immixspace.rs

use super::defrag::StatsForDefrag;
use super::line::*;
use super::{block::*, defrag::Defrag};
use crate::plan::VectorObjectQueue;
use crate::policy::gc_work::{TraceKind, TRACE_KIND_TRANSITIVE_PIN};
use crate::policy::sft::GCWorkerMutRef;
use crate::policy::sft::SFT;
use crate::policy::sft_map::SFTMap;
use crate::policy::space::{CommonSpace, Space};
use crate::util::alloc::allocator::AllocatorContext;
use crate::util::constants::LOG_BYTES_IN_PAGE;
use crate::util::heap::chunk_map::*;
use crate::util::heap::BlockPageResource;
use crate::util::heap::PageResource;
use crate::util::linear_scan::{Region, RegionIterator};
use crate::util::metadata::side_metadata::SideMetadataSpec;
#[cfg(feature = "vo_bit")]
use crate::util::metadata::vo_bit;
use crate::util::metadata::{self, MetadataSpec};
use crate::util::object_enum::ObjectEnumerator;
use crate::util::object_forwarding;
use crate::util::{copy::*, epilogue, object_enum};
use crate::util::{Address, ObjectReference};
use crate::vm::*;
use crate::{
    plan::ObjectQueue,
    scheduler::{GCWork, GCWorkScheduler, GCWorker, WorkBucketStage},
    util::opaque_pointer::{VMThread, VMWorkerThread},
    MMTK,
};
use atomic::Ordering;
use std::sync::{atomic::AtomicU8, atomic::AtomicUsize, Arc};

pub(crate) const TRACE_KIND_FAST: TraceKind = 0;
pub(crate) const TRACE_KIND_DEFRAG: TraceKind = 1;

pub struct ImmixSpace<VM: VMBinding> {
    common: CommonSpace<VM>,
    pr: BlockPageResource<VM, Block>,
    /// Allocation status for all chunks in immix space
    pub chunk_map: ChunkMap,
    /// Current line mark state
    pub line_mark_state: AtomicU8,
    /// Line mark state in previous GC
    line_unavail_state: AtomicU8,
    /// A list of all reusable blocks
    pub reusable_blocks: ReusableBlockPool,
    /// Defrag utilities
    pub(super) defrag: Defrag,
    /// How many lines have been consumed since last GC?
    lines_consumed: AtomicUsize,
    /// Object mark state
    mark_state: u8,
    /// Work packet scheduler
    scheduler: Arc<GCWorkScheduler<VM>>,
    /// Some settings for this space
    space_args: ImmixSpaceArgs,
}

/// Some arguments for Immix Space.
pub struct ImmixSpaceArgs {
    /// Mark an object as unlogged when we trace an object.
    /// Normally we set the log bit when we copy an object with [`crate::util::copy::CopySemantics::PromoteToMature`].
    /// In sticky immix, we 'promote' an object to mature when we trace the object
    /// (no matter we copy an object or not). So we have to use `PromoteToMature`, and instead
    /// just set the log bit in the space when an object is traced.
    pub unlog_object_when_traced: bool,
    /// Reset log bit at the start of a major GC.
    /// Normally we do not need to do this. When immix is used as the mature space,
    /// any object should be set as unlogged, and that bit does not need to be cleared
    /// even if the object is dead. But in sticky Immix, the mature object and
    /// the nursery object are in the same space, we will have to use the
    /// bit to differentiate them. So we reset all the log bits in major GCs,
    /// and unlogged the objects when they are traced (alive).
    pub reset_log_bit_in_major_gc: bool,
    /// Whether this ImmixSpace instance contains both young and old objects.
    /// This affects the updating of valid-object bits.  If some lines or blocks of this ImmixSpace
    /// instance contain young objects, their VO bits need to be updated during this GC.  Currently
    /// only StickyImmix is affected.  GenImmix allocates young objects in a separete CopySpace
    /// nursery and its VO bits can be cleared in bulk.
    // Currently only used when "vo_bit" is enabled.  Using #[cfg(...)] to eliminate dead code warning.
    #[cfg(feature = "vo_bit")]
    pub mixed_age: bool,
}

unsafe impl<VM: VMBinding> Sync for ImmixSpace<VM> {}

impl<VM: VMBinding> SFT for ImmixSpace<VM> {
    fn name(&self) -> &str {
        self.get_name()
    }

    fn get_forwarded_object(&self, object: ObjectReference) -> Option<ObjectReference> {
        // If we never move objects, look no further.
        if super::NEVER_MOVE_OBJECTS {
            return None;
        }

        if object_forwarding::is_forwarded::<VM>(object) {
            Some(object_forwarding::read_forwarding_pointer::<VM>(object))
        } else {
            None
        }
    }

    fn is_live(&self, object: ObjectReference) -> bool {
        // If the mark bit is set, it is live.
        if self.is_marked(object) {
            return true;
        }

        // If we never move objects, look no further.
        if super::NEVER_MOVE_OBJECTS {
            return false;
        }

