Robert Allan JamesandClaude Sonnet 5
d8a195b8d8
Found live during identity-heap-capacity testing (2026-09-07, hotplugging Zuse + 8 identities one at a time and measuring the kernel heap arena via a temporary allocator-stats probe, since reverted): every WIREBIND identity attach births two VMs in sequence -- a "console" VM, then the real "user" VM. When the second birth failed (arena fragmentation under concurrent VM load, a separate, not-yet-fixed capacity issue), capsule_wirebind_try_attach() logged the failure and returned, but the console VM that had *already succeeded* was never torn down. It stays live and registered under the identity's username, consuming its own ~228KB of the fixed 4MB kernel heap arena forever -- nothing ever points a real user at it, since WIREBIND only ever hands the caller the user VM's id. This turns every failed identity attach into a permanent net loss of heap rather than a neutral retry: confirmed live that a failed attach left the arena 228,576 bytes worse off than before the attempt, and every subsequent attempt starts from that worse baseline, compounding. Fix: call capsule_vm_kill(username) on the now-orphaned console VM before returning from the failure path -- the same teardown capsule_wirebind_eject()/capsule_wirebind_unclean_detach() already use elsewhere in this file (vm_cleanup() + sf_free(), confirmed live to actually reclaim per-word dictionary allocations, FABRIC-3.md §IX.2/§IX.3). Verified live with the same allocator-stats probe (written, captured, reverted -- not part of this commit): after the fix, a forced user-VM birth failure now returns the arena to exactly its pre-attempt byte count (3,526,256, matching the baseline precisely) instead of leaking 228,576 bytes. Three-arch clean qemu acceptance (single Zuse device, the standard regression case) passed on amd64, aarch64, and riscv64. The underlying capacity/fragmentation question (why the 7th concurrent identity's arena allocation fails at all despite technically-sufficient free bytes) is a separate, open architecture question -- not addressed here. See project memory for the full measured numbers. Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_014Ec88YKxxhZGG1RNnune78
src/
Hosted StarForth VM implementation (compiled by the root Makefile).
Bare-metal kernel sources live in src/starkernel/; FORTH word
implementations in src/word_source/; the test harness in
src/test_runner/; platform shims in src/platform/.
Entry point / interpreter core
main.c— CLI entry point, VM init, DoE mode dispatch.vm.c— interpreter loop, stacks, dictionary state (the central runtime file).vm_api.c— external VM API implementation.vm_bootstrap.c— VM bootstrap initialization.vm_debug.c— debugging utilities.vm_time.c— time-related VM operations.vm_internal.h— internal-only declarations shared across thevm_*.cfiles, not part of the publicinclude/vm_api.hsurface.repl.c— REPL read-eval-print loop.cli.c— CLI argument parsing.io.c— I/O operations.log.c— logging infrastructure.
Memory / dictionary / blocks
memory_management.c— dictionary allocator.dictionary_management.c— dictionary allocation and search.dictionary_heat_optimization.c— Loop #1 execution-heat tracking.word_registry.c— word registration system.block_subsystem.c— logical→physical block mapper.blkio_file.c,blkio_ram.c,blkio_factory.c— block I/O backends (file-backed, RAM-backed) and the factory that selects between them.stack_management.c— stack operations.
Physics-driven adaptive runtime (7 feedback loops)
physics_runtime.c— main physics coordinator.physics_hotwords_cache.c— Loop #1 hot-words caching.physics_metadata.c— per-word metadata tracking.physics_pipelining_metrics.c— Loop #4 word-transition prediction.physics_execution_hooks.c— execution instrumentation.rolling_window_of_truth.c— Loop #2 circular execution-history buffer.inference_engine.c— Loops #5/#6, statistical inference (window-width, decay-slope).ssm_jacquard.c— L8 Jacquard steady-state mode selector; consumescompudynamics.cfor tuning-word/config lookups.compudynamics.c— generic compudynamics module (cd_tuning_word(),cd_tuning_vm()); the score/UCB/reward/weight constants for the L8 adaptive table live here, not inssm_jacquard.c.heartbeat_export.c— heartbeat metrics export (CSV export function itself not yet implemented — seedocs/working/architecture/heartbeat_csv_export.md).
Math / measurement
math_portable.c— portable math functions.profiler.c— performance profiling.doe_metrics.c— Design of Experiments metrics (2^7 factorial).
Any .bak file alongside a .c file here (doe_metrics.c.bak,
inference_engine.c.bak, vm.c.bak) is a pre-edit backup left by a past
maintenance script (see scripts/remove_loop_conditionals.sh), not a
build input.