- Core O(1) allocator (SmallFloat binning, two-level bitfield)
- Automatic neighbor coalescing on free
- Thread-safe
SyncAllocatorwrapper - Zero heap allocations on hot path
- Enterprise CI (3 OS, lint, formatting, fuzz, Codecov)
- 22 tests, 6 benchmarks, fuzz testing
- FEAT-001:
AllocateAligned(size, alignment)— per-allocation alignment (ADR-001) - Integration with
wgpuVulkan memory pools (replace BuddyAllocator) - Integration with
wgpuDX12 descriptor heaps - Extended statistics (per-bin occupancy, fragmentation metrics)
-
AllocationInfowith bin index and actual allocated size
- Trim approach for aligned allocation (zero-waste prefix reclaim)
- Benchmark suite vs BuddyAllocator (side-by-side comparison)
- Memory layout optimization (cache-friendly node pool)
-
usedBins [8]uint32variant (fewer type casts on hot path)
- API freeze
- Production benchmarks from wgpu + Project Neon consumers
- awesome-go listing