        // If the object is forwarded, it is live, too.
        object_forwarding::is_forwarded::<VM>(object)
    }
    #[cfg(feature = "object_pinning")]
    fn pin_object(&self, object: ObjectReference) -> bool {
        VM::VMObjectModel::LOCAL_PINNING_BIT_SPEC.pin_object::<VM>(object)
    }
    #[cfg(feature = "object_pinning")]
    fn unpin_object(&self, object: ObjectReference) -> bool {
        VM::VMObjectModel::LOCAL_PINNING_BIT_SPEC.unpin_object::<VM>(object)
    }
    #[cfg(feature = "object_pinning")]
    fn is_object_pinned(&self, object: ObjectReference) -> bool {
        VM::VMObjectModel::LOCAL_PINNING_BIT_SPEC.is_object_pinned::<VM>(object)
    }
    fn is_movable(&self) -> bool {
        !super::NEVER_MOVE_OBJECTS
    }

    #[cfg(feature = "sanity")]
    fn is_sane(&self) -> bool {
        true
    }
    fn initialize_object_metadata(&self, _object: ObjectReference, _alloc: bool) {
        #[cfg(feature = "vo_bit")]
        crate::util::metadata::vo_bit::set_vo_bit(_object);
    }
    #[cfg(feature = "is_mmtk_object")]
    fn is_mmtk_object(&self, addr: Address) -> Option<ObjectReference> {
        crate::util::metadata::vo_bit::is_vo_bit_set_for_addr(addr)
    }
    #[cfg(feature = "is_mmtk_object")]
    fn find_object_from_internal_pointer(
        &self,
        ptr: Address,
        max_search_bytes: usize,
    ) -> Option<ObjectReference> {
        // We don't need to search more than the max object size in the immix space.
        let search_bytes = usize::min(super::MAX_IMMIX_OBJECT_SIZE, max_search_bytes);
        crate::util::metadata::vo_bit::find_object_from_internal_pointer::<VM>(ptr, search_bytes)
    }
    fn sft_trace_object(
        &self,
        _queue: &mut VectorObjectQueue,
        _object: ObjectReference,
        _worker: GCWorkerMutRef,
    ) -> ObjectReference {
        panic!("We do not use SFT to trace objects for Immix. sft_trace_object() cannot be used.")
    }
}

impl<VM: VMBinding> Space<VM> for ImmixSpace<VM> {
    fn as_space(&self) -> &dyn Space<VM> {
        self
    }
    fn as_sft(&self) -> &(dyn SFT + Sync + 'static) {
        self
    }
    fn get_page_resource(&self) -> &dyn PageResource<VM> {
        &self.pr
    }
    fn maybe_get_page_resource_mut(&mut self) -> Option<&mut dyn PageResource<VM>> {
        Some(&mut self.pr)
    }
    fn common(&self) -> &CommonSpace<VM> {
        &self.common
    }
    fn initialize_sft(&self, sft_map: &mut dyn SFTMap) {
        self.common().initialize_sft(self.as_sft(), sft_map)
    }
    fn release_multiple_pages(&mut self, _start: Address) {
        panic!("immixspace only releases pages enmasse")
    }
    fn set_copy_for_sft_trace(&mut self, _semantics: Option<CopySemantics>) {
        panic!("We do not use SFT to trace objects for Immix. set_copy_context() cannot be used.")
    }

    fn enumerate_objects(&self, enumerator: &mut dyn ObjectEnumerator) {
        object_enum::enumerate_blocks_from_chunk_map::<Block>(enumerator, &self.chunk_map);
    }
}

impl<VM: VMBinding> crate::policy::gc_work::PolicyTraceObject<VM> for ImmixSpace<VM> {
    fn trace_object<Q: ObjectQueue, const KIND: TraceKind>(
        &self,
        queue: &mut Q,
        object: ObjectReference,
        copy: Option<CopySemantics>,
        worker: &mut GCWorker<VM>,
    ) -> ObjectReference {
        if KIND == TRACE_KIND_TRANSITIVE_PIN {
            self.trace_object_without_moving(queue, object)
        } else if KIND == TRACE_KIND_DEFRAG {
            if Block::containing(object).is_defrag_source() {
                debug_assert!(self.in_defrag());
                debug_assert!(
                    !crate::plan::is_nursery_gc(worker.mmtk.get_plan()),
                    "Calling PolicyTraceObject on Immix in nursery GC"
                );
                self.trace_object_with_opportunistic_copy(
                    queue,
                    object,
                    copy.unwrap(),
                    worker,
                    // This should not be nursery collection. Nursery collection does not use PolicyTraceObject.
                    false,
                )
            } else {
                self.trace_object_without_moving(queue, object)
            }
        } else if KIND == TRACE_KIND_FAST {
            self.trace_object_without_moving(queue, object)
        } else {
            unreachable!()
        }
    }

    fn post_scan_object(&self, object: ObjectReference) {
        if super::MARK_LINE_AT_SCAN_TIME && !super::BLOCK_ONLY {
            debug_assert!(self.in_space(object));
            self.mark_lines(object);
        }
    }

    fn may_move_objects<const KIND: TraceKind>() -> bool {
        if KIND == TRACE_KIND_DEFRAG {
            true
        } else if KIND == TRACE_KIND_FAST || KIND == TRACE_KIND_TRANSITIVE_PIN {
            false
        } else {
            unreachable!()
        }
    }
}

impl<VM: VMBinding> ImmixSpace<VM> {
    #[allow(unused)]
    const UNMARKED_STATE: u8 = 0;
    const MARKED_STATE: u8 = 1;

