Merge branch 'master' into fix/modbuf-global-gc-bug

.github/scripts/ci-common.sh

@@ -9,6 +9,9 @@ cargo_toml=$project_root/Cargo.toml
 
 dummyvm_toml=$project_root/docs/dummyvm/Cargo.toml
 
+# Pin certain deps for our MSRV
+cargo update -p [email protected] --precise 0.5.5 # This can be removed once we move to Rust 1.81 or newer
+
 # Repeat a command for all the features. Requires the command as one argument (with double quotes)
 for_all_features() {
     # without mutually exclusive features

.github/workflows/minimal-tests-core.yml

@@ -87,7 +87,7 @@ jobs:
         target:
           - { os: ubuntu-22.04, triple: x86_64-unknown-linux-gnu }
           - { os: ubuntu-22.04, triple: i686-unknown-linux-gnu }
-          - { os: macos-12, triple: x86_64-apple-darwin }
+          - { os: macos-15, triple: x86_64-apple-darwin }
         rust: ${{ fromJson(needs.setup-test-matrix.outputs.rust )}}
 
     name: style-check/${{ matrix.target.triple }}/${{ matrix.rust }}

CHANGELOG.md

@@ -1,3 +1,23 @@
+0.30.0 (2024-12-20)
+===
+
+## What's Changed
+
+### policy
+* Allow setting object metadata for VM space objects. Expose VO bit under a feature. by @qinsoon in https://github.com/mmtk/mmtk-core/pull/1248
+
+### Misc
+* Fix clippy warnings for Rust 1.83 by @wks in https://github.com/mmtk/mmtk-core/pull/1242
+* Annotate mmap ranges using PR_SET_VMA by @wks in https://github.com/mmtk/mmtk-core/pull/1236
+* Fix warnings for lifetime in MmapAnnotation impl by @qinsoon in https://github.com/mmtk/mmtk-core/pull/1244
+* Collect live bytes per space, and report by space by @qinsoon in https://github.com/mmtk/mmtk-core/pull/1238
+* Minor changes for debugging. by @wks in https://github.com/mmtk/mmtk-core/pull/1245
+* Use macos-15 for style check by @qinsoon in https://github.com/mmtk/mmtk-core/pull/1249
+* Check the option before aggregating live bytes data. Panic if the option is enabled on malloc space. by @qinsoon in https://github.com/mmtk/mmtk-core/pull/1250
+
+
+**Full Changelog**: https://github.com/mmtk/mmtk-core/compare/v0.29.0...v0.30.0
+
 0.29.0 (2024-11-08)
 ===
 

Cargo.toml

@@ -1,6 +1,6 @@
 [package]
 name = "mmtk"
-version = "0.29.0"
+version = "0.30.0"
 authors = ["The MMTk Developers <>"]
 edition = "2021"
 license = "MIT OR Apache-2.0"
@@ -38,7 +38,7 @@ log = { version = "0.4", features = ["max_level_trace", "release_max_level_off"]
 memoffset = "0.9"
 mimalloc-sys = { version = "0.1.6", optional = true }
 # MMTk macros - we have to specify a version here in order to publish the crate, even though we use the dependency from a local path.
-mmtk-macros = { version="0.29.0", path = "macros/" }
+mmtk-macros = { version="0.30.0", path = "macros/" }
 num_cpus = "1.8"
 num-traits = "0.2"
 pfm = { version = "0.1.1", optional = true }
@@ -172,13 +172,27 @@ work_packet_stats = []
 # Count the malloc'd memory into the heap size
 malloc_counted_size = []
 
-# Count the size of all live objects in GC
-count_live_bytes_in_gc = []
-
 # Workaround a problem where bpftrace scripts (see tools/tracing/timeline/capture.bt) cannot
 # capture the type names of work packets.
 bpftrace_workaround = []
 
+# Disable mmap annotations.
+# All invocations of `mmap` in mmtk-core are accompanied by calls of `prctl` with
+# `PR_SET_VMA_ANON_NAME` to annotate the mmap ranges with human-readable names.  It is enabled by
+# default and should work fine even with large heap sizes.  However, if this is causing problems,
+# users can disable such annotations by enabling this Cargo feature.
+no_mmap_annotation = []
+
+# Allow the binding to access Valid Object (VO) bit.
+# MMTk uses VO bit to identify a valid object. Thus VO bit is carefully managed by MMTk in cooperation with the binding.
+# So normally this feature is not needed, and its use should be discouraged.
+# However, in rare cases, a binding may want to directly access and manipulate VO bit. For example, a binding cannot cooperate
+# with MMTk to identify each object for setting the VO bit. This is usually due to the limitation of the VM.
+# In such cases, a binding may use this feature and manipulate the VO bit to their needs. The binding's manipulation on VO bit
+# may violate MMTk's semantics, and may result in undefined behaviors for VO bit related APIs. This feature should
+# only be used if you understand how VO bit works internally. Use at your own risk.
+vo_bit_access = []
+
 # Do not modify the following line - ci-common.sh matches it
 # -- Mutally exclusive features --
 # Only one feature from each group can be provided. Otherwise build will fail.

