Files
aevolution/docs/engineering/VALIDATION.md
T
dingoandClaude Opus 5.5 df0c4f4efb Add M8.2 environment isolation and World0 protection
Per-environment platform layers (dev, stage, world0) with restricted pod
security, quotas, limit ranges, default-deny networking and token-less
service accounts, applied by an administrator rather than Argo. A native
ValidatingAdmissionPolicy denies deleting World0 data, identity and Argo
objects to every requester without the break-glass group. Stage and
World0 Argo projects are scoped to one namespace, World0 blocks
automated sync, and a cleanup tool refuses World0. An offline policy
check runs as a new CI suite; a live verification script uses only
can-i, reads and server dry runs.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-09-29 12:43:58 -04:00

2666 lines
162 KiB
Markdown
Raw Blame History

This file contains ambiguous Unicode characters
This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.
# Validation report
Recorded 2026-09-19; subsequent gate evidence is appended by date below. Latest
historical verified Linux delivery at the time of the foundation report was **run
1402** on `4c2cc04ac47d37c6191a9ef02d4bbba716a930dc`. Native check history and later
gate-specific results are kept separate. Results distinguish executed checks from
pending Linux/cluster gates.
## M5.7 local publication - 2026-09-26
Added the single-writer local filesystem adapter and manifest v1 under ADR-0048.
The implementation is in `include/evo/world/local_files.hpp` and
`src/world/local_files.cpp`; `schemas/local-world-manifest.md` defines the durable
bytes. The commit point is replacement of the inactive manifest slot after the
referenced payload has been written, flushed, installed and directory-flushed.
Recovery validates the snapshot and optional journal before returning a new store.
FNV is only a corruption check; this is not host-loss durability or the M7 region
ownership system.
Executed on Windows x64 / MSVC 19.44.35207:
- `tools/windows-check.ps1 -Preset release -Action check`: configure/build passed;
47/47 CTests passed with `NDEBUG`.
- `tools/windows-check.ps1 -Preset analysis -Action check`: clang-tidy 19 and
warnings-as-errors build passed; 47/47 CTests passed.
- `tools/windows-check.ps1 -Preset asan -Action check`: build passed; 47/47 CTests
passed. The Windows preset is not Linux ASan+UBSan evidence.
- `tools/windows-check.ps1 -Preset release -Action format`: passed.
- `git diff --check`: passed before commit.
New coverage validates checkpoint/journal/recovery, exact compaction over a published
journal, divergent-state rejection, exclusive writer lock, injected publication
boundaries, partial-write/out-of-space outcomes, corrupt manifest and missing payload.
Before-manifest failures recover the previous state; after replacement, reopen
recovers the new state. Actual power-cut/disk-full behavior, POSIX syscall path,
Linux CI/libFuzzer, and filesystem-controller flush guarantees remain unverified
locally. There is no automatic orphan cleanup, production storage adapter,
Kubernetes deployment or World0 change.
### M5.7 review corrections
The follow-up review found that command high-water alone allowed a divergent
checkpoint to replace an active journal. Checkpoint now compares its canonical bytes
with the fully recovered journal result before compaction. Poisoned handles reject
recovery; reopen resolves ambiguous post-replacement directory-sync failures.
Interruption tests now use a different edited payload and compare recovered canonical
state and debris across the commit point. Deterministic partial-write/out-of-space
injection covers payload and manifest temporaries, including retry, and a directory
sync failure hook verifies poison/reopen behavior. The injection models a partial
write outcome; it is not a physical disk-full or power-cut test.
Revalidated after these corrections on Windows x64 / MSVC 19.44.35207:
- Release configure/build and 47/47 CTests passed.
- clang-tidy analysis build and 47/47 CTests passed.
- ASan build and 47/47 CTests passed.
- clang-format and `git diff --check` passed.
## Tools and host
- Windows NT 10.0.26200, AMD Ryzen 7 9800X3D, 8 cores / 16 logical processors.
- MSVC 19.44.35217.0, CMake 3.31.6-msvc6, Ninja 1.12.1.
- clang-format and clang-tidy 19.1.5 (Visual Studio LLVM tools).
- kubectl 1.31.4 / Kustomize 5.4.2 for offline manifest rendering.
## Executed native checks
| Command | Observed result |
|---|---|
| `./tools/windows-check.ps1 -Preset debug -Action check` | Configure/build passed; 5/5 CTests passed |
| `./tools/windows-check.ps1 -Preset release -Action check` | Configure/build passed; 5/5 CTests passed under NDEBUG |
| `./tools/windows-check.ps1 -Preset analysis -Action check` | clang-tidy/warnings-as-errors build passed; 5/5 CTests passed |
| `./tools/windows-check.ps1 -Preset asan -Action check` | MSVC AddressSanitizer build passed; 5/5 CTests passed; this is NOT UBSan |
| `./tools/windows-check.ps1 -Preset debug -Action format` | Passed with clang-format 19; repeated twice in one shell after helper fix |
| `./build/release/evo-sim --seed 123456 --entities 8 --ticks 1000000 --verify-replay` | Exit 0; replay_verified=true; hash fab22c2e8ae1c379 |
| `./build/release/evo-benchmark` | Exit 0; MSVC 1944 Release, seed 123456, 10000 entities, 1000 ticks, 0.0210666 seconds, hash 3bad6db1612cbd1 |
| Git Bash `bash -n tools/ci/linux-check.sh` and `bash -n tools/ci/publish-image.sh` | Syntax checks passed; no Linux execution implied |
| `kubectl kustomize deploy/kubernetes/dev` | Rendered Job, ServiceAccount and NetworkPolicy in evolution-dev; final render pins the verified run-1402 image |
CTest covers the independently frozen 1000-tick fingerprint d61ad5bf0c928c76,
seeded replay properties, malformed/semantic snapshot rejection, entity lifecycle,
capacity/overflow, telemetry and CLI file continuation/alias protection. The
expected-failure harness confirms tests remain active in Release.
The benchmark is a single uncontrolled wall-clock observation, not a performance
threshold or an MMO capacity estimate. Its initial compiler output accidentally
used hexadecimal; the stream now returns to decimal and reports MSVC 1944.
## Gitea evidence
The maintainer's [run 1399](https://git.aeternalisgames.com/dingo/AEvolution/actions/runs/1399)
on `8ab48d3a5639121693f5881b600482e870b49f05` completed successfully. API job records
confirm verify (configure/build/CTest) and publish (Docker build/push) succeeded
on runner theraborn-runners-runner-0, label ubuntu-22.04. This is evidence for the
original workflow, not the expanded sanitizer/analysis jobs added in this change.
Exact Linux tool versions and expanded-run results appear below; a runner label
is not a compiler version.
## Limitations and next gates
- Local Docker Desktop has no running Linux engine. Container checks must run on
the configured Gitea runner; no local container success is claimed.
- Expanded Linux GCC14/Clang19, ASan/UBSan, TSan and libFuzzer evidence is recorded
below. TSan currently validates a sequential kernel, not future concurrency.
- Sandbox process restrictions stalled native Ninja and blocked Git Bash signal
pipes. Those attempts were stopped; checks reran successfully outside sandbox.
- Repeated VsDevCmd initialization grew PATH beyond cmd's limit. The helper now
reuses the matching x64 environment and deduplicates PATH; repeated checks pass.
- The maintainer reports Argo setup, but this workspace has no Kubernetes context.
No dev Job completion or cluster admission result is recorded yet.
- No biology, network/load, durable server recovery or backup restore exists.
Those remain M2, M7/M8 and later gates, not failing or passing foundation tests.
Update this report with run URL, tested commit, image digest and Job result when
those checks execute. Never overwrite a golden merely to make CI pass.
## Expanded Linux run 1400
[Run 1400](https://git.aeternalisgames.com/dingo/AEvolution/actions/runs/1400),
commit 9e412736d91ad103b7a951f99b811c629c5ac61e: GCC14 Debug and Release, Clang19
analysis and ASan/UBSan each passed 5/5 tests. Formatting passed. The seeded fuzz
smoke completed 10000 cases with no reported failure. CMake 3.28.3, GCC 14.2.0,
Clang/format/tidy 19.1.1, Linux 6.8.0-110-generic x86-64. Million-tick hash matched
Windows (fab22c2e8ae1c379); benchmark hash matched too (3bad6db1612cbd1), with
0.0674131 seconds observed for 10000 entities * 1000 ticks on the CI host.
TSan compiled but all executions failed before main: its runtime could not call
personality(ADDR_NO_RANDOMIZE) inside the Docker build sandbox. The image job was
skipped, as required. This is not recorded as a TSan pass. The follow-up isolates
execution in a container with the narrowly constrained profile from ADR-0020;
its result must be recorded after execution.
## Successful delivery run 1402
[Run 1402](https://git.aeternalisgames.com/dingo/AEvolution/actions/runs/1402),
commit 4c2cc04ac47d37c6191a9ef02d4bbba716a930dc: every job succeeded. The general
checks reused Docker's valid cache for unchanged checked sources; their full
execution also passed in run 1401. TSan executed all five CTests successfully in
2.71 seconds using the scoped profile and writable tmpfs outputs. Run 1401 had
first exposed a non-root tmpfs permission error; setting explicit mode=1777 on
these isolated scratch mounts fixed it. No test was disabled or ignored.
The runtime image built, ran the restricted replay smoke test and was pushed to:
`registry.aeternalisgames.com/aeternalis/aevolution:sha-4c2cc04ac47d37c6191a9ef02d4bbba716a930dc-run-6-1`
Registry digest:
`sha256:8fce2b462c9907f38d3fb832afeb70d0bff068dd832b2171e861186dbcb8e6b9`.
The dev manifest pins this digest. This proves build/test/image delivery; Argo
application mapping and Kubernetes Job completion remain unverified locally.
## M1.5 native ECS experiment
Opt-in target evo-ecs-experiment (EnTT4 pinned in ADR-0021), based on 1c6e6e1 plus
this experiment's changes. MSVC Release contract checks passed, including 72
seed/density/count combinations, hand-calculated arithmetic, canonical equality,
restore continuation, retired-ID rejection, malformed snapshot checks and
allocator balance. The final clang-tidy build/test passed after tightening the
process-boundary diagnostic catch. MSVC ASan passed on the final source.
Baseline Release CTest remained 5/5
and format-check included the experiment. No simulation rule/golden changed.
Commands: configure release/analysis/asan with `-B build/ecs[-analysis|-asan]
-DEVO_BUILD_ECS_EXPERIMENT=ON`, build target evo-ecs-experiment, then
`ctest --test-dir <directory> -R ecs-experiment --output-on-failure`.
Seven measured samples plus warmup are retained in results/ecs-msvc1944.jsonl;
protocol, build observations and scope limits are in ECS_EVALUATION.md. The
Linux experiment CI job is pending its first pushed run; no Linux experiment
success is claimed yet.
## M1.5 Linux evidence
[Run 1404](https://git.aeternalisgames.com/dingo/AEvolution/actions/runs/1404),
source bf7d60b7cd174736d38aea42616064e55e0c5f05: every job passed. The new ci-ecs
stage explicitly prepared the pinned header/license, then built and ran the
experiment under GCC14 Release, Clang19 ASan/UBSan and Clang19 analysis. All three
contract invocations passed. Raw measured GCC samples are retained at
results/ecs-gcc142.jsonl; all 56 fingerprints match corresponding Windows samples.