    /// Get side metadata specs
    fn side_metadata_specs() -> Vec<SideMetadataSpec> {
        metadata::extract_side_metadata(&if super::BLOCK_ONLY {
            vec![
                MetadataSpec::OnSide(Block::DEFRAG_STATE_TABLE),
                MetadataSpec::OnSide(Block::MARK_TABLE),
                MetadataSpec::OnSide(ChunkMap::ALLOC_TABLE),
                *VM::VMObjectModel::LOCAL_MARK_BIT_SPEC,
                *VM::VMObjectModel::LOCAL_FORWARDING_BITS_SPEC,
                *VM::VMObjectModel::LOCAL_FORWARDING_POINTER_SPEC,
                #[cfg(feature = "object_pinning")]
                *VM::VMObjectModel::LOCAL_PINNING_BIT_SPEC,
            ]
        } else {
            vec![
                MetadataSpec::OnSide(Line::MARK_TABLE),
                MetadataSpec::OnSide(Block::DEFRAG_STATE_TABLE),
                MetadataSpec::OnSide(Block::MARK_TABLE),
                MetadataSpec::OnSide(ChunkMap::ALLOC_TABLE),
                *VM::VMObjectModel::LOCAL_MARK_BIT_SPEC,
                *VM::VMObjectModel::LOCAL_FORWARDING_BITS_SPEC,
                *VM::VMObjectModel::LOCAL_FORWARDING_POINTER_SPEC,
                #[cfg(feature = "object_pinning")]
                *VM::VMObjectModel::LOCAL_PINNING_BIT_SPEC,
            ]
        })
    }

    pub fn new(
        args: crate::policy::space::PlanCreateSpaceArgs<VM>,
        space_args: ImmixSpaceArgs,
    ) -> Self {
        #[cfg(feature = "immix_non_moving")]
        info!(
            "Creating non-moving ImmixSpace: {}. Block size: 2^{}",
            args.name,
            Block::LOG_BYTES
        );

        if space_args.unlog_object_when_traced || space_args.reset_log_bit_in_major_gc {
            assert!(
                args.constraints.needs_log_bit,
                "Invalid args when the plan does not use log bit"
            );
        }

        super::validate_features();
        #[cfg(feature = "vo_bit")]
        vo_bit::helper::validate_config::<VM>();
        let vm_map = args.vm_map;
        let scheduler = args.scheduler.clone();
        let common =
            CommonSpace::new(args.into_policy_args(true, false, Self::side_metadata_specs()));
        ImmixSpace {
            pr: if common.vmrequest.is_discontiguous() {
                BlockPageResource::new_discontiguous(
                    Block::LOG_PAGES,
                    vm_map,
                    scheduler.num_workers(),
                )
            } else {
                BlockPageResource::new_contiguous(
                    Block::LOG_PAGES,
                    common.start,
                    common.extent,
                    vm_map,
                    scheduler.num_workers(),
                )
            },
            common,
            chunk_map: ChunkMap::new(),
            line_mark_state: AtomicU8::new(Line::RESET_MARK_STATE),
            line_unavail_state: AtomicU8::new(Line::RESET_MARK_STATE),
            lines_consumed: AtomicUsize::new(0),
            reusable_blocks: ReusableBlockPool::new(scheduler.num_workers()),
            defrag: Defrag::default(),
            // Set to the correct mark state when inititialized. We cannot rely on prepare to set it (prepare may get skipped in nursery GCs).
            mark_state: Self::MARKED_STATE,
            scheduler: scheduler.clone(),
            space_args,
        }
    }

    /// Flush the thread-local queues in BlockPageResource
    pub fn flush_page_resource(&self) {
        self.reusable_blocks.flush_all();
        #[cfg(target_pointer_width = "64")]
        self.pr.flush_all()
    }

    /// Get the number of defrag headroom pages.
    pub fn defrag_headroom_pages(&self) -> usize {
        self.defrag.defrag_headroom_pages(self)
    }

    /// Check if current GC is a defrag GC.
    pub fn in_defrag(&self) -> bool {
        self.defrag.in_defrag()
    }

    /// check if the current GC should do defragmentation.
    pub fn decide_whether_to_defrag(
        &self,
        emergency_collection: bool,
        collect_whole_heap: bool,
        collection_attempts: usize,
        user_triggered_collection: bool,
        full_heap_system_gc: bool,
    ) -> bool {
        self.defrag.decide_whether_to_defrag(
            emergency_collection,
            collect_whole_heap,
            collection_attempts,
            user_triggered_collection,
            self.reusable_blocks.len() == 0,
            full_heap_system_gc,
        );
        self.defrag.in_defrag()
    }

    /// Get work packet scheduler
    fn scheduler(&self) -> &GCWorkScheduler<VM> {
        &self.scheduler
    }

    pub fn prepare(&mut self, major_gc: bool, plan_stats: StatsForDefrag) {
        if major_gc {
            // Update mark_state
            if VM::VMObjectModel::LOCAL_MARK_BIT_SPEC.is_on_side() {
                self.mark_state = Self::MARKED_STATE;
            } else {
                // For header metadata, we use cyclic mark bits.
                unimplemented!("cyclic mark bits is not supported at the moment");
            }

            // Prepare defrag info
            if super::DEFRAG {
                self.defrag.prepare(self, plan_stats);
            }