docs/team/release.md

@@ -13,9 +13,9 @@ MMTk releases include MMTk core and the officially supported bindings. They shar
 The current releases include the following bindings:
 * OpenJDK
 * JikesRVM
+* Julia
 
 The current releases do not include the following bindings:
-* Julia: We have made good progress on the binding development, and we will include it soon.
 * Ruby: We have made good progress on the binding development, and we will include it soon.
 * V8: We currently lack of resources to work on the binding.
 

docs/userguide/src/migration/prefix.md

@@ -30,6 +30,44 @@ Notes for the mmtk-core developers:
 
 <!-- Insert new versions here -->
 
+## 0.30.0
+
+### `live_bytes_in_last_gc` becomes a runtime option, and returns a map for live bytes in each space
+
+```admonish tldr
+`count_live_bytes_in_gc` is now a runtime option instead of a features (build-time), and we collect
+live bytes statistics per space. Correspondingly, `memory_manager::live_bytes_in_last_gc` now returns a map for
+live bytes in each space.
+```
+
+API changes:
+
+-   module `util::options`
+    +   `Options` includes `count_live_bytes_in_gc`, which defaults to `false`. This can be turned on at run-time.
+    +   The old `count_live_bytes_in_gc` feature is removed.
+-   module `memory_manager`
+    +   `live_bytes_in_last_gc` now returns a `HashMap<&'static str, LiveBytesStats>`. The keys are
+        strings for space names, and the values are statistics for live bytes in the space.
+
+See also:
+
+-   PR: <https://github.com/mmtk/mmtk-core/pull/1238>
+
+
+### mmap-related functions require annotation
+
+```admonish tldr
+Memory-mapping functions in `mmtk::util::memory` now take an additional `MmapAnnotation` argument.
+```
+
+API changes:
+
+-   module `util::memory`
+    +   The following functions take an additional `MmapAnnotation` argument.
+        *   `dzmmap`
+        *   `dzmmap_noreplace`
+        *   `mmap_noreserve`
+
 ## 0.28.0
 
 ### `handle_user_collection_request` returns `bool`

macros/Cargo.toml

@@ -1,7 +1,7 @@
 [package]
 name = "mmtk-macros"
 # the macro crate uses the same version as mmtk-core
-version = "0.29.0"
+version = "0.30.0"
 edition = "2021"
 license = "MIT OR Apache-2.0"
 description = "MMTk macros provides procedural macros used by mmtk-core."

src/global_state.rs

@@ -1,9 +1,9 @@
+use atomic_refcell::AtomicRefCell;
+use std::collections::HashMap;
 use std::sync::atomic::{AtomicBool, AtomicUsize, Ordering};
 use std::sync::Mutex;
 use std::time::Instant;
 
-use atomic_refcell::AtomicRefCell;
-
 /// This stores some global states for an MMTK instance.
 /// Some MMTK components like plans and allocators may keep an reference to the struct, and can access it.
 // This used to be a part of the `BasePlan`. In that case, any component that accesses
@@ -45,9 +45,8 @@ pub struct GlobalState {
     /// A counteer that keeps tracks of the number of bytes allocated by malloc
     #[cfg(feature = "malloc_counted_size")]
     pub(crate) malloc_bytes: AtomicUsize,
-    /// This stores the size in bytes for all the live objects in last GC. This counter is only updated in the GC release phase.
-    #[cfg(feature = "count_live_bytes_in_gc")]
-    pub(crate) live_bytes_in_last_gc: AtomicUsize,
+    /// This stores the live bytes and the used bytes (by pages) for each space in last GC. This counter is only updated in the GC release phase.
+    pub(crate) live_bytes_in_last_gc: AtomicRefCell<HashMap<&'static str, LiveBytesStats>>,
 }
 
 impl GlobalState {
@@ -183,16 +182,6 @@ impl GlobalState {
     pub(crate) fn decrease_malloc_bytes_by(&self, size: usize) {
         self.malloc_bytes.fetch_sub(size, Ordering::SeqCst);
     }
-
-    #[cfg(feature = "count_live_bytes_in_gc")]
-    pub fn get_live_bytes_in_last_gc(&self) -> usize {
-        self.live_bytes_in_last_gc.load(Ordering::SeqCst)
-    }
-
-    #[cfg(feature = "count_live_bytes_in_gc")]
-    pub fn set_live_bytes_in_last_gc(&self, size: usize) {
-        self.live_bytes_in_last_gc.store(size, Ordering::SeqCst);
-    }
 }
 
