(
+ thingKind, ShouldCheckThresholds::CheckThresholds);
+}
+
+/* static */
+void* GCRuntime::refillFreeListInGC(Zone* zone, AllocKind thingKind) {
+ // Called by compacting GC to refill a free list while we are in a GC.
+ MOZ_ASSERT(JS::RuntimeHeapIsCollecting());
+ MOZ_ASSERT_IF(!JS::RuntimeHeapIsMinorCollecting(),
+ !zone->runtimeFromMainThread()->gc.isBackgroundSweeping());
+
+ return zone->arenas.refillFreeListAndAllocate(
+ thingKind, ShouldCheckThresholds::DontCheckThresholds);
+}
+
+void* ArenaLists::refillFreeListAndAllocate(
+ AllocKind thingKind, ShouldCheckThresholds checkThresholds) {
+ MOZ_ASSERT(freeLists().isEmpty(thingKind));
+
+ JSRuntime* rt = runtimeFromAnyThread();
+
+ mozilla::Maybe<AutoLockGCBgAlloc> maybeLock;
+
+ // See if we can proceed without taking the GC lock.
+ if (concurrentUse(thingKind) != ConcurrentUse::None) {
+ maybeLock.emplace(rt);
+ }
+
+ Arena* arena = arenaList(thingKind).takeNextArena();
+ if (arena) {
+ // Empty arenas should be immediately freed.
+ MOZ_ASSERT(!arena->isEmpty());
+
+ return freeLists().setArenaAndAllocate(arena, thingKind);
+ }
+
+ // Parallel threads have their own ArenaLists, but chunks are shared;
+ // if we haven't already, take the GC lock now to avoid racing.
+ if (maybeLock.isNothing()) {
+ maybeLock.emplace(rt);
+ }
+
+ TenuredChunk* chunk = rt->gc.pickChunk(maybeLock.ref());
+ if (!chunk) {
+ return nullptr;
+ }
+
+ // Although our chunk should definitely have enough space for another arena,
+ // there are other valid reasons why TenuredChunk::allocateArena() may fail.
+ arena = rt->gc.allocateArena(chunk, zone_, thingKind, checkThresholds,
+ maybeLock.ref());
+ if (!arena) {
+ return nullptr;
+ }
+
+ ArenaList& al = arenaList(thingKind);
+ MOZ_ASSERT(al.isCursorAtEnd());
+ al.insertBeforeCursor(arena);
+
+ return freeLists().setArenaAndAllocate(arena, thingKind);
+}
+
+inline void* FreeLists::setArenaAndAllocate(Arena* arena, AllocKind kind) {
+#ifdef DEBUG
+ auto old = freeLists_[kind];
+ if (!old->isEmpty()) {
+ old->getArena()->checkNoMarkedFreeCells();
+ }
+#endif
+
+ FreeSpan* span = arena->getFirstFreeSpan();
+ freeLists_[kind] = span;
+
+ Zone* zone = arena->zone;
+ if (MOZ_UNLIKELY(zone->isGCMarkingOrSweeping())) {
+ arena->arenaAllocatedDuringGC();
+ }
+
+ TenuredCell* thing = span->allocate(Arena::thingSize(kind));
+ MOZ_ASSERT(thing); // This allocation is infallible.
+
+ return thing;
+}
+
+void Arena::arenaAllocatedDuringGC() {
+ // Ensure that anything allocated during the mark or sweep phases of an
+ // incremental GC will be marked black by pre-marking all free cells in the
+ // arena we are about to allocate from.
+
+ MOZ_ASSERT(zone->isGCMarkingOrSweeping());
+ for (ArenaFreeCellIter cell(this); !cell.done(); cell.next()) {
+ MOZ_ASSERT(!cell->isMarkedAny());
+ cell->markBlack();
+ }
+}
+
+// /////////// TenuredChunk -> Arena Allocator ///////////////////////////////
+
+bool GCRuntime::wantBackgroundAllocation(const AutoLockGC& lock) const {
+ // To minimize memory waste, we do not want to run the background chunk
+ // allocation if we already have some empty chunks or when the runtime has
+ // a small heap size (and therefore likely has a small growth rate).
+ return allocTask.enabled() &&
+ emptyChunks(lock).count() < minEmptyChunkCount(lock) &&
+ (fullChunks(lock).count() + availableChunks(lock).count()) >= 4;
+}
+
+Arena* GCRuntime::allocateArena(TenuredChunk* chunk, Zone* zone,
+ AllocKind thingKind,
+ ShouldCheckThresholds checkThresholds,
+ const AutoLockGC& lock) {
+ MOZ_ASSERT(chunk->hasAvailableArenas());
+
+ // Fail the allocation if we are over our heap size limits.