            // Prepare each block for GC
            let threshold = self.defrag.defrag_spill_threshold.load(Ordering::Acquire);
            // # Safety: ImmixSpace reference is always valid within this collection cycle.
            let space = unsafe { &*(self as *const Self) };
            let work_packets = self.chunk_map.generate_tasks(|chunk| {
                Box::new(PrepareBlockState {
                    space,
                    chunk,
                    defrag_threshold: if space.in_defrag() {
                        Some(threshold)
                    } else {
                        None
                    },
                })
            });
            self.scheduler().work_buckets[WorkBucketStage::Prepare].bulk_add(work_packets);

            if !super::BLOCK_ONLY {
                self.line_mark_state.fetch_add(1, Ordering::AcqRel);
                if self.line_mark_state.load(Ordering::Acquire) > Line::MAX_MARK_STATE {
                    self.line_mark_state
                        .store(Line::RESET_MARK_STATE, Ordering::Release);
                }
            }
        }

        #[cfg(feature = "vo_bit")]
        if vo_bit::helper::need_to_clear_vo_bits_before_tracing::<VM>() {
            let maybe_scope = if major_gc {
                // If it is major GC, we always clear all VO bits because we are doing full-heap
                // tracing.
                Some(VOBitsClearingScope::FullGC)
            } else if self.space_args.mixed_age {
                // StickyImmix nursery GC.
                // Some lines (or blocks) contain only young objects,
                // while other lines (or blocks) contain only old objects.
                if super::BLOCK_ONLY {
                    // Block only.  Young objects are only allocated into fully empty blocks.
                    // Only clear unmarked blocks.
                    Some(VOBitsClearingScope::BlockOnly)
                } else {
                    // Young objects are allocated into empty lines.
                    // Only clear unmarked lines.
                    let line_mark_state = self.line_mark_state.load(Ordering::SeqCst);
                    Some(VOBitsClearingScope::Line {
                        state: line_mark_state,
                    })
                }
            } else {
                // GenImmix nursery GC.  We do nothing to the ImmixSpace because the nursery is a
                // separate CopySpace.  It'll clear its own VO bits.
                None
            };

            if let Some(scope) = maybe_scope {
                let work_packets = self
                    .chunk_map
                    .generate_tasks(|chunk| Box::new(ClearVOBitsAfterPrepare { chunk, scope }));
                self.scheduler.work_buckets[WorkBucketStage::ClearVOBits].bulk_add(work_packets);
            }
        }
    }

    /// Release for the immix space.
    pub fn release(&mut self, major_gc: bool) {
        if major_gc {
            // Update line_unavail_state for hole searching after this GC.
            if !super::BLOCK_ONLY {
                self.line_unavail_state.store(
                    self.line_mark_state.load(Ordering::Acquire),
                    Ordering::Release,
                );
            }
        }
        // Clear reusable blocks list
        if !super::BLOCK_ONLY {
            self.reusable_blocks.reset();
        }
        // Sweep chunks and blocks
        let work_packets = self.generate_sweep_tasks();
        self.scheduler().work_buckets[WorkBucketStage::Release].bulk_add(work_packets);

        self.lines_consumed.store(0, Ordering::Relaxed);
    }

    /// This is called when a GC finished.
    /// Return whether this GC was a defrag GC, as a plan may want to know this.
    pub fn end_of_gc(&mut self) -> bool {
        let did_defrag = self.defrag.in_defrag();
        if super::DEFRAG {
            self.defrag.reset_in_defrag();
        }
        did_defrag
    }

    /// Generate chunk sweep tasks
    fn generate_sweep_tasks(&self) -> Vec<Box<dyn GCWork<VM>>> {
        self.defrag.mark_histograms.lock().clear();
        // # Safety: ImmixSpace reference is always valid within this collection cycle.
        let space = unsafe { &*(self as *const Self) };
        let epilogue = Arc::new(FlushPageResource {
            space,
            counter: AtomicUsize::new(0),
        });
        let tasks = self.chunk_map.generate_tasks(|chunk| {
            Box::new(SweepChunk {
                space,
                chunk,
                epilogue: epilogue.clone(),
            })
        });
        epilogue.counter.store(tasks.len(), Ordering::SeqCst);
        tasks
    }

    /// Release a block.
    pub fn release_block(&self, block: Block) {
        block.deinit();
        self.pr.release_block(block);
    }

    /// Allocate a clean block.
    pub fn get_clean_block(&self, tls: VMThread, copy: bool, no_gc_on_fail: bool) -> Option<Block> {
        let block_address = self.acquire(tls, Block::PAGES, no_gc_on_fail);
        if block_address.is_zero() {
            return None;
        }
        self.defrag.notify_new_clean_block(copy);
        let block = Block::from_aligned_address(block_address);
        block.init(copy);
        self.chunk_map.set(block.chunk(), ChunkState::Allocated);
        self.lines_consumed
            .fetch_add(Block::LINES, Ordering::SeqCst);
        Some(block)
    }

    /// Pop a reusable block from the reusable block list.
    pub fn get_reusable_block(&self, copy: bool) -> Option<Block> {
        if super::BLOCK_ONLY {
            return None;
        }
        loop {
            if let Some(block) = self.reusable_blocks.pop() {
                // Skip blocks that should be evacuated.
                if copy && block.is_defrag_source() {
                    continue;
                }