 impl Default for GlobalState {
@@ -213,8 +202,7 @@ impl Default for GlobalState {
             allocation_bytes: AtomicUsize::new(0),
             #[cfg(feature = "malloc_counted_size")]
             malloc_bytes: AtomicUsize::new(0),
-            #[cfg(feature = "count_live_bytes_in_gc")]
-            live_bytes_in_last_gc: AtomicUsize::new(0),
+            live_bytes_in_last_gc: AtomicRefCell::new(HashMap::new()),
         }
     }
 }
@@ -225,3 +213,15 @@ pub enum GcStatus {
     GcPrepare,
     GcProper,
 }
+
+/// Statistics for the live bytes in the last GC. The statistics is per space.
+#[derive(Copy, Clone, Debug)]
+pub struct LiveBytesStats {
+    /// Total accumulated bytes of live objects in the space.
+    pub live_bytes: usize,
+    /// Total pages used by the space.
+    pub used_pages: usize,
+    /// Total bytes used by the space, computed from `used_pages`.
+    /// The ratio of live_bytes and used_bytes reflects the utilization of the memory in the space.
+    pub used_bytes: usize,
+}

src/lib.rs

@@ -46,6 +46,7 @@ pub(crate) use mmtk::MMAPPER;
 pub use mmtk::MMTK;
 
 mod global_state;
+pub use crate::global_state::LiveBytesStats;
 
 mod policy;
 

src/memory_manager.rs

@@ -26,6 +26,8 @@ use crate::vm::slot::MemorySlice;
 use crate::vm::ReferenceGlue;
 use crate::vm::VMBinding;
 
+use std::collections::HashMap;
+
 /// Initialize an MMTk instance. A VM should call this method after creating an [`crate::MMTK`]
 /// instance but before using any of the methods provided in MMTk (except `process()` and `process_bulk()`).
 ///
@@ -531,16 +533,18 @@ pub fn free_bytes<VM: VMBinding>(mmtk: &MMTK<VM>) -> usize {
     mmtk.get_plan().get_free_pages() << LOG_BYTES_IN_PAGE
 }
 
-/// Return the size of all the live objects in bytes in the last GC. MMTk usually accounts for memory in pages.
+/// Return a hash map for live bytes statistics in the last GC for each space.
+///
+/// MMTk usually accounts for memory in pages by each space.
 /// This is a special method that we count the size of every live object in a GC, and sum up the total bytes.
-/// We provide this method so users can compare with `used_bytes` (which does page accounting), and know if
-/// the heap is fragmented.
+/// We provide this method so users can use [`crate::LiveBytesStats`] to know if
+/// the space is fragmented.
 /// The value returned by this method is only updated when we finish tracing in a GC. A recommended timing
 /// to call this method is at the end of a GC (e.g. when the runtime is about to resume threads).
-#[cfg(feature = "count_live_bytes_in_gc")]
-pub fn live_bytes_in_last_gc<VM: VMBinding>(mmtk: &MMTK<VM>) -> usize {
-    use std::sync::atomic::Ordering;
-    mmtk.state.live_bytes_in_last_gc.load(Ordering::SeqCst)
+pub fn live_bytes_in_last_gc<VM: VMBinding>(
+    mmtk: &MMTK<VM>,
+) -> HashMap<&'static str, crate::LiveBytesStats> {
+    mmtk.state.live_bytes_in_last_gc.borrow().clone()
 }
 