No production rule or snapshot-format change occurred. This supersedes the
pending Linux experiment note above. The final follow-up edits only record
results and decision scope; production ECS adoption remains deferred.
## M1.6 sequential scheduler (2026-09-19)
Native commands: `tools/windows-check.ps1 -Preset release -Action configure`,
clang-format on changed C++ files, `cmake --build --preset release`,
`ctest --preset release`; then `tools/windows-check.ps1 -Preset analysis -Action check`,
`tools/windows-check.ps1 -Preset asan -Action check` and Release `-Action format`.
All three configurations passed 5/5 CTests; clang-tidy and format-check passed.
Tests execute under NDEBUG and cover six batch permutations across 32 seeds,
an independent direct-lifecycle reference, ordered receipts, restore continuation,
rejected conflicts and malformed admission, the 1,024-command boundary, full entity
capacity and tick/ID exhaustion. Golden `d61ad5bf0c928c76` remains unchanged.
The copy-and-commit implementation allocates before publishing world state and
statically checks noexcept publication/receipt moves. Allocation-failure injection
was not run. No scheduler throughput claim, concurrency validation, durable input
journal or Kubernetes execution is implied.
[Gitea run 1407](https://git.aeternalisgames.com/dingo/AEvolution/actions/runs/1407)
passed all four jobs for source `b620edfd6f8097172acf055402a24620cb078358`:
Linux checks, the ECS experiment, separate TSan, and image smoke/publication.
This closes the pending Linux validation for M1.6. The maintainer's independent
dev image pin from `20e609d` was preserved during rebase. Kustomize renders the
three expected dev resources successfully; no cluster execution is claimed.
## M1.7 spatial queries (2026-09-19)
Release and MSVC ASan passed 5/5 CTests. Tests cover periodic-image reference
comparison (32 seeds × 3 lifecycle/restore phases × 24 queries), hand-checked
seams/corners/inclusive radii, empty/full capacity, invalid centers and maximum
u32 radius. Queries leave snapshots unchanged; existing v1 goldens remain intact.
Native commands use `tools/windows-check.ps1 -Preset <release|analysis|asan>
-Action check` and Release `-Action format`. Initial analysis flagged an unnecessary
test-helper vector copy; the helper now takes a const reference. Final analysis,
Release, ASan and formatting passed after applying the helper's required line
wrapping. Linux validation is recorded below.
`build/release/evo-spatial-benchmark.exe` produced 72 measured records plus
metadata in results/spatial-msvc1944.jsonl. One warmup and three samples per
scenario; 1k/10k/100k entities, 128 queries, uniform/clustered positions, four radii.
Timing includes result allocation, destruction and checksum work, excludes setup;
no performance threshold. Desktop CPU load was not isolated. No cluster changes
are required for this library increment.
[Gitea run 1409](https://git.aeternalisgames.com/dingo/AEvolution/actions/runs/1409)
passed all four jobs for source `9b19509673d88683ee72e093e42534bfe151f928`:
Linux checks, ECS regression, separate TSan and image smoke/publication.
All 72 spatial benchmark checksums and match counts agree with Windows.
Raw Linux measurements are in results/spatial-gcc142.jsonl; 24 native scenarios
also have identical checksums across their three repetitions. No deployment
manifest, rules version, snapshot format or golden was changed.
## M2.1 finite inventory ledger (2026-09-19)
Native Release, analysis and MSVC ASan each passed 6/6 CTests; formatting passed.
Commands: `tools/windows-check.ps1 -Preset <release|analysis|asan> -Action check`
and Release `-Action format`. Initial clang-tidy rejected the Element enum's
oversized base type; changed to u8 and reran all configurations successfully.
The new independent ledger test checks typed energy/mass separation, finite-source
exhaustion, explicit heat transfer, intermediate overdraft, whole-transaction
rollback, retry/duplicate action semantics, invalid reservoir IDs, maximum batch
and reservoir counts, opening-total overflow and exact u64-limit transfers.
32 seeded runs × 31,250 transactions = one million transactions, each checked
against independent reference balances, a replay branch and recomputed per-channel
totals. This is transfer replay, not serialized recovery or an ecosystem run.
Final ledger tests took 0.63s Release and 7.62s ASan locally; not controlled
performance measurements. Existing particle goldens remain unchanged.
Allocation-failure injection, durable ledger recovery and cross-ledger ownership
are not tested or implemented. Rates and conversions are absent, so rounding
residuals cannot occur in this transfer-only scope.
[Gitea run 1411](https://git.aeternalisgames.com/dingo/AEvolution/actions/runs/1411),
source `8fd3591b0e37aaecf00a6410a5926745a031cb6f`, passed
the Linux checks, ECS regression and separate TSan jobs. Each main Linux
configuration passed 6/6 CTests. Ledger-suite times were 3.34s GCC Debug, 0.37s GCC
Release, 2.46s Clang analysis and 8.78s Clang ASan/UBSan; these are CI test
durations, not controlled performance benchmarks. The image smoke/publication
job also passed, completing all four jobs. No deployment image pin was changed.
## M2.2 finite producer process (2026-09-19)
Native Release, clang-tidy and MSVC ASan passed 7/7 CTests; format-check passed.
Commands: `tools/windows-check.ps1 -Preset <release|analysis|asan> -Action check`
and Release `-Action format`. The producer suite compares hand-calculated ticks
and 32 seeds × 1,000 ticks against a small-domain cumulative arithmetic reference,
an independent replay branch and per-tick energy/element conservation audits.
It covers invalid config/ticks with unchanged state, fractional capture/growth,
efficiency endpoints, finite source exhaustion, all four limiting elements,
zero capture, maximum u64 capture with 50% carried conversion, and maximum recipe
and growth-cost multiplication. Particle goldens remain unchanged.
Allocation-failure injection and tick-counter exhaustion were not exercised;
the latter has an explicit guard but no arbitrary-state restore API. Tests use
in-memory copies, not serialized recovery. No photosynthetic chemistry, metabolic
model, database integration, throughput guarantee or deployment is claimed.
[Gitea run 1413](https://git.aeternalisgames.com/dingo/AEvolution/actions/runs/1413)
passed all four jobs for source `af293b8b50cbb506a49dfff10c6cc92307e16aca`:
Linux checks, ECS regression, separate TSan and runtime image smoke/publication.
The initial push was delayed by a tool approval-review usage limit; retry succeeded
after the maintainer resumed work. No deployment image pin was changed.
## M2.3 shared feeding community (2026-09-20)
Native Release, clang-tidy and MSVC ASan each passed 8/8 CTests; formatting passed.
Commands: `tools/windows-check.ps1 -Preset <release|analysis|asan> -Action check`
and Release `-Action format`. The new suite checks hand-calculated capture/feeding,
producer-to-consumer and consumer-to-consumer predation, heat/waste outputs,
terminal actor rejection, full-tick rollback, competing resource requests,
duplicate sequence/consumer admission, per-channel intake caps, invalid actors,
config/capacity limits and u64 assimilation endpoints. 32 seeds × 200 ticks compare
reordered input batches, receipts, full balances and producer progress, with
independently recomputed energy and elemental totals. Standalone producer and
existing particle golden tests remain unchanged and passing.
No serialized recovery, allocation-failure injection, spatial reach/perception,
physical injury, corpse handling or persistent-world integration was tested or
implemented. These are co-located synthetic process fixtures.
[Gitea run 1415](https://git.aeternalisgames.com/dingo/AEvolution/actions/runs/1415)
passed all four jobs for source `b7dd466f57f385d6a8ffccfaa69f639077545587`:
Linux checks, ECS regression, separate TSan and image smoke/publication.
No deployment image pin was changed.
## M2.4 terminal inventory and decomposition (2026-09-20)
Native Release, analysis and MSVC ASan passed 8/8 CTests; formatting passed.
Commands: `tools/windows-check.ps1 -Preset <release|analysis|asan> -Action check`
and Release `-Action format`. Tests preserve corpse inventories on death, suppress
producer activity immediately, reject repeated/invalid terminal events and roll
back death with later invalid feeding. Sixteen seeds × 256 ticks compare a
closed-form depletion reference, reversed terminal-event order, per-tick budgets,
inventories and conservation. Additional cases cover shared corpse/waste limits,
mixed and zero channel caps, u64-maximum inventories, diagnostic rollback, and
returned nutrients funding producer structure on the next tick.
The existing feeding, standalone producer, ledger and particle tests pass unchanged.
This does not validate reaction chemistry, actual death causes, corpse scavenging,
serialized recovery, allocation-failure injection or persistent history. There is
no deployment change.
[Gitea run 1417](https://git.aeternalisgames.com/dingo/AEvolution/actions/runs/1417)
passed all four jobs for source `e17a9a7279d6bde736261c0c33abe46108e1f622`:
Linux checks, ECS regression, separate TSan and image smoke/publication.
## M2.5 funded reproduction (2026-09-21)
Native Release, analysis and MSVC ASan passed 8/8 CTests; formatting passed.
Commands: `tools/windows-check.ps1 -Preset <release|analysis|asan> -Action check`
and debug `-Action format`. The first format run rejected three wrapped test
declarations; clang-format 19.1.5 rewrote them and Release was rebuilt and rerun.
Tests cover hand-calculated endowment and construction overhead, the endowment
floor, per-channel insufficiency with abundant energy, zero and oversized
investments, reserved slots rejected as feeding targets, corpses and ancestry
subjects, the producer rejected as a parent, terminated and unknown parents,
zero/duplicate sequences, two intents naming one parent, oversized batches,
capacity exhaustion rejecting the whole tick, absent identity reuse after a child
dies and decomposes, ancestry surviving both deaths, and construction endpoints of
0 and 1000 including a paid-for zero-endowment child. 32 seeds × 4 ticks × 2 births
compare reversed intent order against balances, producer progress, birth receipts
and independently recomputed conservation. A u64-maximum investment endows exactly
and conserves. Default zero capacity leaves the M2.4 reservoir layout, feeding,
producer, ledger and particle tests unchanged; golden `d61ad5bf0c928c76` is intact.
Review of `c84a701` found one defect (2026-09-22): births validated each parent
against the progressively modified candidate, so a child allocated earlier in the
same batch was already active and could fund a second birth. Two generations could
therefore appear in one tick, which ADR-0028 forbids. Eligibility is now fixed on
entry to the reproduction phase, after terminal facts and feeding, and checked for
every intent before any birth is applied. The contract itself was already correct
and did not change. The regression test admits one intent whose parent is created
by the preceding intent and requires `unborn_actor` with the entire tick unchanged;
it was confirmed to fail against the unguarded implementation and to pass with the
guard. The same pair succeeds across two ticks, recording the grandchild ancestry.
All four native configurations were rerun after the fix.
Allocation-failure injection, serialized recovery of states/parents/origins and
long-run ecosystem behavior were not exercised. This fixture defines no
reproductive strategy, fecundity, mutation, gestation, spatial placement or account
ownership, and no deployment manifest or image pin changed. Linux validation is
pending its first pushed run; no Gitea result is claimed for this increment yet.
## M2.6 measured ecosystem episodes (2026-09-22)
Native Release, analysis and MSVC ASan each passed 11/11 CTests; formatting passed.
Commands: `tools/windows-check.ps1 -Preset <release|analysis|asan> -Action check`
and debug `-Action format`. Formatting rejected 25 wrapped declarations on the
first run; clang-format 19.1.5 rewrote them and Release was rebuilt and rerun.