                // Get available lines. Do this before block.init which will reset block state.
                let lines_delta = match block.get_state() {
                    BlockState::Reusable { unavailable_lines } => {
                        Block::LINES - unavailable_lines as usize
                    }
                    BlockState::Unmarked => Block::LINES,
                    _ => unreachable!("{:?} {:?}", block, block.get_state()),
                };
                self.lines_consumed.fetch_add(lines_delta, Ordering::SeqCst);

                block.init(copy);
                return Some(block);
            } else {
                return None;
            }
        }
    }

    /// Trace and mark objects without evacuation.
    pub fn trace_object_without_moving(
        &self,
        queue: &mut impl ObjectQueue,
        object: ObjectReference,
    ) -> ObjectReference {
        #[cfg(feature = "vo_bit")]
        vo_bit::helper::on_trace_object::<VM>(object);

        if self.attempt_mark(object, self.mark_state) {
            // Mark block and lines
            if !super::BLOCK_ONLY {
                if !super::MARK_LINE_AT_SCAN_TIME {
                    self.mark_lines(object);
                }
            } else {
                Block::containing(object).set_state(BlockState::Marked);
            }

            #[cfg(feature = "vo_bit")]
            vo_bit::helper::on_object_marked::<VM>(object);

            // Visit node
            queue.enqueue(object);
            self.unlog_object_if_needed(object);
            return object;
        }
        object
    }

    /// Trace object and do evacuation if required.
    #[allow(clippy::assertions_on_constants)]
    pub fn trace_object_with_opportunistic_copy(
        &self,
        queue: &mut impl ObjectQueue,
        object: ObjectReference,
        semantics: CopySemantics,
        worker: &mut GCWorker<VM>,
        nursery_collection: bool,
    ) -> ObjectReference {
        let copy_context = worker.get_copy_context_mut();
        debug_assert!(!super::BLOCK_ONLY);

        #[cfg(feature = "vo_bit")]
        vo_bit::helper::on_trace_object::<VM>(object);

        let forwarding_status = object_forwarding::attempt_to_forward::<VM>(object);
        if object_forwarding::state_is_forwarded_or_being_forwarded(forwarding_status) {
            // We lost the forwarding race as some other thread has set the forwarding word; wait
            // until the object has been forwarded by the winner. Note that the object may not
            // necessarily get forwarded since Immix opportunistically moves objects.
            #[allow(clippy::let_and_return)]
            let new_object =
                object_forwarding::spin_and_get_forwarded_object::<VM>(object, forwarding_status);
            #[cfg(debug_assertions)]
            {
                if new_object == object {
                    debug_assert!(
                        self.is_marked(object) || self.defrag.space_exhausted() || self.is_pinned(object),
                        "Forwarded object is the same as original object {} even though it should have been copied",
                        object,
                    );
                } else {
                    // new_object != object
                    debug_assert!(
                        !Block::containing(new_object).is_defrag_source(),
                        "Block {:?} containing forwarded object {} should not be a defragmentation source",
                        Block::containing(new_object),
                        new_object,
                    );
                }
            }
            new_object
        } else if self.is_marked(object) {
            // We won the forwarding race but the object is already marked so we clear the
            // forwarding status and return the unmoved object
            object_forwarding::clear_forwarding_bits::<VM>(object);
            object
        } else {
            // We won the forwarding race; actually forward and copy the object if it is not pinned
            // and we have sufficient space in our copy allocator
            let new_object = if self.is_pinned(object)
                || (!nursery_collection && self.defrag.space_exhausted())
            {
                self.attempt_mark(object, self.mark_state);
                object_forwarding::clear_forwarding_bits::<VM>(object);
                Block::containing(object).set_state(BlockState::Marked);

                #[cfg(feature = "vo_bit")]
                vo_bit::helper::on_object_marked::<VM>(object);

                if !super::MARK_LINE_AT_SCAN_TIME {
                    self.mark_lines(object);
                }

                object
            } else {
                // We are forwarding objects. When the copy allocator allocates the block, it should
                // mark the block. So we do not need to explicitly mark it here.

                object_forwarding::forward_object::<VM>(
                    object,
                    semantics,
                    copy_context,
                    |_new_object| {
                        #[cfg(feature = "vo_bit")]
                        vo_bit::helper::on_object_forwarded::<VM>(_new_object);
                    },
                )
            };
            debug_assert_eq!(
                Block::containing(new_object).get_state(),
                BlockState::Marked
            );

            queue.enqueue(new_object);
            debug_assert!(new_object.is_live());
            self.unlog_object_if_needed(new_object);
            new_object
        }
    }

    fn unlog_object_if_needed(&self, object: ObjectReference) {
        if self.space_args.unlog_object_when_traced {
            // Make sure the side metadata for the line can fit into one byte. For smaller line size, we should
            // use `mark_as_unlogged` instead to mark the bit.
            const_assert!(
                Line::BYTES
                    >= (1
                        << (crate::util::constants::LOG_BITS_IN_BYTE
                            + crate::util::constants::LOG_MIN_OBJECT_SIZE))
            );
            const_assert_eq!(
                crate::vm::object_model::specs::VMGlobalLogBitSpec::LOG_NUM_BITS,
                0
            ); // We should put this to the addition, but type casting is not allowed in constant assertions.