+ if ((checkThresholds != ShouldCheckThresholds::DontCheckThresholds) &&
+ (heapSize.bytes() >= tunables.gcMaxBytes())) {
+ return nullptr;
+ }
+
+ Arena* arena = chunk->allocateArena(this, zone, thingKind, lock);
+ zone->gcHeapSize.addGCArena(heapSize);
+
+ // Trigger an incremental slice if needed.
+ if (checkThresholds != ShouldCheckThresholds::DontCheckThresholds) {
+ maybeTriggerGCAfterAlloc(zone);
+ }
+
+ return arena;
+}
+
+Arena* TenuredChunk::allocateArena(GCRuntime* gc, Zone* zone,
+ AllocKind thingKind,
+ const AutoLockGC& lock) {
+ if (info.numArenasFreeCommitted == 0) {
+ commitOnePage(gc);
+ MOZ_ASSERT(info.numArenasFreeCommitted == ArenasPerPage);
+ }
+
+ MOZ_ASSERT(info.numArenasFreeCommitted > 0);
+ Arena* arena = fetchNextFreeArena(gc);
+
+ arena->init(zone, thingKind, lock);
+ updateChunkListAfterAlloc(gc, lock);
+
+ verify();
+
+ return arena;
+}
+
+template <size_t N>
+static inline size_t FindFirstBitSet(
+ const mozilla::BitSet<N, uint32_t>& bitset) {
+ MOZ_ASSERT(!bitset.IsEmpty());
+
+ const auto& words = bitset.Storage();
+ for (size_t i = 0; i < words.Length(); i++) {
+ uint32_t word = words[i];
+ if (word) {
+ return i * 32 + mozilla::CountTrailingZeroes32(word);
+ }
+ }
+
+ MOZ_CRASH("No bits found");
+}
+
+void TenuredChunk::commitOnePage(GCRuntime* gc) {
+ MOZ_ASSERT(info.numArenasFreeCommitted == 0);
+ MOZ_ASSERT(info.numArenasFree >= ArenasPerPage);
+
+ uint32_t pageIndex = FindFirstBitSet(decommittedPages);
+ MOZ_ASSERT(decommittedPages[pageIndex]);
+
+ if (DecommitEnabled()) {
+ MarkPagesInUseSoft(pageAddress(pageIndex), PageSize);
+ }
+
+ decommittedPages[pageIndex] = false;
+
+ for (size_t i = 0; i < ArenasPerPage; i++) {
+ size_t arenaIndex = pageIndex * ArenasPerPage + i;
+ MOZ_ASSERT(!freeCommittedArenas[arenaIndex]);
+ freeCommittedArenas[arenaIndex] = true;
+ arenas[arenaIndex].setAsNotAllocated();
+ ++info.numArenasFreeCommitted;
+ gc->updateOnArenaFree();
+ }
+
+ verify();
+}
+
+inline void GCRuntime::updateOnFreeArenaAlloc(const TenuredChunkInfo& info) {
+ MOZ_ASSERT(info.numArenasFreeCommitted <= numArenasFreeCommitted);
+ --numArenasFreeCommitted;
+}
+
+Arena* TenuredChunk::fetchNextFreeArena(GCRuntime* gc) {
+ MOZ_ASSERT(info.numArenasFreeCommitted > 0);
+ MOZ_ASSERT(info.numArenasFreeCommitted <= info.numArenasFree);
+
+ size_t index = FindFirstBitSet(freeCommittedArenas);
+ MOZ_ASSERT(freeCommittedArenas[index]);
+
+ freeCommittedArenas[index] = false;
+ --info.numArenasFreeCommitted;
+ --info.numArenasFree;
+ gc->updateOnFreeArenaAlloc(info);
+
+ return &arenas[index];
+}
+
+// /////////// System -> TenuredChunk Allocator //////////////////////////////
+
+TenuredChunk* GCRuntime::getOrAllocChunk(AutoLockGCBgAlloc& lock) {
+ TenuredChunk* chunk = emptyChunks(lock).pop();
+ if (chunk) {
+ // Reinitialize ChunkBase; arenas are all free and may or may not be
+ // committed.
+ SetMemCheckKind(chunk, sizeof(ChunkBase), MemCheckKind::MakeUndefined);
+ chunk->initBase(rt, nullptr);
+ MOZ_ASSERT(chunk->unused());
+ } else {
+ void* ptr = TenuredChunk::allocate(this);
+ if (!ptr) {
+ return nullptr;
+ }
+
+ chunk = TenuredChunk::emplace(ptr, this, /* allMemoryCommitted = */ true);
+ MOZ_ASSERT(chunk->info.numArenasFreeCommitted == 0);
+ }
+
+ if (wantBackgroundAllocation(lock)) {
+ lock.tryToStartBackgroundAllocation();
+ }
+
+ return chunk;
+}
+
+void GCRuntime::recycleChunk(TenuredChunk* chunk, const AutoLockGC& lock) {
+#ifdef DEBUG
+ MOZ_ASSERT(chunk->unused());
+ chunk->verify();
+#endif
+
+ // Poison ChunkBase to catch use after free.