Three CTest cases are new: `ecosystem-contracts`, `cli-ecosystem` pinned to
fingerprint `2cd4a15f9fbe9949`, and `cli-ecosystem-conflict`, which requires the
CLI to reject `--ecosystem` combined with proof-run options.
The suite checks a solitary founder that only ages, a hand-checked predation
retirement, births exhausting exactly the reserved slots, tick-limit termination
with survivors, configuration rejection, and two episodes interleaved without
observing each other. Across 64 seeds it steps tick by tick, requiring that the
driver never provokes a rejected batch and that conserved totals are identical at
every tick, then replays each seed and requires an identical report. That ensemble
advanced 22,804 ticks with 879 births, 730 predation deaths, 277 lifespan deaths
and extinction in all 64 episodes, episode length 73 to 463 ticks.
`build/release/evo-sim.exe --ecosystem --seed 0 --episodes 3000` advanced
**1,093,007 ticks** in 1.34 s, with 41,138 births, 33,984 predation deaths, 13,154
lifespan deaths, extinction in all 3,000 episodes, episode length 50 to 497 ticks
and fingerprint `130d7725c0dc8b16`. Deaths equal founders plus births exactly
(6,000 + 41,138 = 47,138), so every process died once and no producer was grazed
out: the producer always retains captured energy, so its whole inventory never
reaches zero. The elapsed time is one uncontrolled wall-clock observation, not a
performance threshold.
Review of `eaf58f2` found one defect (2026-09-22): the CLI validated mode options
in one direction only. `--ticks` was accepted and ignored under `--ecosystem`, and
`--episodes`/`--tick-limit` were accepted and ignored without it, so a measurement
script could exit zero while running the other workload. The parser now records
which options were supplied and rejects any that do not belong to the selected
mode, in both directions. Six CLI cases cover this; `cli-ecosystem-rejects-ticks`
and `cli-episodes-require-ecosystem` are the two genuine regressions, while the
entities, save and verify-replay cases were already rejected and add coverage.
The remaining shared options were audited: `--seed` and `--output` are valid in
both modes, and `--load` conflicts were already enforced. CTest is now 16 cases,
all four native configurations were rerun, and `2cd4a15f9fbe9949` is unchanged.
This measures arithmetic conservation and turnover, not ecological plausibility.
Strategy constants, lifespans and the targeting rule are chosen inputs. There is
no maintenance cost, starvation rule, perception, spatial gate, genome or
heritable variation, so lineages differ only by founder strategy. Episode state is
not serialized, allocation failure is not injected, and no population abstraction
exists. Golden `d61ad5bf0c928c76` and every earlier suite are unchanged. No
deployment manifest or image pin changed. Linux validation is pending its first
pushed run; no Gitea result or cross-platform fingerprint agreement is claimed yet.
## M3.1 immutable genome record (2026-09-22)
Native Release, analysis and MSVC ASan each passed 17/17 CTests; formatting
passed. Commands: `tools/windows-check.ps1 -Preset <release|analysis|asan>
-Action check` and debug `-Action format`. Formatting rejected three wrapped
declarations on the first run; clang-format 19.1.5 rewrote them and Release was
rebuilt and rerun. `genome-contracts` is the one new CTest case, covering a new
`evo::biology` library that nothing else links.
Tests cover founder and descendant records, unordered drafts producing byte-equal
canonical records, digests differing for identical capabilities from different
origins, and every ancestry rule: zero identity, zero origin, zero development
version, a zero parent, a self parent, two parents and a repeated parent. Schema
rules cover an empty record, each of the three required capabilities removed in
turn, a duplicate kind, an unknown kind, an unknown prerequisite bit, and both
inclusive boundaries of all seven declared ranges accepted with one below and one
above each rejected. Graph rules cover a declared prerequisite naming an absent
capability, self, mutual and three-node cycles, a valid ordering using the same
edges without the closing one, and a cycle closed by a mandated rather than a
declared edge.
Encoding tests cover the round trip, byte-exact re-encoding, truncation at every
length from zero to one byte short, one trailing byte, an oversized buffer, a
corrupted magic, and every single-byte corruption of the full record. Three cases
reseal the checksum after editing so the later checks are reached rather than
masked: a rewritten genome ID is caught by the digest, a future schema version
fails closed, and non-canonical record order is caught as a duplicate kind. A
parent count beyond the reserved room and a capability count beyond the schema
are both refused before the corresponding region is read.
Coverage limitation: in schema 1 every mandated capability is also a required
one, so the presence half of the mandate check is reachable only through declared
prerequisites, which is tested. The mandated edges do take part in the cycle
closure, which is tested. A later schema that mandates a capability without
requiring it would exercise the remaining branch; this is recorded rather than
covered by a test that could not fail.
Review of `8584eb1` found two defects (2026-09-22). First, `Genome` had implicit
move operations, so moving one emptied the source's capabilities while leaving
its identity, origin and digest intact: a record that still reported ID 7 but
whose encoded bytes no longer decoded. That contradicted the documented invariant
directly. The type is now copy-only, which suppresses the implicit moves so a
move decays to a copy, and the invariant holds for every operation. The new move
tests were confirmed to fail without the fix: the Debug and ASan builds exit 1 at
runtime, and the Release build does not compile at all, because MSVC proves the
moved-from comparison always throws and rejects the unreachable code under
warnings-as-errors. clang-tidy then flagged the deliberate decaying moves in the
tests as `performance-move-const-arg`; those two lines carry a NOLINT explaining
that the decay is the contract under test.
Second, the encoding tests compared the current encoder against the current
decoder, so a coordinated change to layout or digest would have passed while
breaking existing v1 records. A frozen 83-byte fixture now holds one schema 1
founder record. Every structural byte was verified by hand against
schemas/genome-v1.md before being frozen: magic, both versions, ID 7, origin 1,
zero parents, three records, kinds 0/1/5 with parameters 400/120/5000, and the
`56 + 9N` length. The digest `9b2cdedef16ac87f` and the trailing checksum are
computed values frozen alongside them. The fixture is asserted in both
directions: the encoder must still produce those bytes, and the decoder must
still read them into the expected record. It is a golden and must not be edited
to make a test pass.
No genome affects any simulation, and nothing links the new library except its
own tests. The M2 process fixtures still duplicate founder strategies and do not
read the record. Development, phenotype derivation, mutation, recombination,
multi-parent heredity, migration between schema versions and any persistent
genome archive are not implemented. The encoding is a local record format, not a
network protocol or durable storage, and neither the digest nor the checksum is
an authenticity claim. Golden `d61ad5bf0c928c76` and the ecosystem fingerprint
`2cd4a15f9fbe9949` are unchanged. No deployment manifest or image pin changed.
Linux validation is pending its first pushed run; no Gitea result is claimed yet.
## M3.2 development and phenotype (2026-09-22)
Native Release, analysis and MSVC ASan each passed 18/18 CTests; formatting
passed. Commands: `tools/windows-check.ps1 -Preset <release|analysis|asan>
-Action check` and debug `-Action format`. `development-contracts` is the one new
CTest case, in the existing `evo::biology` library, which still nothing else
links. No existing suite changed.
The fixture genome uses a target size of 8,000,000 fg, which is an exact 20
micrometre cube at the declared density, so every derived value is hand-calculable
and is asserted as a literal: structural mass 8,000,000, extent 20, surface 2,400,
membrane mass 28,800,000, total 36,800,000 and maintenance 160,000 picojoules per
tick. Half-built, damaged and maximum-size cases are hand-calculated the same way,
including the 1e9 fg body that is an exact 100 micrometre cube with an uptake rate
of exactly the reference surface, and the 999,999,999 fg body one cubic micrometre
short of it, whose extent is 99.
Determinism is checked over 45 growth and temperature combinations, each derived
twice and compared, with more growth never yielding less structure. Damage is
checked to change the phenotype while leaving the genome's encoded bytes and
content digest identical, which is the M3.2 risk stated in the roadmap. Capability
prerequisites are checked both ways: without the capability a component is zero at
any size, and with it uptake scales with surface while movement speed does not.
Thermal shortfall is checked inside tolerance, symmetrically outside it, and
without the capability, and is confirmed to change no other derived value.
Unsupported rule versions and out-of-range growth fail closed through both entry
points.
Plans are checked for the full funded step, for the exact largest affordable step
one femtogram short of it, for no growth with no material, for a finished organism,
for clamping the last step to the target rather than overshooting, and for never
repairing damage. The declared flooring is pinned rather than hidden: a body whose
mass floors to zero advances its growth counter for no material, which grants no
structure because none exists at that size either way.
Coverage limitation: the overflow guards are unreachable through the public API at
schema 1 ranges, because the largest admissible genome produces a total mass of
about 7.2e11 femtograms, eleven orders of magnitude below the u64 limit. They are
kept as guards for later schema ranges and are recorded here rather than covered
by a test that could not fail.
No organism develops, nothing links the library except its own tests, and no
inventory moves: a plan reserves nothing and pays for nothing. The transfer half
of ApplyDevelopment, energetic budgets, repair, ageing, starvation, mutation and
morphology are not implemented. Golden `d61ad5bf0c928c76`, ecosystem fingerprint
`2cd4a15f9fbe9949` and the frozen genome fixture are unchanged. No deployment
manifest or image pin changed. Linux validation is pending its first pushed run.
## M3.3 metabolic budgets (2026-09-23)
Native Release, analysis and MSVC ASan each passed 19/19 CTests; formatting
passed. Commands: `tools/windows-check.ps1 -Preset <release|analysis|asan>
-Action check` and debug `-Action format`. `metabolism-contracts` is the one new
CTest case. `evo_biology` now links `evo::accounting` and `evo_processes` links
`evo::biology`; no existing suite changed and no existing fixture reads genomes.
Every budget is asserted as a hand-calculated literal against the same 20
micrometre cube fixture as M3.2: body 24,160,000 / 2,640,000 / 3,280,000 /
6,720,000 femtograms across C/N/P/other, maintenance 160,000 and movement 184,000
picojoules per tick, intake 240,000 picojoules and 2400/720/240/1440 femtograms,
and a reproduction investment of 11,040,000 femtograms priced at 441,600
picojoules. Construction is checked to compose each compartment separately rather
than by one blended ratio, and to agree exactly with what `plan_development`
reports as material and energy.
The non-uniformity the increment exists to demonstrate is tested directly.
Doubling membrane thickness raises body matter to 65,600,000 femtograms,
maintenance to 304,000 picojoules and the movement cost, shifts the carbon share
upward because membrane is the carbon-rich compartment, and leaves intake limit,
speed, assimilation and structural mass identical; the mass and maintenance ratios
are asserted to differ, so no single multiplier could produce this. Growth is
tested the same way: uptake follows surface while maintenance follows mass, and
their ratios differ.
Mass splits are checked to conserve at four awkward target sizes and four growth
fractions, where every share floors and the last channel absorbs the remainder.
An absent capability is checked to yield a zero budget rather than a default, and
unknown rule versions and out-of-range growth fail closed.
The genome-to-ledger path is exercised end to end: a derived `ConsumerConfig`
drives a real `FeedingCommunity` tick, the ledger totals are identical before and
after, and a bite one femtogram past the derived cap is refused by the existing
M2.3 contract. The opening inventory is checked to be exactly the caller's.
