This release contains over 160 commits, focusing on the technical debts
cleaning including refactorings, bug fixes, test coverage improvement, and
option cleanups. The release also includes portability improvements per
upstream issues report.
New features:
- Add
EXTENT_ALLOC_FLAG_PINNED so custom extent-allocation hooks can
mark non-reclaimable mappings, such as HugeTLB pages, for preferential
reuse outside the decay and purge pipeline. Add the mallctl interfaces
stats.pinned, stats.arenas.<i>.pinned,
stats.arenas.<i>.extents.<j>.npinned,
stats.arenas.<i>.extents.<j>.pinned_bytes, and
stats.arenas.<i>.mutexes.extents_pinned.{counter} to report
pinned-memory usage and mutex statistics. (@binliu19: be2de8c)
- Allow resuming per-CPU arena selection via
thread.arena.
(@Algunenano: 68c35f6)
- Better align the contents of human-readable and JSON malloc statistics.
(@spredolac: 68fe1be, 30b1a41, 04aad97)
- Replace the runtime
experimental_infallible_new option with the
compile-time --enable-cxx-infallible-new option, enabling
compiler-level optimizations and optimization in move constructors,
and fix the new(std::nothrow) contract (@spredolac: 160ab9d, fe33667)
Incompatible changes:
- Adapt tcache fill and retention targets per bin to demand observed
between GC events, replacing the fixed refill/flush policy. Remove
seven legacy non-experimental controls: lg_tcache_nslots_mul,
tcache_nslots_small_min, tcache_nslots_small_max,
tcache_nslots_large, tcache_gc_delay_bytes,
lg_tcache_flush_small_div, and lg_tcache_flush_large_div.
Corresponding malloc_conf settings are silently ignored, matching
opt.* mallctls return ENOENT, and tcache_ncached_max remains
supported. (@spredolac: d13fe91)
Bug fixes:
- Preserve
errno across free, free_sized, and free_aligned_sized,
and across process_madvise-based page purging. (@spredolac: 3a77966,
86f0582)
- Fix numeric overflow checks in size classes. (@spredolac: 6b24522)
- Accept NULL in
free_sized() and free_aligned_sized() (C23
correctness). (@bigbruno: 7ce8b91)
- Fix TSD lifecycle edge cases by 1) initializing thread-cache bins before
marking the cache enabled, preventing reentrant bootstrap allocations
from using uninitialized state, and 2) avoiding TSD recreation for late
deallocations after thread teardown on generic-TSD platforms.
(@fzakaria: 54f22c8, @spredolac: fb5499a)
- Use
O_CLOEXEC when opening the THP sysfs file in init_thp_state.
(@ibookstein: d070554)
- Fix duplicate
opt.stats_print and opt.stats_print_opts fields in
malloc_stats_print output. (@spredolac: dfe3a2e)
- Fix a potential deadlock during
arena_reset. (@guangli-dai: 6957341)
- Fix a prof-sampling / guard-page interaction bug in the SAN. (@gctony:
e36a0fa)
Optimizations and refactors:
- Modularize jemalloc's front end by extracting arena management,
initialization, fork orchestration, and allocation dispatch from
jemalloc.c; untangle tcache/arena ownership; and consolidate the
internal header graph to eliminate circular dependencies. (@spredolac:
ba1e2fe, 8874597, 9d75722, de9ad14, ...)
- Simplify ctl dispatch by refactoring arena helpers, internalizing an
implementation-only control, replacing control-flow macros with typed
helpers, and organizing ctl.c by subsystem. (@spredolac: 9e14346,
901365d, 4e903a0, 5bc8d6e)
- Cap the base-block growth heuristic to avoid virtual-memory exhaustion
under rare racy conditions. (@guangli-dai: 2f4db8c)
- Move background-thread lifecycle and state operations into the
background-thread module, clarifying ownership independently of PAC/HPA
callers. (@guangli-dai: f1f0792)
- Simplify the page-allocation boundary by replacing PAI vtable dispatch
with direct PAC/HPA calls, removing the obsolete pai_t/pai.h
abstraction, and moving deferred-work and decay orchestration out of the
arena. (@guangli-dai: 1dfa6f7, 8edd101, d410f43)
- Refactor statistics collection and rendering into separate
gather/emission stages and descriptor-driven tables. (@spredolac:
184f304, 80c8fcb)
- Introduce an OS abstraction layer and move platform-dependent
file/process I/O, time, synchronization, CPU, virtual-memory, atfork,
error-handling, profiling, thread-yield, and configuration-access
operations out of allocator core code. (@guangli-dai: c4158ac,
c8e2e01, ...)
Portability improvements:
- Replace the
std::__throw_bad_alloc call with standard C++ (#2900).
(@lexprfuncall: 1a15fe3)
- Make
arena_s use a flexible array member (bin_t all_bins[]) for C99
or newer. (@grueninger: 300b58b)
- Fix rdtscp detection with
--with-lg-vaddr. (@xinydev: e8a0d2b)
- Fix
malloc_getcpu on macOS to read the current CPU number correctly.
(@Algunenano: 5aabbc8)
- Use
CLOCK_MONOTONIC for background-thread sleep to prevent
clock-rollback stalls, and detect monotonic-condvar support at configure
time. (@antonio2368: ebacec3, 8361239)
- Fix compilation warnings on macOS. (@gctony: abb0a8a)
- Fix thread-exit TSD cleanup on MinGW builds. (@gctony: 1e92317)
- Fix GCC 16 build warnings by removing
-Wpedantic violations in macro
and function syntax and explicitly NUL-terminating profiling thread-name
copies to resolve -Wstringop-truncation. (@grueninger: 5acdcee,
c22d929, @jasonangelov: 6a245e0)
- Parse PID-namespace symlinks without glibc-dependent
strtok/atol,
and return identifiers as uint64_t. (@guangli-dai: 278d90a)