            // Every immix line is 256 bytes, which is mapped to 4 bytes in the side metadata.
            // If we have one object in the line that is mature, we can assume all the objects in the line are mature objects.
            // So we can just mark the byte.
            VM::VMObjectModel::GLOBAL_LOG_BIT_SPEC
                .mark_byte_as_unlogged::<VM>(object, Ordering::Relaxed);
        }
    }

    /// Mark all the lines that the given object spans.
    #[allow(clippy::assertions_on_constants)]
    pub fn mark_lines(&self, object: ObjectReference) {
        debug_assert!(!super::BLOCK_ONLY);
        Line::mark_lines_for_object::<VM>(object, self.line_mark_state.load(Ordering::Acquire));
    }

    /// Atomically mark an object.
    fn attempt_mark(&self, object: ObjectReference, mark_state: u8) -> bool {
        loop {
            let old_value = VM::VMObjectModel::LOCAL_MARK_BIT_SPEC.load_atomic::<VM, u8>(
                object,
                None,
                Ordering::SeqCst,
            );
            if old_value == mark_state {
                return false;
            }

            if VM::VMObjectModel::LOCAL_MARK_BIT_SPEC
                .compare_exchange_metadata::<VM, u8>(
                    object,
                    old_value,
                    mark_state,
                    None,
                    Ordering::SeqCst,
                    Ordering::SeqCst,
                )
                .is_ok()
            {
                break;
            }
        }
        true
    }

    /// Check if an object is marked.
    fn is_marked_with(&self, object: ObjectReference, mark_state: u8) -> bool {
        let old_value = VM::VMObjectModel::LOCAL_MARK_BIT_SPEC.load_atomic::<VM, u8>(
            object,
            None,
            Ordering::SeqCst,
        );
        old_value == mark_state
    }

    pub(crate) fn is_marked(&self, object: ObjectReference) -> bool {
        self.is_marked_with(object, self.mark_state)
    }

    /// Check if an object is pinned.
    fn is_pinned(&self, _object: ObjectReference) -> bool {
        #[cfg(feature = "object_pinning")]
        return self.is_object_pinned(_object);

        #[cfg(not(feature = "object_pinning"))]
        false
    }

    /// Hole searching.
    ///
    /// Linearly scan lines in a block to search for the next
    /// hole, starting from the given line. If we find available lines,
    /// return a tuple of the start line and the end line (non-inclusive).
    ///
    /// Returns None if the search could not find any more holes.
    #[allow(clippy::assertions_on_constants)]
    pub fn get_next_available_lines(&self, search_start: Line) -> Option<(Line, Line)> {
        debug_assert!(!super::BLOCK_ONLY);
        let unavail_state = self.line_unavail_state.load(Ordering::Acquire);
        let current_state = self.line_mark_state.load(Ordering::Acquire);
        let block = search_start.block();
        let mark_data = block.line_mark_table();
        let start_cursor = search_start.get_index_within_block();
        let mut cursor = start_cursor;
        // Find start
        while cursor < mark_data.len() {
            let mark = mark_data.get(cursor);
            if mark != unavail_state && mark != current_state {
                break;
            }
            cursor += 1;
        }
        if cursor == mark_data.len() {
            return None;
        }
        let start = search_start.next_nth(cursor - start_cursor);
        // Find limit
        while cursor < mark_data.len() {
            let mark = mark_data.get(cursor);
            if mark == unavail_state || mark == current_state {
                break;
            }
            cursor += 1;
        }
        let end = search_start.next_nth(cursor - start_cursor);
        debug_assert!(RegionIterator::<Line>::new(start, end)
            .all(|line| !line.is_marked(unavail_state) && !line.is_marked(current_state)));
        Some((start, end))
    }

    pub fn is_last_gc_exhaustive(did_defrag_for_last_gc: bool) -> bool {
        if super::DEFRAG {
            did_defrag_for_last_gc
        } else {
            // If defrag is disabled, every GC is exhaustive.
            true
        }
    }

    pub(crate) fn get_pages_allocated(&self) -> usize {
        self.lines_consumed.load(Ordering::SeqCst) >> (LOG_BYTES_IN_PAGE - Line::LOG_BYTES as u8)
    }

    /// Post copy routine for Immix copy contexts
    fn post_copy(&self, object: ObjectReference, _bytes: usize) {
        // Mark the object
        VM::VMObjectModel::LOCAL_MARK_BIT_SPEC.store_atomic::<VM, u8>(
            object,
            self.mark_state,
            None,
            Ordering::SeqCst,
        );
        // Mark the line
        if !super::MARK_LINE_AT_SCAN_TIME {
            self.mark_lines(object);
        }
    }
}

/// A work packet to prepare each block for a major GC.
/// Performs the action on a range of chunks.
pub struct PrepareBlockState<VM: VMBinding> {
    pub space: &'static ImmixSpace<VM>,
    pub chunk: Chunk,
    pub defrag_threshold: Option<usize>,
}

impl<VM: VMBinding> PrepareBlockState<VM> {
    /// Clear object mark table
    fn reset_object_mark(&self) {
        // NOTE: We reset the mark bits because cyclic mark bit is currently not supported, yet.
        // See `ImmixSpace::prepare`.
        if let MetadataSpec::OnSide(side) = *VM::VMObjectModel::LOCAL_MARK_BIT_SPEC {
            side.bzero_metadata(self.chunk.start(), Chunk::BYTES);
        }
        if self.space.space_args.reset_log_bit_in_major_gc {
            if let MetadataSpec::OnSide(side) = *VM::VMObjectModel::GLOBAL_LOG_BIT_SPEC {
                // We zero all the log bits in major GC, and for every object we trace, we will mark the log bit again.
                side.bzero_metadata(self.chunk.start(), Chunk::BYTES);
            } else {
                // If the log bit is not in side metadata, we cannot bulk zero. We can either
                // clear the bit for dead objects in major GC, or clear the log bit for new
                // objects. In either cases, we do not need to set log bit at tracing.
                unimplemented!("We cannot bulk zero unlogged bit.")
            }
        }
    }
}