Review of `be71d4f` found one defect (2026-09-23): construction composed the mass
difference rather than differencing the two composed bodies, which discards the
rounding the existing body already carries. At target mass 1001 growing from 10 to
60, it funded C25 N7 P2 other16: the right total, but nitrogen and phosphorus one
short each and other two over, so spending the budget could not produce the body
the same rules derive. Construction is now the per-channel difference between the
composed bodies, which funds C25 N8 P3 other14 for that case.
Because the remainder rule is not monotone per channel, that difference can be
negative on a channel, so construction became a transition with an intake and a
release rather than a single requirement: a body growing from 13 to 14 femtograms
acquires one carbon and one nitrogen while freeing one from other. The release is
a rounding artifact bounded by a few femtograms, not excretion.
A follow-up review found the affordability half of the same problem: a transition
could require more of an element than was available, because the check compared
net mass. A step growing a body from 13 to 14 femtograms nets one femtogram but
acquires a carbon and a nitrogen while freeing one of other, and released other
pays for neither. The environment now states availability per ledger channel
rather than as one pool, since a pool cannot say that carbon is plentiful while
nitrogen is not. `plan_development` keeps a total-mass bound, sound because net
never exceeds gross but not an authority, and `metabolic_budget` narrows it to the
largest step whose own intake fits channel by channel and reports the growth that
step funds. That revises ADR-0031 and is recorded there.
The affordability tests place a single femtogram in each channel in turn and
require that no step needing a different element is funded, that nothing is
acquired beyond what is present, and that the mass-only plan would have admitted
the step; a carbon and a nitrogen together then fund it exactly. Across six target
sizes by seven growth fractions by four stock levels, every reported intake is
checked to fit the availability channel by channel and never to exceed the step
the plan proposed. These tests were confirmed to fail with the per-channel check
removed.
A third review found the availability bound itself could wrap: summing four
unrestricted u64 channels with 2^62 in each reaches exactly 2^64, so the total
read as zero and both planning and budgeting refused growth that a billion
femtograms per channel funded. The sum now saturates, which is also the right
answer since no step can need more than u64. Regressions in both suites use that
exact 2^62 case and a single saturated channel, and require the same step a
billion per channel funds; both were confirmed to fail without saturation. The
equivalent helper in `metabolism.cpp` is applied only to inventories derived from
bounded phenotype masses, never to caller input, so it is left unsaturated rather
than given a branch no test could reach.
The invariant `held + intake - release == reached` is now asserted per channel for
the reviewer's exact case, for the releasing case, across six target sizes by
seven growth fractions by four material budgets, and across a forty-step walk from
nothing to full growth where the accumulated intake and release are checked
against the whole history. The net of every transition is also asserted to equal
what `plan_development` charges material for. These tests were confirmed to fail
against the previous implementation.
Coverage limitation: as in M3.2, the overflow guards are unreachable through the
public API at schema 1 ranges and are recorded rather than covered by a test that
could not fail. The small integer helpers are duplicated between `development.cpp`
and `metabolism.cpp` rather than shared through an internal header.
No organism spends any budget: nothing consumes maintenance, movement or
construction, and no starvation, death or repair rule exists. The constants are
chosen for legible magnitudes, not balance. Golden `d61ad5bf0c928c76`, ecosystem
fingerprint `2cd4a15f9fbe9949` and the frozen genome fixture are unchanged. No
deployment manifest or image pin changed. Linux validation is pending its first
pushed run; no Gitea result is claimed for any increment since `b01a824`.
## M3.4 seeded mutation candidates (2026-09-23)
Native Release, analysis and MSVC ASan each passed 20/20 CTests; formatting
passed. Commands: `tools/windows-check.ps1 -Preset <release|analysis|asan>
-Action check` and debug `-Action format`. `mutation-contracts` is the one new
CTest case, in the existing `evo::biology` library. No existing suite changed.
The eligible set for the founder fixture is pinned by hand rather than observed:
nine operators, derived from the schema by the rules that make them eligible.
Energy conversion and membrane can adjust either way, movement, ingestion,
reproduction and thermal tolerance can be acquired, target size can adjust up but
not down because its 9,999,999 step would fall below the minimum from 8,000,000,
and no required capability can be shed. Over 2,000 seeds the offers are exactly
that set and nothing else.
Uniformity is checked to within a fifth of each operator's exact share, about six
standard deviations for 8,000 offers across nine operators. That test was
confirmed to fail when the draw was replaced by taking the first eligible
operators, which is the shape a biased offering would take. The rejection
sampling that removes modulo bias is correct by construction rather than
test-verified: over nine bins with 64-bit draws the bias it removes is about one
part in 10^18 and no test at this scale could observe it. That is recorded rather
than claimed as covered.
Stability is checked both ways: the same context always returns the same
candidates, while a new event index, seed, stream or growth state returns a
different draw. Selection is checked to refuse an operator that was never offered,
a moved event index, a changed growth state, a tampered offer and a different
parent genome, each with its own error. Boundary operators are checked to be
absent rather than clamped at both ends of a range, and a capability another one
declares as a prerequisite is checked to be unsheddable while the dependent
capability itself remains sheddable. Every offered candidate is selected in turn
and its successor checked to validate, to record parent 7, and to survive a
decode; the reported maintenance and mass are checked to equal what
`metabolic_budget` charges that successor.
Review of `5497f4d` found one defect (2026-09-23): selection established a
stale context by comparing regenerated candidate lists, which cannot establish it.
Four of nine operators is only 126 distinct lists, so different contexts draw the
same list routinely, and an offer from event 1 was accepted under event 239. An
offer now records the parent's identity and digest and the context itself, and
selection compares those directly before anything else. Because they then
determine the draw, a list that still differs means the offer was altered, which
is now its own error rather than a context having moved.
The regression searches for a context whose draw collides with the offer's,
asserts that such a collision exists rather than assuming it, and requires that
selection still rejects it. It was confirmed to fail against the previous
implementation. Rewriting the offer's recorded context to name the colliding one
is checked to be accepted, and should be: those candidates genuinely are what that
context draws, so the result is a legitimate offer for it rather than a forgery.
Editing a candidate is checked to be caught as an altered offer.
No organism mutates and no fixture calls any of this. Candidate count, context
window, mutation cost as a game rule, inherited versus acquired expression, NPC
mutation and recombination remain open and are not settled here. Golden
`d61ad5bf0c928c76`, ecosystem fingerprint `2cd4a15f9fbe9949` and the frozen genome
fixture are unchanged. No deployment manifest or image pin changed. Linux
validation is pending; no Gitea result is claimed since `b01a824`.
## M3.5 lineage retention (2026-09-23)
Native Release, analysis and MSVC ASan each passed 21/21 CTests; formatting
passed. Commands: `tools/windows-check.ps1 -Preset <release|asan|analysis>
-Action check` and debug `-Action format`. `lineage-contracts` is the one new
CTest case, in the existing `evo::biology` library. No existing suite changed.
Retention is asserted as a property of the type, not only of its behaviour: four
concepts check that the archive offers no `erase`, `remove`, `clear` or `prune`,
with a std::map positive control so the checks are known to be live rather than
vacuously true. An earlier attempt used bare requires-expressions, which MSVC
diagnoses eagerly for a non-dependent type, and a std::vector control, which was
wrong because `vector::erase` takes an iterator rather than a key; the concept
correctly reported false and the control was corrected rather than the assertion
weakened.
A branching fixture of five genomes over three generations is used throughout.
Tests cover a genome outliving both organisms that carried it with its record,
ancestry, children and organism terminal ticks intact; ancestry and descendants
resolving through a genome whose only carrier died; common ancestors for
siblings, cousins, a genome against itself and two separate founders; refusal of
a duplicate identity, a child admitted before its parent and a full archive;
death being terminal against repeat, reversal and rebirth; and absence being
reported for every lineage that has no living carrier, including one that gains a
carrier again afterwards. The capacity case admits the full 4,096 genomes as one
chain, checks that the refusal leaves every earlier record intact, and walks the
4,096-deep ancestry.
The implementation was changed twice under clang-tidy rather than suppressed.
`std::map` members made the archive's move potentially throwing, because MSVC's
map move constructor allocates a sentinel node; sorted vectors replaced them,
which also matches the rest of the codebase. A wrapper struct holding a `Genome`
by value then had the same problem, because ADR-0030 made `Genome` copy-only, so
genomes are now stored directly and identified by their own `id()`. The redundant
per-genome carrier tally was dropped in the same change: carriers are counted from
the organism records, so there is no second tally that could drift.
Review of `8020347` found one defect (2026-09-23): the retention claim covered
methods named like removal and nothing else, so `archive = LineageArchive{}`
discarded every record, and assigning a copy taken before a death restored that
organism's living state. Both forms of assignment are now deleted, which also
makes the archive unswappable, and the copy constructor is declared so the
implicit moves are suppressed and a move decays to a copy. Copying stays available
because a clone removes nothing and GENOME_MODEL expects cloned-world rehearsals.
The interface tests now cover assignment, replacement from a temporary and
swapping as well as the named methods, each with a std::vector positive control,
and check at runtime that a copy taken before a death does not disturb the
original's terminal record and that a moved-from archive still holds everything.
All four static assertions were confirmed to fail with the deleted assignments
removed.
Linux CI on `29ce255` reported one build blocker (2026-09-23), the first Linux
result recorded since `b01a824`. GCC Release passed 21/21, then the Clang 19
analysis build failed: a bare `{1, 2, 3, 4, 5}` deduces `initializer_list<int>`
and the sign-changing conversion to each `std::uint64_t` element trips
`-Wconversion` under `-Werror`. The literals are now unsigned.
MSVC `/W4 /WX` does not diagnose this, and GCC's `-Wconversion` does not warn when
a constant value is known to fit, so neither the local checks nor the GCC job
could have caught it; only Clang did. Attempting to reproduce Clang's diagnostic
locally through clang-tidy failed, because the warning flags are not honoured in
its cl driver mode and no clang driver is installed on this host. The whole tree
was instead audited by hand for the pattern: every other range-for over a braced
literal list either deduces through `auto` or already carries suffixes matching
its declared type, and this was the only instance. That audit is a one-off, not a
check that will catch the next one; the standing gap is that no Clang or GCC
compile happens on this host before a push.
Nothing is serialized, so no history survives a process, and no fixture calls any
of this. Terminal extinction is not claimed anywhere, durable retention and
migration are M7 and M8, and NPC mutation remains undecided. Golden
`d61ad5bf0c928c76`, ecosystem fingerprint `2cd4a15f9fbe9949` and the frozen genome
fixture are unchanged. No deployment manifest or image pin changed. Linux
validation is pending; no Gitea result is claimed since `b01a824`.
## M3.6 genome ecology ensembles
Executed on this Windows host (2026-09-23): Release configure, build and `ctest`
33/33; the analysis preset (clang-tidy 19, warnings as errors) built clean; the
ASan preset configured, built and ran 33/33; `clang-format` 19.1.5 reported no
violations after the new files were formatted. All four checks ran and passed on
the committed tree. No Clang or GCC compile happens on this host, so the standing
gap recorded for `29ce255` is unchanged and Linux CI remains the first Clang and
GCC result for this increment.