+ AlwaysPoison(chunk, JS_FREED_CHUNK_PATTERN, sizeof(ChunkBase),
+ MemCheckKind::MakeNoAccess);
+
+ emptyChunks(lock).push(chunk);
+}
+
+TenuredChunk* GCRuntime::pickChunk(AutoLockGCBgAlloc& lock) {
+ if (availableChunks(lock).count()) {
+ return availableChunks(lock).head();
+ }
+
+ TenuredChunk* chunk = getOrAllocChunk(lock);
+ if (!chunk) {
+ return nullptr;
+ }
+
+#ifdef DEBUG
+ chunk->verify();
+ MOZ_ASSERT(chunk->unused());
+ MOZ_ASSERT(!fullChunks(lock).contains(chunk));
+ MOZ_ASSERT(!availableChunks(lock).contains(chunk));
+#endif
+
+ availableChunks(lock).push(chunk);
+
+ return chunk;
+}
+
+BackgroundAllocTask::BackgroundAllocTask(GCRuntime* gc, ChunkPool& pool)
+ : GCParallelTask(gc, gcstats::PhaseKind::NONE),
+ chunkPool_(pool),
+ enabled_(CanUseExtraThreads() && GetCPUCount() >= 2) {
+ // This can occur outside GCs so doesn't have a stats phase.
+}
+
+void BackgroundAllocTask::run(AutoLockHelperThreadState& lock) {
+ AutoUnlockHelperThreadState unlock(lock);
+
+ AutoLockGC gcLock(gc);
+ while (!isCancelled() && gc->wantBackgroundAllocation(gcLock)) {
+ TenuredChunk* chunk;
+ {
+ AutoUnlockGC unlock(gcLock);
+ void* ptr = TenuredChunk::allocate(gc);
+ if (!ptr) {
+ break;
+ }
+ chunk = TenuredChunk::emplace(ptr, gc, /* allMemoryCommitted = */ true);
+ }
+ chunkPool_.ref().push(chunk);
+ }
+}
+
+/* static */
+void* TenuredChunk::allocate(GCRuntime* gc) {
+ void* chunk = MapAlignedPages(ChunkSize, ChunkSize);
+ if (!chunk) {
+ return nullptr;
+ }
+
+ gc->stats().count(gcstats::COUNT_NEW_CHUNK);
+ return chunk;
+}
+
+static inline bool ShouldDecommitNewChunk(bool allMemoryCommitted,
+ const GCSchedulingState& state) {
+ if (!DecommitEnabled()) {
+ return false;
+ }
+
+ return !allMemoryCommitted || !state.inHighFrequencyGCMode();
+}
+
+TenuredChunk* TenuredChunk::emplace(void* ptr, GCRuntime* gc,
+ bool allMemoryCommitted) {
+ /* The chunk may still have some regions marked as no-access. */
+ MOZ_MAKE_MEM_UNDEFINED(ptr, ChunkSize);
+
+ /*
+ * Poison the chunk. Note that decommitAllArenas() below will mark the
+ * arenas as inaccessible (for memory sanitizers).
+ */
+ Poison(ptr, JS_FRESH_TENURED_PATTERN, ChunkSize, MemCheckKind::MakeUndefined);
+
+ TenuredChunk* chunk = new (mozilla::KnownNotNull, ptr) TenuredChunk(gc->rt);
+
+ if (ShouldDecommitNewChunk(allMemoryCommitted, gc->schedulingState)) {
+ // Decommit the arenas. We do this after poisoning so that if the OS does
+ // not have to recycle the pages, we still get the benefit of poisoning.
+ chunk->decommitAllArenas();
+ } else {
+ // The chunk metadata is initialized as decommitted regardless, to avoid
+ // having to initialize the arenas at this time.
+ chunk->initAsDecommitted();
+ }
+
+ chunk->verify();
+
+ return chunk;
+}
+
+void TenuredChunk::decommitAllArenas() {
+ MOZ_ASSERT(unused());
+ MarkPagesUnusedSoft(&arenas[0], ArenasPerChunk * ArenaSize);
+ initAsDecommitted();
+}
+
+void TenuredChunkBase::initAsDecommitted() {
+ // Set the state of all arenas to free and decommitted. They might not
+ // actually be decommitted, but in that case the re-commit operation is a
+ // no-op so it doesn't matter.
+ decommittedPages.SetAll();
+ freeCommittedArenas.ResetAll();
+ info.numArenasFree = ArenasPerChunk;
+ info.numArenasFreeCommitted = 0;
+}