 /// Return the starting address of the heap. *Note that currently MMTk uses

src/mmtk.rs

@@ -12,6 +12,7 @@ use crate::util::address::ObjectReference;
 use crate::util::analysis::AnalysisManager;
 use crate::util::finalizable_processor::FinalizableProcessor;
 use crate::util::heap::gc_trigger::GCTrigger;
+use crate::util::heap::layout::heap_parameters::MAX_SPACES;
 use crate::util::heap::layout::vm_layout::VMLayout;
 use crate::util::heap::layout::{self, Mmapper, VMMap};
 use crate::util::heap::HeapMeta;
@@ -26,6 +27,7 @@ use crate::util::statistics::stats::Stats;
 use crate::vm::ReferenceGlue;
 use crate::vm::VMBinding;
 use std::cell::UnsafeCell;
+use std::collections::HashMap;
 use std::default::Default;
 use std::sync::atomic::{AtomicBool, Ordering};
 use std::sync::Arc;
@@ -526,4 +528,72 @@ impl<VM: VMBinding> MMTK<VM> {
             space.enumerate_objects(&mut enumerator);
         })
     }
+
+    /// Aggregate a hash map of live bytes per space with the space stats to produce
+    /// a map of live bytes stats for the spaces.
+    pub(crate) fn aggregate_live_bytes_in_last_gc(
+        &self,
+        live_bytes_per_space: [usize; MAX_SPACES],
+    ) -> HashMap<&'static str, crate::LiveBytesStats> {
+        use crate::policy::space::Space;
+        let mut ret = HashMap::new();
+        self.get_plan().for_each_space(&mut |space: &dyn Space<VM>| {
+            let space_name = space.get_name();
+            let space_idx = space.get_descriptor().get_index();
+            let used_pages = space.reserved_pages();
+            if used_pages != 0 {
+                let used_bytes = crate::util::conversions::pages_to_bytes(used_pages);
+                let live_bytes = live_bytes_per_space[space_idx];
+                debug_assert!(
+                    live_bytes <= used_bytes,
+                    "Live bytes of objects in {} ({} bytes) is larger than used pages ({} bytes), something is wrong.",
+                    space_name, live_bytes, used_bytes
+                );
+                ret.insert(space_name, crate::LiveBytesStats {
+                    live_bytes,
+                    used_pages,
+                    used_bytes,
+                });
+            }
+        });
+        ret
+    }
+
+    /// Print VM maps.  It will print the memory ranges used by spaces as well as some attributes of
+    /// the spaces.
+    ///
+    /// -   "I": The space is immortal.  Its objects will never die.
+    /// -   "N": The space is non-movable.  Its objects will never move.
+    ///
+    /// Arguments:
+    /// *   `out`: the place to print the VM maps.
+    /// *   `space_name`: If `None`, print all spaces;
+    ///                   if `Some(n)`, only print the space whose name is `n`.
+    pub fn debug_print_vm_maps(
+        &self,
+        out: &mut impl std::fmt::Write,
+        space_name: Option<&str>,
+    ) -> Result<(), std::fmt::Error> {
+        let mut result_so_far = Ok(());
+        self.get_plan().for_each_space(&mut |space| {
+            if result_so_far.is_ok()
+                && (space_name.is_none() || space_name == Some(space.get_name()))
+            {
+                result_so_far = crate::policy::space::print_vm_map(space, out);
+            }
+        });
+        result_so_far
+    }
+
+    /// Initialize object metadata for a VM space object.
+    /// Objects in the VM space are allocated/managed by the binding. This function provides a way for
+    /// the binding to set object metadata in MMTk for an object in the space.
+    #[cfg(feature = "vm_space")]
+    pub fn initialize_vm_space_object(&self, object: crate::util::ObjectReference) {
+        use crate::policy::sft::SFT;
+        self.get_plan()
+            .base()
+            .vm_space
+            .initialize_object_metadata(object, false)
+    }
 }

src/plan/global.rs

@@ -390,7 +390,7 @@ pub struct CreateSpecificPlanArgs<'a, VM: VMBinding> {
     pub global_side_metadata_specs: Vec<SideMetadataSpec>,
 }
 
-impl<'a, VM: VMBinding> CreateSpecificPlanArgs<'a, VM> {
+impl<VM: VMBinding> CreateSpecificPlanArgs<'_, VM> {
     /// Get a PlanCreateSpaceArgs that can be used to create a space
     pub fn get_space_args(
         &mut self,

src/plan/markcompact/gc_work.rs

@@ -51,7 +51,6 @@ impl<VM: VMBinding> GCWork<VM> for UpdateReferences<VM> {
         mmtk.slot_logger.reset();
 
         // We do two passes of transitive closures. We clear the live bytes from the first pass.
-        #[cfg(feature = "count_live_bytes_in_gc")]
         mmtk.scheduler
             .worker_group
             .get_and_clear_worker_live_bytes();

src/plan/mutator_context.rs

@@ -81,7 +81,6 @@ impl<VM: VMBinding> std::fmt::Debug for MutatorConfig<VM> {
 /// A mutator is a per-thread data structure that manages allocations and barriers. It is usually highly coupled with the language VM.
 /// It is recommended for MMTk users 1) to have a mutator struct of the same layout in the thread local storage that can be accessed efficiently,
 /// and 2) to implement fastpath allocation and barriers for the mutator in the VM side.
-
 // We are trying to make this struct fixed-sized so that VM bindings can easily define a type to have the exact same layout as this struct.
 // Currently Mutator is fixed sized, and we should try keep this invariant:
 // - Allocators are fixed-length arrays of allocators.
@@ -256,7 +255,6 @@ impl<VM: VMBinding> Mutator<VM> {
 
 /// Each GC plan should provide their implementation of a MutatorContext. *Note that this trait is no longer needed as we removed
 /// per-plan mutator implementation and we will remove this trait as well in the future.*
-
 // TODO: We should be able to remove this trait, as we removed per-plan mutator implementation, and there is no other type that implements this trait.
 // The Mutator struct above is the only type that implements this trait. We should be able to merge them.
 pub trait MutatorContext<VM: VMBinding>: Send + 'static {