`evo-ensemble-tests` covers eight properties: that a derived strategy carries
exactly ADR-0032's budget quantities and invents no margin, that divergence is
reproducible and refuses its bounds rather than truncating, that conservation
holds at every tick of every cell, that a run is reproducible and a different seed
is not, that the archive ends holding exactly the founders and every organism that
lived with no living carriers, that the two seed halves agree within the stated
band, that each gradient axis alone changes the outcome, and that a thermally
tolerant strategy keeps more scheduled life at the extremes. `evo-sim --ensemble`
is pinned by fingerprint `c2b96c86101be628`, confirmed stable across three runs
and distinct from a different seed's `8f88e72cf2715eac`.
Three defects were found and fixed during implementation, all by measurement
rather than by review. `EnsembleReport` and `CellReport` were default-initialized
rather than value-initialized: both are aggregates whose vector members give them
a non-trivial default constructor, which leaves the scalar counters indeterminate,
and the episode total read as garbage. Both are now `{}`.
The other two are findings about the model, and are recorded in ADR-0035 rather
than tuned away. The header claimed the resource level entered both the derived
strategies and the episode's nutrient pool. It does not enter the derivation at
all: founders are derived fully grown, so their construction transition is empty
and ADR-0031's per-channel `available` cannot reach intake, assimilation or
reproduction investment. The comment was corrected rather than the code, because
the code was right and the claim was false.
The resource axis was then measured to be decorative at the original constants and
non-monotonic once it was not. A ten-point sweep at the original fixture returned
41 to 45 organisms across three orders of magnitude of nutrients with identical
tick totals and eight extinctions everywhere: every birth came from the opening
endowment, because a body of 353,600,000 fg needed 73,776,000 fg of carbon to
reproduce while feeding supplied 8,000 fg per tick before assimilation, or roughly
23,000 ticks against a 60-tick lifespan. Rescaling so a birth can be earned
required pinning the investment fraction to the genome schema's floor of one per
thousand, because intake scales with surface and investment with body mass, so
the ticks needed to fund a birth fall only as the cube root of mass. That a
reproduction is affordable only at the schema floor is a finding about the
ADR-0031/0032 cost model and is now the second priority in STATUS.md.
After rescaling, a sweep of 200 to 200,000 femtograms gave 45, 45, 45, 51, 61, 81,
44, 44, 42 and 42 organisms: a peak near 8,000 and a fall away on both sides. The
mechanism is ADR-0026's feeding rule that a target may be bitten by only one
consumer per tick, which caps community income at one producer bite however large
the pool; scarcity empties the producer, consumers bite each other, and that
concentrates mass into individuals that reach the threshold, while abundance
spreads the same income too thin to fund a birth. The gradient test therefore
asserts only that an axis moves the outcome, never a direction. An earlier draft
of that test asserted `rich_organisms > lean_organisms`; that assertion held at
the two chosen cells but implied a monotonicity the sweep disproves, so it was
replaced rather than satisfied by picking two cells that flattered it.
Coverage limits. The grid is four cells and the tests use eight seeds per cell, so
every number describes this fixture and nothing wider. The split-half band
(`high * 2 <= low * 3 + 2`) is a chosen tolerance, not a derived one. No organism
mutates during an episode, so nothing here exercises in-episode divergence and NPC
mutation stays open. The thermal penalty lives in the harness and no contract
knows about it, so no contract test covers it. Nothing is serialized.
The eleven new CLI tests are new-behaviour tests for a new mode, not regressions
of a fixed defect; only the M2.6 bidirectional mode check they extend was ever a
regression. They were confirmed to fail the run, which is what `WILL_FAIL`
asserts, but each one was not separately confirmed to fail through the branch it
names, so a rejection arriving from a different branch would still pass. Golden `d61ad5bf0c928c76`, ecosystem fingerprint `2cd4a15f9fbe9949` and the
frozen genome fixture are unchanged, and ADR-0029's episode driver was not
touched. No deployment manifest or image pin changed. Linux validation is pending;
no Gitea result is claimed since `b01a824`.
Review of `89ad95b` found three defects (2026-09-24), each now fixed with a
regression confirmed to fail against the unfixed code.
The harness recorded a death for every organism still alive when an episode
stopped. A one-tick run therefore reported zero extinctions while writing terminal
death records for founders that had not died, and every such organism's watched
time was added to the same total as a completed lifetime. Ending observation is
not death, and ADR-0034 makes a recorded death terminal, so there was nothing for
the archive to be terminal about. Survivors now stay living, are counted in
`survivors`, and their watched time accumulates in `observed_ticks`, which is a
lower bound on a lifetime the run never saw end. `deaths` and `lived_ticks` cover
only organisms observed to die, and the two are never added together. The run and
archive test now checks that deaths plus survivors equals every organism recorded
and that a lineage is absent exactly when all its carriers were seen to die.
A draw that produced a genome without the reproduction capability derived a zero
reproduction investment, which ADR-0029's driver refused outright, so
`--ensemble --seed 10` exited with invalid_arguments; seeds 12 and 15 were
confirmed to fail the same way and seed 11 to succeed. Filtering such variants out
before sampling was rejected as the worse repair: the measured set would then have
been the fertile subset chosen by a guard, and an inability to reproduce is a
result rather than a reason for exclusion. The driver now accepts the strategy and
proposes no birth for it, since an empty investment would otherwise clear both
reproduction bounds trivially and request a birth transferring nothing. ADR-0029
is amended rather than silently contradicted, and ADR-0035's claim that the driver
does not change was corrected to say what did.
Only configurations the driver previously rejected outright behave differently, so
the M2.6 pin was expected to hold and was checked rather than assumed:
`--ecosystem --seed 0 --episodes 32` still reports `2cd4a15f9fbe9949`, and the
ecosystem, feeding and producer suites pass unchanged. The ecosystem test that
asserted a zero investment is rejected now asserts the opposite fact, that such an
episode runs and reports no births.
A half's thermal shortfall was assigned rather than accumulated, so a field
documented as a total across the half's episodes carried one episode's reading.
Two episodes at 15,000 millikelvin reported 15,000 rather than 30,000. It now
accumulates per episode. The tolerance test's expectations were per-episode
readings and are now stated against the episode count, with an assertion that the
count exceeds one so a sum and a single reading cannot look alike.
`cli-ensemble` was repinned from `c2b96c86101be628` to `5ec59d390909b526`,
confirmed stable across three runs and distinct from seed 7's `090b83efb85ae477`.
This is a change in reported facts, not a test accommodation: survivors are no
longer counted as deaths, the outcome record gained `deaths`, `survivors` and
`observed_ticks`, and the shortfall is now a sum, so the previous value describes
a report this build does not produce. The rationale is recorded beside the pin.
Each fix was confirmed to be load-bearing by reverting it alone and rebuilding.
Restoring the survivor-as-death behaviour failed the archive test at the survivor
count and, when that test was run second, the dedicated regression at `deaths ==
0`. Restoring the zero-investment rejection failed the sterile-strategy regression
at `EcosystemEpisode::create`. Restoring the assignment of thermal shortfall
failed the summation regression. One of my own expectations was wrong rather than
the code: a one-tick episode's survivors include organisms born during that tick,
watched for none of it, so observed time equals one tick per founder rather than
one per survivor. The test now says that.
Coverage limits unchanged from the increment above, with two additions. Censored
lifetimes mean a run's mean lifetime is not computable from `lived_ticks` alone
and no test asserts one. Across a run the archive now ends holding living
carriers, which is accurate about what was observed and is not a claim that
anything persists; nothing is serialized.
## M4.1 cohort state and ownership
Executed on this Windows host (2026-09-24): Release configure, build and `ctest`
34/34; the analysis preset (clang-tidy 19, warnings as errors) built clean; the
ASan preset configured, built and ran 34/34; `clang-format` 19.1.5 reported no
violations. All four checks ran and passed on the committed tree. No Clang or GCC
compile happens on this host, so Linux CI remains the first result for the new
`evo::ecology` library.
`evo-cohort-tests` covers six properties: that a cohort keeps exactly the fields
it was given, that totals equal the summed classes, that two cohorts with
identical totals but different structure are distinguishable, that every
construction invariant is rejected by its own named error, that one habitat and
genome holds one record, and that only the recorded owner under a current epoch
may replace it. A sixth checks that a cohort names a genome the archive resolves
ancestry for without storing that ancestry.
Every guard was confirmed load-bearing rather than assumed. Each of the nine was
removed in turn and the suite rebuilt and rerun: the duplicate, stale epoch, wrong
owner, unowned reserve, residual bound, class ordering, empty class and class
range guards each produced a failing run, and removing the overflow guard fails to
compile, because the running totals then have no reader and the Release build
treats that warning as an error. An earlier attempt at this sweep reported all
nine as not load-bearing; that result was false and was discarded. The harness
wrote to `/tmp`, which Git Bash and PowerShell resolve to different directories,
so the mutation script was never found and the source was never modified. The
sweep was rerun against an absolute path.
Two defects in my own test code were found and fixed before the suite was trusted.
The single-argument `CHECK` macro split on the comma inside a braced initializer
such as `Owner{11, 2}`, which is a compile error rather than a silent one, and the
macro is now variadic. More seriously, the rejection tests originally asserted on
`.error()` alone, which is undefined when the call unexpectedly succeeded and would
have let a removed guard pass quietly; they now check the failure first through a
`refused` helper. A regex rewrite applied to make that change spanned lines and
corrupted several positive assertions, so the test file was rewritten rather than
patched.
Coverage limits. Nothing here is serialized, so `schemas/population-v1.md` pins
field meaning and invariants and there is no encoder, no round trip and no
migration to test. The approximation bound is carried and never computed or
checked, so no test asserts that aggregation is faithful; that is M4.5.
Conservation between a cohort and the rest of the world is not asserted, because
this type holds an inventory without participating in any transfer. Ownership is
tested as a single-process invariant only: epochs make a stale writer fail locally
and nothing here exercises fencing between workers, which is M7 and M10. Negative
reserves are not representable at all, since every quantity is unsigned, so what
the suite rejects instead is a declared state that would require one, and two
static assertions record that the underlying types are unsigned. `max_cohorts` is
enforced but not exercised at its limit, since installing 4,096 records would add
nothing to what the duplicate and lookup tests already establish.
No existing contract changed. The proof golden `d61ad5bf0c928c76`, the ecosystem
fingerprint `2cd4a15f9fbe9949`, the ensemble fingerprint `5ec59d390909b526` and
the frozen genome fixture are all unchanged, and the other 33 tests pass
untouched. No deployment manifest or image pin changed. Linux validation is
pending; no Gitea result is claimed since `b01a824`.
Review of `1c8b292` found two defects (2026-09-24), both fixed with regressions
confirmed to fail against the unfixed code.
Replacement checked the ownership epoch and nothing that ordered two writes made
inside one. A record installed at tick 1,000 holding 50 picojoules could be
replaced by its own earlier tick-900 record holding 100, restoring inventory the
installed record had already spent, because within an epoch the two were
interchangeable. A cohort now carries a revision that strictly increases within an
epoch, and a repeated revision is refused as well as an older one, since a write
has to be strictly later rather than merely not earlier. A newer epoch may restart
the series, because re-acquiring ownership begins a new one. The interval boundary
is checked independently of both, so neither advancing a revision nor re-acquiring
ownership is a way to rewind simulated time. A later revision at the same boundary
stays legitimate: that is a correction, not a rewind.
The implicit move emptied a cohort's age classes while keeping its habitat, genome
and owner, so a moved-from record was the invalid cohort `create` refuses to build
and the registry accepted through a move what it rejects through construction. The
type is now copy-only, which suppresses the implicit moves so a move decays to a
copy. ADR-0030 made the genome copy-only for exactly this reason in M3.1; the
precedent existed and was not applied when this type was written.