impl<VM: VMBinding> GCWork<VM> for PrepareBlockState<VM> {
    fn do_work(&mut self, _worker: &mut GCWorker<VM>, _mmtk: &'static MMTK<VM>) {
        // Clear object mark table for this chunk
        self.reset_object_mark();
        // Iterate over all blocks in this chunk
        for block in self.chunk.iter_region::<Block>() {
            let state = block.get_state();
            // Skip unallocated blocks.
            if state == BlockState::Unallocated {
                continue;
            }
            // Check if this block needs to be defragmented.
            let is_defrag_source = if !super::DEFRAG {
                // Do not set any block as defrag source if defrag is disabled.
                false
            } else if super::DEFRAG_EVERY_BLOCK {
                // Set every block as defrag source if so desired.
                true
            } else if let Some(defrag_threshold) = self.defrag_threshold {
                // This GC is a defrag GC.
                block.get_holes() > defrag_threshold
            } else {
                // Not a defrag GC.
                false
            };
            block.set_as_defrag_source(is_defrag_source);
            // Clear block mark data.
            block.set_state(BlockState::Unmarked);
            debug_assert!(!block.get_state().is_reusable());
            debug_assert_ne!(block.get_state(), BlockState::Marked);
        }
    }
}

/// Chunk sweeping work packet.
struct SweepChunk<VM: VMBinding> {
    space: &'static ImmixSpace<VM>,
    chunk: Chunk,
    /// A destructor invoked when all `SweepChunk` packets are finished.
    epilogue: Arc<FlushPageResource<VM>>,
}

impl<VM: VMBinding> GCWork<VM> for SweepChunk<VM> {
    fn do_work(&mut self, _worker: &mut GCWorker<VM>, mmtk: &'static MMTK<VM>) {
        assert_eq!(self.space.chunk_map.get(self.chunk), ChunkState::Allocated);

        let mut histogram = self.space.defrag.new_histogram();
        let line_mark_state = if super::BLOCK_ONLY {
            None
        } else {
            Some(self.space.line_mark_state.load(Ordering::Acquire))
        };
        // Hints for clearing side forwarding bits.
        let is_moving_gc = mmtk.get_plan().current_gc_may_move_object();
        let is_defrag_gc = self.space.defrag.in_defrag();
        // number of allocated blocks.
        let mut allocated_blocks = 0;
        // Iterate over all allocated blocks in this chunk.
        for block in self
            .chunk
            .iter_region::<Block>()
            .filter(|block| block.get_state() != BlockState::Unallocated)
        {
            // Clear side forwarding bits.
            // In the beginning of the next GC, no side forwarding bits shall be set.
            // In this way, we can omit clearing forwarding bits when copying object.
            // See `GCWorkerCopyContext::post_copy`.
            // Note, `block.sweep()` overwrites `DEFRAG_STATE_TABLE` with the number of holes,
            // but we need it to know if a block is a defrag source.
            // We clear forwarding bits before `block.sweep()`.
            if let MetadataSpec::OnSide(side) = *VM::VMObjectModel::LOCAL_FORWARDING_BITS_SPEC {
                if is_moving_gc {
                    let objects_may_move = if is_defrag_gc {
                        // If it is a defrag GC, we only clear forwarding bits for defrag sources.
                        block.is_defrag_source()
                    } else {
                        // Otherwise, it must be a nursery GC of StickyImmix with copying nursery.
                        // We don't have information about which block contains moved objects,
                        // so we have to clear forwarding bits for all blocks.
                        true
                    };
                    if objects_may_move {
                        side.bzero_metadata(block.start(), Block::BYTES);
                    }
                }
            }

            if !block.sweep(self.space, &mut histogram, line_mark_state) {
                // Block is live. Increment the allocated block count.
                allocated_blocks += 1;
            }
        }
        probe!(mmtk, sweep_chunk, allocated_blocks);
        // Set this chunk as free if there is not live blocks.
        if allocated_blocks == 0 {
            self.space.chunk_map.set(self.chunk, ChunkState::Free)
        }
        self.space.defrag.add_completed_mark_histogram(histogram);
        self.epilogue.finish_one_work_packet();
    }
}

/// Count number of remaining work pacets, and flush page resource if all packets are finished.
struct FlushPageResource<VM: VMBinding> {
    space: &'static ImmixSpace<VM>,
    counter: AtomicUsize,
}

impl<VM: VMBinding> FlushPageResource<VM> {
    /// Called after a related work packet is finished.
    fn finish_one_work_packet(&self) {
        if 1 == self.counter.fetch_sub(1, Ordering::SeqCst) {
            // We've finished releasing all the dead blocks to the BlockPageResource's thread-local queues.
            // Now flush the BlockPageResource.
            self.space.flush_page_resource()
        }
    }
}

impl<VM: VMBinding> Drop for FlushPageResource<VM> {
    fn drop(&mut self) {
        epilogue::debug_assert_counter_zero(&self.counter, "FlushPageResource::counter");
    }
}

use crate::policy::copy_context::PolicyCopyContext;
use crate::util::alloc::Allocator;
use crate::util::alloc::ImmixAllocator;