Both fixes were confirmed load-bearing by reverting each alone. Removing the
revision check fails the new ordering regression, as does removing the backward
boundary check. Removing the copy declarations fails to compile in Release, and
not because of the static assertion: with that assertion also removed, MSVC proves
the code after `source.classes().size() == 3` unreachable, having established that
a moved-from vector is empty and the check therefore always throws. That is a
compile-time proof of the assertion rather than an execution of it, so the runtime
path was checked separately under the ASan preset, where the same mutation exits 1.
Coverage limits are unchanged from the increment above, with one addition. The
revision is a caller-supplied counter, not one the registry issues, so nothing
prevents a caller from choosing revisions that are unrelated to its own write
order; what is enforced is only that the registry refuses a record that is not
strictly later than the one it holds. Issuing revisions, and fencing between
workers, remain M7 and M10.
No existing contract changed. The proof golden `d61ad5bf0c928c76`, the ecosystem
fingerprint `2cd4a15f9fbe9949`, the ensemble fingerprint `5ec59d390909b526` and
the frozen genome fixture are unchanged. Linux validation is pending; no Gitea
result is claimed since `b01a824`.
## M4.2 collapse to cohort
Executed on this Windows host (2026-09-24): Release configure, build and `ctest`
35/35; the analysis preset (clang-tidy 19, warnings as errors) built clean; the
ASan preset configured, built and ran 35/35; `clang-format` 19.1.5 reported no
violations. All four checks ran and passed on the committed tree. No Clang or GCC
compile happens on this host, so Linux CI remains the first such result.
`evo-collapse-tests` covers seven properties: that a plan totals its members
exactly and moves each member's whole reserve, that ages and the per-class reserve
distribution survive the move without redistribution, that every ineligible state
and every malformed request refuses the whole plan by name, that a committed
collapse leaves the ledger's conserved total unchanged while emptying every member
reservoir and crediting the destination, that every rejection leaves both the
ledger and the registry exactly as they were, that one plan cannot be committed
twice, and that a moved-from plan is still whole.
All ten guards were confirmed load-bearing by removing each in turn and rebuilding:
the ineligible, foreign, duplicate, self-destination, age classification, stale
sample, occupied destination, empty class and commit ordering guards, and the
copy-only declaration. Two needed a second pass. The first age mutation did not
compile, which is not evidence about the test, so it was rewritten to make the
class predicate always match, which places every member in the first class and is
detected at runtime. Removing the copy declarations fails to compile in Release
for the reason recorded for the cohort; with the static assertion also removed the
same mutation was rerun under ASan, where it exits 1, so the runtime assertions
are load-bearing independently of the compile-time proof.
The commit ordering mutation is worth naming: moving the ledger transaction ahead
of the destination check is detected, because a rejected conversion then leaves
biomass in a reservoir nothing owns while the members are already empty. That is
the dual-representation failure this milestone is named after, and the ordering is
what prevents it rather than a comment saying it does not happen.
One expectation of my own was wrong rather than the code. A replay of a committed
plan was expected to report `destination_occupied`, but the staleness check ran
first and reported `stale_plan`, which is equally true. Both are pure checks made
before anything moves, so safety was unaffected and only the diagnosis differed;
the destination check now runs first, because "this conversion already happened"
says more about a replay than "the world moved". The ordering rationale is
recorded beside it.
`accounting::ReservoirId` and `accounting::Transfer` gained the defaulted
comparisons that every other value type in the project already carries, so a
plan's legs can be compared. This is additive, no behaviour depends on it, and
ADR-0037's status was corrected to say so rather than claim no M1 change.
Coverage limits. Completeness of the offered member set is not checkable here and
is not checked: nothing establishes that the caller offered every member it had.
Emptying a reservoir is not withdrawing the right to use it, since the ledger has
no removal, so a later transfer could credit an emptied reservoir and no test
asserts otherwise. The before-and-after property is demonstrated across rejections
inside one process; there is no persistence, so a crash between the ledger
transaction and the registry install is outside what an in-memory registry can
show, and durable conversion sequences, escrow and replay stay M7 and M8. Plans
are bounded to 256 members by the ledger's transaction limit and nothing
coordinates several conversions. The ledger-failure and destination-unavailable
branches are reachable only past bounds the tests do not build, so they are
refused-by-construction rather than exercised.
No existing behaviour changed. The proof golden `d61ad5bf0c928c76`, the ecosystem
fingerprint `2cd4a15f9fbe9949`, the ensemble fingerprint `5ec59d390909b526` and
the frozen genome fixture are unchanged, and the other 34 tests pass untouched.
No deployment manifest or image pin changed. Linux validation is pending; no Gitea
result is claimed since `b01a824`.
Review of `1a59b63` found one defect (2026-09-24), now fixed with a regression
confirmed to fail against the unfixed code.
Commit proved the destination reservoir existed and that the registry slot was
free, but never that the reservoir was empty. A conversion into an occupied
reservoir therefore succeeded and mixed inventory it had never sampled into the
cohort's backing store, leaving the cohort's recorded reserve understating what
backed it and the surplus owned by nobody. The reported case was reproduced
against the unfixed code rather than assumed: gathering into reservoir 5, which
the fixture already seeds with seven picojoules, commits successfully and leaves
the ledger holding 67 against a cohort recording 60.
A destination holding anything is now refused. The success test additionally
asserts that the destination balance equals the installed cohort's reserve, so a
cohort's reserve is the whole of its reservoir rather than a part of it, and the
new regression asserts that the refusal leaves the ledger balances, the ledger
action counter and the registry all exactly as they were, then that emptying the
reservoir first makes the same plan committable.
The check runs after the staleness loop rather than beside the other destination
checks. Both run before anything moves, so this is a reporting choice and not a
safety one: replaying a committed plan into a fresh registry would be refused by
either, and that the members no longer hold what was sampled is the more useful
answer than that the reservoir is occupied. The ordering rationale is recorded
beside the check.
The guard was confirmed load-bearing by removing it alone, which fails the new
regression. An earlier attempt to add that mutation reported the guard as not
load-bearing; the result was void and discarded, because a Bash heredoc had
mangled the escaped newlines in the mutation script and the script failed to
parse, so nothing was ever mutated. It is the second time this environment quirk
has produced a false negative in a mutation sweep, and both times the printed
output rather than the exit status is what revealed it.
## M4.3 interval accounting
Executed on this Windows host (2026-09-24): Release configure, build and `ctest`
36/36; the analysis preset (clang-tidy 19, warnings as errors) built clean; the
ASan preset configured, built and ran 36/36; `clang-format` 19.1.5 reported no
violations. All four checks ran and passed on the committed tree. No Clang or GCC
compile happens on this host, so Linux CI remains the first such result.
`evo-population-tests` covers ten properties, and `evo-cohort-tests` gained two
for the amendments. An interval with no flux advances the clock and submits no
ledger transaction. Births change the population and leave the reserve untouched.
Death, predation, immigration and emigration each move biomass once to the named
counterparty with the conserved total unchanged. A loss exceeding the population
or any one inventory channel is refused rather than clamped, including at the
boundary of a single millionth of a member, while taking exactly everything is
vacancy rather than refusal. A population below one member stays a population and
its record stays installed. A declared result that does not match the fluxes is
refused in both directions, for counts and for biomass. Empty and backward
intervals, a foreign owner and a repeated revision are refused. A commit against a
record that has moved on, or a reservoir that no longer holds what was counted,
changes nothing in either the ledger or the registry, and one plan cannot be
committed twice.
The seeded ensemble runs twenty-four seeds of forty proposed intervals each,
deliberately proposing losses unbounded relative to the cohort so that some must
be refused. Every accepted interval is checked for the conserved total being
unchanged and for the cohort's reservoir equalling its recorded reserve. The test
asserts that both the accepted and the refused path actually occurred, so a change
that made everything refuse, or everything pass, fails rather than looking clean.
All fourteen guards were confirmed load-bearing by removing each in turn and
rerunning: the two no-clamp guards, the two mismatch guards, the interval, owner,
revision and counterparty guards, the three commit staleness guards, and the three
amendments to the cohort record. The two clamp mutations are the ones worth
naming, because they are the defect rather than a missing check: replacing the
refusals with saturating subtraction, which is what a clamp is, is detected.
Two of my own expectations were wrong rather than the code. The escrow arithmetic
in the migration test was miscalculated: the reservoir receives two hundred matter
and returns one hundred, so from seven it stands at a hundred and seven, not a
hundred and ninety-seven. And the dwindling-population test shed all but one
femtogram of the reserve while declaring a result that held none, which is a real
`biomass_mismatch`; the test now sheds the whole reserve. That second one surfaced
a genuine tension worth recording: a class reduced to a fraction of a member may
hold no reserve, because ADR-0036 gives reserve to members and a fraction is not
one yet, so that biomass has to go somewhere in the same interval. It is recorded
in ADR-0038's costs rather than worked around.
ADR-0036 was amended twice, and both were forced by this milestone rather than
chosen. Threading the reservoir through each request instead of binding it to the
record would have let a caller advance one cohort against another's store, and a
balance comparison cannot distinguish two equal reserves, so the binding is a
correctness fix and not tidiness. Vacancy is required because a decline to exactly
zero was otherwise unrepresentable and the only way through would have been the
clamp this milestone exists to refuse. `Cohort::vacant` is a separate factory
precisely so that an empty class list in a draft stays an error.
Coverage limits. Closure is checked and stability is not: an interval whose
arithmetic balances and whose ecology is nonsense is accepted, and no test asserts
otherwise, because comparing matched ensembles is M4.5. The declared age structure
is not verified to be what ageing would produce, since that needs the within-class
distribution the record discards; nothing here checks that members did not get
younger. Emigration escrows into a reservoir nothing settles with. Ownership
remains a single-process invariant. The overflow and ledger-failure branches are
reachable only past bounds the tests do not build, so they are refused by
construction rather than exercised. Nothing is serialized.
The proof golden `d61ad5bf0c928c76`, the ecosystem fingerprint
`2cd4a15f9fbe9949`, the ensemble fingerprint `5ec59d390909b526` and the frozen
genome fixture are unchanged, and the other 33 tests pass untouched. No deployment
manifest or image pin changed. Linux validation is pending; no Gitea result is
claimed since `b01a824`.
## M4.4 materialization
Executed on this Windows host (2026-09-24): Release configure, build and `ctest`
37/37; the analysis preset (clang-tidy 19, warnings as errors) built clean; the
ASan preset configured, built and ran 37/37; `clang-format` 19.1.5 reported no
violations. All four checks ran and passed on the committed tree. No Clang or GCC
compile happens on this host, so Linux CI remains the first such result.
`evo-materialize-tests` covers ten properties: that a drawn member receives the
floor share while the rounding remainder stays with the class, that a class drawn
empty hands over everything it held, that neither a residual nor a member that is
not there can be drawn, that taking everyone leaves a vacant record rather than an
extinction, that ages come from the top of a class and are stable when regathered,
that a round trip is exact at one member per class, that a coarse class conserves
the total while smoothing individuals, that repeated oscillation never pays, that
malformed draws and stale views are refused without changing anything, and that a
moved-from plan is whole.
The oscillation test runs twelve collapse and materialize cycles over three
members of unequal reserve and asserts three things every cycle: the ledger's
conserved total never moves, no member ever holds more than the largest starting
member held, and no member is younger than the corresponding member of the
previous cycle. Those three together are what stops a round trip being a way to
heal or to reset age. It also ends by asserting every member has been pushed to a
class ceiling, which records the cost of the conservative age rule rather than
leaving it as a claim.
Eleven guards were confirmed load-bearing by removing each in turn and rerunning.
Two of them are the named risk rather than ordinary validation, and both are
detected: rounding the share up instead of down, which creates biomass, and
drawing ages from the bottom of the class instead of the top, which returns
members younger than they were and makes oscillation profitable.
The free-healing risk has two sources and only one is closed. The arithmetic
source is prevented outright and tested. The other is the declared resolution
itself: within a class members are indistinguishable, so a coarse class heals a
starving member at a well-provisioned member's expense while conserving the total
exactly. That is not fixable by accounting and is asserted rather than avoided,
with a companion test showing a round trip is exact when each class holds one
member. Reading conservation of the total as evidence that any individual is right
would be a mistake this pair of tests exists to prevent.
Coverage limits. Materialization is not claimed to be the inverse of collapse and
no test asserts that it is beyond the one-member-per-class case. Nothing checks
that a proposed draw is ecologically sensible. A full draw yields at most one unit
per member more than a partial one, which is an ordering artifact bounded by the
class count and is not separately tested at every count. Injury, fertility and
incubation are not reconstructed because a cohort never carried them. Ownership
remains a single-process invariant, emigration escrow is still unsettled, and
nothing is serialized. The overflow and ledger-failure branches are reachable only
past bounds the tests do not build.
No existing behaviour changed. The proof golden `d61ad5bf0c928c76`, the ecosystem
fingerprint `2cd4a15f9fbe9949`, the ensemble fingerprint `5ec59d390909b526` and
the frozen genome fixture are unchanged, and the other 36 tests pass untouched. No
deployment manifest or image pin changed. Linux validation is pending; no Gitea
result is claimed since `b01a824`.
Review of `380d609` found two defects (2026-09-24), both fixed with regressions
confirmed to fail against the unfixed code.
Materialization treated an empty reservoir as an available one. A vacant cohort, or
one simply holding nothing, still owns its store, so a draw could credit that store
while the owning record went on recording nothing: the same inventory in two places,
reached from the opposite direction to the destination defect found in `1a59b63`.
Destinations are now checked against the registry's ownership as well as their
balance, and the registry answers whether a reservoir is used rather than callers
inferring availability from what it contains.
The same hole existed on the collapse side in a worse form, and was found by
checking whether the reported defect had a sibling rather than by being reported.
Nothing verified that the destination reservoir was unowned, so the shared-reservoir
rule added in M4.3 would have refused the install only after the ledger transfer had
already run, leaving the members emptied and their biomass in a reservoir nothing
accounted for. That is exactly the stranding the M4.2 ordering was chosen to prevent,
defeated by a check that ran too late. Whichever registry step will follow a transfer
is now proved to succeed before it, through new `may_install` and `may_replace`
queries that `install` and `replace` themselves delegate to, so there is one rule
rather than a restatement that could drift.
The second finding was a functional gap the oscillation test had been hiding by
discarding its registry every cycle. A full draw leaves a vacant record, which
ADR-0036 retains because absence is not extinction, and collapse then refused that
habitat and genome as occupied. A full materialization was therefore a one-way door
out of a habitat: the lineage that had been there could never return. Collapse now
repopulates a vacant record in place, continuing its ownership and revision series
rather than restarting them, and a record that still holds a population is still
refused, because merging two silently is not a conversion. The oscillation test keeps
one registry for all twelve cycles and asserts the revision continues across each
repopulation, so the round trip is exercised against the world it left behind.
Four of the five new guards were confirmed load-bearing by removing each alone: the
ownership check on both sides, the collapse pre-check ordering, and the repopulation
path. The fifth is not, and is recorded rather than presented as tested. The
`may_replace` pre-check in `commit_materialize` is unreachable given the checks
before it: the installed record has already been proved equal to the one the plan
was built from, and the plan's result carries the same habitat, genome, reservoir and
tick with a strictly later revision, so the replacement cannot be refused. Removing
it changes nothing the tests can see. It is kept because it is the structurally
correct place for the guard and its sibling on the collapse path is genuinely
reachable, not because it is exercised.
An environment note for the record: a Bash heredoc containing apostrophes failed to
parse, so the larger test edits in this increment were written to a scratchpad file
and run from there. That is the third distinct way heredocs have mangled content in
this project, after backslash escapes and `/tmp` resolving differently between
shells.
## M4.5 resolution comparison
Executed on this Windows host (2026-09-24): Release configure, build and `ctest`
38/38; the analysis preset (clang-tidy 19, warnings as errors) built clean; the ASan
preset configured, built and ran 38/38; `clang-format` 19.1.5 reported no
violations. All four checks ran and passed on the committed tree. No Clang or GCC
compile happens on this host, so Linux CI remains the first such result.
The increment adds no library code apart from one defect fix described below. One
rule is written once and applied at both resolutions; the individual run gives it a
member at a time and the aggregate run gives it a class average and multiplies by
the count. The aggregate run sub-steps at one tick, so interval-length error is not
confounded with averaging error, and interval-length error is therefore unmeasured.
The control is the part that makes the rest mean anything. With one member per class
a class average is that member, so the two runs must agree exactly; the test checks
population and every channel at every tick over forty ticks, and checks the class
count stays inside the sixty-four-class bound. Introducing a one-unit drift into the
aggregate upkeep is caught by it, so the instrument was verified rather than assumed.
Without this control a measured divergence would be indistinguishable from a
transcription mistake in one of the two runs.
Measured, and reported rather than smoothed into a tolerance. Class widths ten and
fifty: the aggregate population survives ten thousand ticks, with population gaps of
250 and 375 parts per thousand and energy gaps of 24 and 43. Class widths
twenty-five, one hundred and two hundred: the aggregate population reaches zero while
the individual run sustains seventy-five members. That is a false extinction caused
by resolution alone. The numbers are in
`docs/engineering/results/resolution-equivalence-msvc1944.jsonl`.
The collapse is not monotone in class width: twenty-five destroys the population and
fifty sustains it. An earlier version of the test asserted that coarser is never
better, and the measurement refuted it; the assertion was removed rather than the
widths being chosen to make it hold, and the non-monotonicity is now asserted
instead, because it is surprising and because the warning written around it could
otherwise become wrong without failing. Five widths locate the effect and do not map
it; no mechanism beyond whole-class starvation is claimed.
Conservation was exact at every tick on both sides of all five runs, including the
three whose populations died out. Four of five conserve perfectly and three are
ecologically worthless, which is the clearest available demonstration of
ECOLOGY_MODEL's point that conservation is insufficient ecological evidence.
Two flaws in my own experiment were found and fixed before any number was believed.
The aggregate model originally shifted class bounds every tick and opened newborn
classes at the full declared width, which produced overlapping classes: the model
being measured was not one a cohort record could represent, and the first numbers
measured were therefore of a model the library could not hold. Newborns now extend
the youngest class down while it is narrower than the width, keeping the structure
contiguous and ascending, and every step of the aggregate run is now checked to
construct a valid `Cohort`. The earlier figures were discarded.
The comparison also found a real defect in M4.3. `plan_advance` checked affordability
with arrivals added before losses were taken, but emitted the ledger legs with the
outflows first. The ledger applies legs in sequence, so an interval affordable as a
whole underflowed leg by leg and was refused at commit with `ledger_failure`: a
fifty-tick interval shedding 10,480 energy against 6,950 held while drawing 13,080
in reproduced it exactly. Arrivals are now the first leg, matching the order the
arithmetic was validated in. The fix was confirmed load-bearing by restoring the old
order, which fails both suites, and a dedicated regression in `evo-population-tests`
covers an interval affordable only with its arrivals. One existing test had seeded
its escrow assuming the old order and was corrected, with its comment; that comment
had asserted the old ordering was what made the interval fundable, which was true
and is now true of the opposite order.
Coverage limits. One genome, so the mixed-genome diversity case ECOLOGY_MODEL also
asks for is untested. One model, one scarce axis, one initial distribution and one
capacity, so every number describes this fixture. The control runs over tens of
ticks because one class per age exhausts the class bound, while the measurement runs
over ten thousand: the instrument is checked at a shorter horizon than it is used at.
The asserted envelope bounds are round numbers above the measurement with room for
platform variation, so they are a regression guard and not a theoretical limit.
Interval-length error, crash behaviour at conversion boundaries beyond one process,
distribution across owners and persistence are all out of scope here.
The proof golden `d61ad5bf0c928c76`, the ecosystem fingerprint `2cd4a15f9fbe9949`,
the ensemble fingerprint `5ec59d390909b526` and the frozen genome fixture are
unchanged. No deployment manifest or image pin changed. Linux validation is pending;
no Gitea result is claimed since `b01a824`.
Review of `b4c3da5` found one defect (2026-09-25), now fixed with the controls
extended to catch it.
The two runs allocated contested breeding slots in different orders. The individual
run iterated members in insertion order; the aggregate run walked classes as stored,
which is youngest first. When eligible parents outnumber free slots that decides
which parents reproduce, and it does so for a reason that has nothing to do with
averaging, so the published divergence was not attributable to resolution. The
control never reached the capacity boundary and so could not have seen it, which is
the part that made ADR-0040's central claim unsupported rather than merely
imprecise.
The allocation order is now part of the stated rule, oldest first, honoured at both
resolutions: classes partition ages, so the oldest class holds the oldest members,
and ordering inside a class remains something only the individual run can see, which
is a resolution effect rather than an implementation difference.
A second control was added at the capacity boundary, where twelve eligible parents
contend for one free slot. Both controls now compare the whole distribution of
members rather than censuses. That change matters on its own: whoever breeds hands
the same investment to a child either way, so every total agrees while the
distribution does not, and a census-only control cannot see a different parent
having bred. The reported probe diverging only by tick 44 is exactly that effect.
The attribution was then verified by mutation rather than argued. Restoring the
individual run's insertion order fails the new control; restoring the aggregate
run's youngest-first order fails it; and the earlier one-unit drift in the aggregate
upkeep still fails the first control. Since either allocation order alone fails,
the divergence reported in the results file is attributable to averaging.
One of my own attempts at the control was wrong and was caught by the same mutation
sweep. The contested fixture first gave ages that descended with the index, on the
reasoning that this made insertion order differ from age order; it does the
opposite, since descending ages make insertion order identical to oldest-first, and
the individual-side mutation passed. Ages now ascend with the index, so oldest-first
disagrees with both the insertion order and the stored class order, and each
mutation alone is caught. Had the sweep not been run, the new control would have
looked like it established attribution while covering only half of it.
The measurements were regenerated and are unchanged: the same five widths, the same
two regimes, the same gaps. That is worth stating plainly rather than presenting as
confirmation, because the finding was about what the numbers could be attributed to
and not about their values.
Coverage limit added. Attribution rests on the controls covering every stage where
the two runs could order things differently. Breeding allocation is the only
order-sensitive stage, and the environment is sized so intake never depletes it,
which the fixture asserts rather than assumes. A future stage introducing another
order-sensitive decision would need its own control before these numbers could be
trusted again.
## M4.6 local validation - 2026-09-26
Base 55a728a plus working M4.6 changes. Integrated the prior agent's untracked
`tests/fixtures/resolution_model.hpp`; no production library changes or format
version changes. Existing contested-parent and round-trip controls remain active.