/// Normal immix copy context. It has one copying Immix allocator.
/// Most immix plans use this copy context.
pub struct ImmixCopyContext<VM: VMBinding> {
    allocator: ImmixAllocator<VM>,
}

impl<VM: VMBinding> PolicyCopyContext for ImmixCopyContext<VM> {
    type VM = VM;

    fn prepare(&mut self) {
        self.allocator.reset();
    }
    fn release(&mut self) {
        self.allocator.reset();
    }
    fn alloc_copy(
        &mut self,
        _original: ObjectReference,
        bytes: usize,
        align: usize,
        offset: usize,
    ) -> Address {
        self.allocator.alloc(bytes, align, offset)
    }
    fn post_copy(&mut self, obj: ObjectReference, bytes: usize) {
        self.get_space().post_copy(obj, bytes)
    }
}

impl<VM: VMBinding> ImmixCopyContext<VM> {
    pub(crate) fn new(
        tls: VMWorkerThread,
        context: Arc<AllocatorContext<VM>>,
        space: &'static ImmixSpace<VM>,
    ) -> Self {
        ImmixCopyContext {
            allocator: ImmixAllocator::new(tls.0, Some(space), context, true),
        }
    }

    fn get_space(&self) -> &ImmixSpace<VM> {
        self.allocator.immix_space()
    }
}

/// Hybrid Immix copy context. It includes two different immix allocators. One with `copy = true`
/// is used for defrag GCs, and the other is used for other purposes (such as promoting objects from
/// nursery to Immix mature space). This is used by generational immix.
pub struct ImmixHybridCopyContext<VM: VMBinding> {
    copy_allocator: ImmixAllocator<VM>,
    defrag_allocator: ImmixAllocator<VM>,
}

impl<VM: VMBinding> PolicyCopyContext for ImmixHybridCopyContext<VM> {
    type VM = VM;

    fn prepare(&mut self) {
        self.copy_allocator.reset();
        self.defrag_allocator.reset();
    }
    fn release(&mut self) {
        self.copy_allocator.reset();
        self.defrag_allocator.reset();
    }
    fn alloc_copy(
        &mut self,
        _original: ObjectReference,
        bytes: usize,
        align: usize,
        offset: usize,
    ) -> Address {
        if self.get_space().in_defrag() {
            self.defrag_allocator.alloc(bytes, align, offset)
        } else {
            self.copy_allocator.alloc(bytes, align, offset)
        }
    }
    fn post_copy(&mut self, obj: ObjectReference, bytes: usize) {
        self.get_space().post_copy(obj, bytes)
    }
}

impl<VM: VMBinding> ImmixHybridCopyContext<VM> {
    pub(crate) fn new(
        tls: VMWorkerThread,
        context: Arc<AllocatorContext<VM>>,
        space: &'static ImmixSpace<VM>,
    ) -> Self {
        ImmixHybridCopyContext {
            copy_allocator: ImmixAllocator::new(tls.0, Some(space), context.clone(), false),
            defrag_allocator: ImmixAllocator::new(tls.0, Some(space), context, true),
        }
    }

    fn get_space(&self) -> &ImmixSpace<VM> {
        // Both copy allocators should point to the same space.
        debug_assert_eq!(
            self.defrag_allocator.immix_space().common().descriptor,
            self.copy_allocator.immix_space().common().descriptor
        );
        // Just get the space from either allocator
        self.defrag_allocator.immix_space()
    }
}

#[cfg(feature = "vo_bit")]
#[derive(Clone, Copy)]
enum VOBitsClearingScope {
    /// Clear all VO bits in all blocks.
    FullGC,
    /// Clear unmarked blocks, only.
    BlockOnly,
    /// Clear unmarked lines, only.  (i.e. lines with line mark state **not** equal to `state`).
    Line { state: u8 },
}

/// A work packet to clear VO bit metadata after Prepare.
#[cfg(feature = "vo_bit")]
struct ClearVOBitsAfterPrepare {
    chunk: Chunk,
    scope: VOBitsClearingScope,
}

#[cfg(feature = "vo_bit")]
impl<VM: VMBinding> GCWork<VM> for ClearVOBitsAfterPrepare {
    fn do_work(&mut self, _worker: &mut GCWorker<VM>, _mmtk: &'static MMTK<VM>) {
        match self.scope {
            VOBitsClearingScope::FullGC => {
                vo_bit::bzero_vo_bit(self.chunk.start(), Chunk::BYTES);
            }
            VOBitsClearingScope::BlockOnly => {
                self.clear_blocks(None);
            }
            VOBitsClearingScope::Line { state } => {
                self.clear_blocks(Some(state));
            }
        }
    }
}

#[cfg(feature = "vo_bit")]
impl ClearVOBitsAfterPrepare {
    fn clear_blocks(&mut self, line_mark_state: Option<u8>) {
        for block in self
            .chunk
            .iter_region::<Block>()
            .filter(|block| block.get_state() != BlockState::Unallocated)
        {
            block.clear_vo_bits_for_unmarked_regions(line_mark_state);
        }
    }
}