- Native MSVC 19.44 Release configure/build and CTest: 39/39 passed.
- Native analysis (clang-tidy 19, including fixture headers): build and 39/39 passed.
Initial analysis found a throwing error-reporting boundary in the benchmark;
replaced catch-handler stream writes with stdio diagnostics and reran successfully.
- Native AddressSanitizer: build and 39/39 passed. Windows ASan does not cover UBSan.
- clang-format 19 format-check passed, now including tests/*.hpp. Diff whitespace
check passed.
- Full Release benchmark: 15 scenarios, one warmup plus three measured repetitions
per representation. Raw results and host/method caveats are in
[AGGREGATE_BUDGET.md](AGGREGATE_BUDGET.md). All scenarios conserved every channel;
none qualified across all scarcity inputs under the proposed research gate.
- Linux CI, UBSan, target-server profiling and production deployment were not run.
No World0 resources or user source documents were changed.
The final exception-handler-only adjustment was compiled in analysis; Release and
ASan benchmark targets were then rebuilt and their resolution tests rerun. No
performance or model algorithm changed after the archived timing run.
## M5.1 local voxel contract - 2026-09-26
Added ADR-0042, the local voxel contract and dependency-free `evo::world`. No
existing simulation golden, codec or runtime dependency changed. The earlier
uncommitted M5-M9 roadmap expansion is preserved in the working tree.
Executed on Windows x64 / MSVC 19.44 with the repository helper:
- `tools/windows-check.ps1 -Preset release -Action check`: configure/build,
warnings-as-errors and 40/40 CTest passed.
- `tools/windows-check.ps1 -Preset analysis -Action check`: clang-tidy 19 build
and 40/40 CTest passed.
- `tools/windows-check.ps1 -Preset asan -Action check`: AddressSanitizer build
and 40/40 CTest passed. Native Windows ASan is not UBSan coverage.
- `tools/windows-check.ps1 -Preset release -Action format`: passed.
- `git diff --check`: passed.
New Release-active tests enumerate independent signed chunk intervals, all 299,520
logical cell ordinals across edges 8/16/32/64, envelope corners, each-axis invalid
bounds including INT64_MIN/MAX and UINT32_MAX, metric coordinates and all valid
positive fill numerators. Typed rejection is checked before reading expected values.
No allocation/terrain-generation/performance claim is made for the address envelope.
Linux GCC/Clang CI and UBSan have not been run for this increment. No container,
Argo or World0 deployment occurred. M5.2 storage benchmarks remain the next task.
## M5.2 compact chunk comparison - 2026-09-26
Added `evo-chunk-layout`, Release-active CTest checks, ADR-0043 and the
[measured report](CHUNK_LAYOUT_EVALUATION.md). No production world API, persisted
format or simulation golden changed.
Executed on Windows x64 / MSVC 19.44:
- `tools/windows-check.ps1 -Preset release -Action check`: 41/41 passed.
- `tools/windows-check.ps1 -Preset analysis -Action check`: clang-tidy 19 and
41/41 passed after fixing a multiplication-width warning and reserving edit input.
- `tools/windows-check.ps1 -Preset asan -Action check`: 41/41 passed.
- `tools/windows-check.ps1 -Preset release -Action format`: passed.
- Release `evo-chunk-layout`: completed; archived 180 measured samples after warmups.
Independently checked matching initial/edited hashes across all layouts and repeats
for each of 20 edge/fixture pairs. Full per-cell oracle checks also run internally.
- `git diff --check`: passed.
Memory figures are owned capacities, not RSS or true allocation peaks; encoding
sizes are estimates, not a codec. See the report for excluded buffers and timing
limits. Native ASan does not cover UBSan. Linux CI, R730xd benchmarking and Kubernetes
execution were not run locally. No Argo/World0 deployment occurred. Next: M5.3.
## M5.3 deterministic abiotic generation - 2026-09-26
Implemented ADR-0044 in evo::world with a versioned inert material table and pure
point/chunk queries. The fixture creates no biological state. No existing simulation
version, golden or persistence codec changed.
Executed on Windows x64 / MSVC 19.44:
- `tools/windows-check.ps1 -Preset release -Action check`: 42/42 CTest passed.
- `tools/windows-check.ps1 -Preset analysis -Action check`: clang-tidy 19 build and
42/42 CTest passed.
- `tools/windows-check.ps1 -Preset asan -Action check`: 42/42 CTest passed.
- `tools/windows-check.ps1 -Preset release -Action format`: passed.
- `python tools/world_generation_reference.py`: computed the 12 frozen fixture
fingerprints independently before C++ execution; all match in CTest.
- `git diff --check`: passed.
New tests cover 262144 point/chunk comparisons across signed seams, query order,
stratigraphy, known/unknown materials, version/edge rejection and extreme addresses.
Fingerprint bytes are explicitly little endian logical fields. Linux CI and UBSan
were not run locally; no cross-platform execution result is claimed. No deployment
or World0 state change occurred. Next: M5.4 bounded sparse chunk ownership.
## M5.4 bounded chunk ownership - 2026-09-26
Added CompactCells, a fixed-slot ChunkStore, revision-validated read tokens and
conservative dirty pinning under ADR-0045. No cell edit, persistence, distributed
lease or clean acknowledgement API was introduced. Existing goldens remain unchanged.
Executed on Windows x64 / MSVC 19.44:
- `tools/windows-check.ps1 -Preset release -Action check`: 43/43 CTest passed.
- `tools/windows-check.ps1 -Preset analysis -Action check`: clang-tidy 19 build and
43/43 CTest passed.
- `tools/windows-check.ps1 -Preset asan -Action check`: 43/43 CTest passed.
- `tools/windows-check.ps1 -Preset release -Action format`: passed.
- `git diff --check`: passed.
The initial test oracle mixed unsigned local offsets with a negative integer literal;
corrected to explicit int64 arithmetic. Analysis required explicit optional checks
at reads/eviction and identified an integer-width expression in the test. One narrow
use-after-move suppression exercises the explicitly guaranteed empty-void source.
Final checks above ran after those corrections.
Coverage includes palette counts through 256 then dense fallback, full logical cell
equality, material/fill distinctions, move validity, extreme/far addresses, exact
accounted budget boundaries, failed-load state preservation, slot pressure, clean
regeneration, stale/foreign tokens and dirty eviction refusal. Counter exhaustion is
guarded but not injected; no allocation-failure fault injector is claimed. Memory
accounting excludes allocator/runtime overhead and is not an RSS cap.
Linux CI/UBSan and target-host performance have not been run locally. No deployment
or World0 change occurred. Next: M5.5 accounted transactional edits.
## M5.5 transactional excavation - 2026-09-26
Added bounded atomic excavation, finite per-material debris volume, full-payload
retry identity and a 64-command replay ring under ADR-0046. Existing biology/world
base goldens and persisted formats were not changed. Formatter coverage now includes
private headers under src, including the shared chunk-state definitions.
Executed on Windows x64 / MSVC 19.44:
- `tools/windows-check.ps1 -Preset release -Action check`: 44/44 passed, including
the final late-source-failure test and nonthrowing commit static assertions.
- `tools/windows-check.ps1 -Preset analysis -Action check`: clang-tidy 19 build and
44/44 passed. After the final targeted additions, rebuilt analysis and reran
`ctest --test-dir build/analysis -R world-edit-contracts --output-on-failure`: passed.
- `tools/windows-check.ps1 -Preset asan -Action check`: 44/44 passed. After those
additions, rebuilt ASan and reran the same targeted test in build/asan: passed.
- `tools/windows-check.ps1 -Preset release -Action format`: passed.
- `git diff --check`: passed.
Coverage includes seam-crossing excavation, four chunks/64 cuts, partial fills,
rock/regolith volume conservation, all-or-nothing capacity/memory/source failures,
missing/stale revisions, duplicate/changed requests, malformed bounds and retention
expiry. Late source failure verifies an earlier prepared chunk is not published.
No allocator failure injection, revision exhaustion injection or crash recovery is
claimed. The source/replacement commit operations have compile-time nonthrowing guards.
Linux CI and UBSan were not run locally. These are typed volume transfers, not a
validated density/mass/chemistry model. No deployment or World0 modification occurred.
Next: M5.6 canonical snapshots/edit journals preserving geometry, debris and retry state.
## M5.6 canonical world snapshots and edit journals - 2026-09-26
Added ADR-0047, byte schemas, canonical snapshot/restore and detached atomic journal
replay. State includes base versions, identity, membership, revisions, geometry,
debris, successful edit count and exact bounded retry records. No existing golden
or biological simulation format changed. New snapshot v1 has an independently encoded
132-byte empty fixture from tools/world_snapshot_reference.py.
Executed on Windows x64 / MSVC 19.44:
- Release configure/build and 46/46 CTest passed on the final source/tests.
- Analysis configure/build (clang-tidy 19) and 46/46 CTest passed after explicit
size/iterator conversion fixes. Final history-corruption additions rebuilt and
passed both tests selected by `ctest --test-dir build/analysis -R world-codec
--output-on-failure`.
- ASan configure/build and 46/46 CTest passed. After final size/iterator/golden and
history-corruption additions, rebuilt ASan and reran both world-codec tests: passed.
- `tools/windows-check.ps1 -Preset release -Action format`: passed.
- `python tools/world_snapshot_reference.py`: generated independent golden bytes;
encoder equality and decoder acceptance pass in CTest.
- `git diff --check`: passed.
New coverage includes 65-edit history rollover, deterministic continuation, historical
retry receipts, exact re-encoding, snapshot+journal replay, wrong predecessor/final
state, every fixture truncation, oversized/trailing data, version/identity/budget
rejection, recomputed-checksum geometry/debris/receipt/revision/history corruption,
and a deterministic 128-input semantic fuzz smoke. The fuzz harness mutates valid
fixtures and repairs checksums so validation below the envelope is exercised.
Linux libFuzzer/UBSan jobs are wired for 10000 cases but were not executed locally.
Native Windows ASan is not UBSan. Full-size throughput, allocation-failure injection
and crash-atomic files remain unverified/later gates. Codecs write no files, clear no
dirty flags and issue no durable acknowledgements. No Argo/World0 operation occurred.
Next: M5.7 local snapshot/journal publication, compaction and recovery.
## M5.7 review and M5.8 bounded world proof - 2026-09-26
Reviewed a2a12fdc696c005bd3feb496f6efcdc58c28262c against the three prior findings.
Committed-journal compaction compares exact recovered canonical state; poisoned
handles refuse recovery; interruption tests use changed payloads and include partial
write failures. No new blocking finding identified in that correction's scope.
M5.8 adds ADR-0049, evo-world-proof and world-process-restart CTest. It launches fresh
prepare/compact/verify processes for three workloads, checks all cells and conserved
debris independently, compares canonical bytes, enforces frozen hashes and checks
clean regeneration. Existing formats and simulation goldens are unchanged.
Executed on Windows x64 / MSVC 19.44:
- tools/windows-check.ps1 -Preset release -Action check: 48/48 passed, including the
final source with frozen hashes and nonthrowing boundary diagnostics.
- tools/windows-check.ps1 -Preset analysis -Action check: clang-tidy 19 and 48/48
passed. Initial widening-conversion and exception-boundary diagnostics were fixed.
- tools/windows-check.ps1 -Preset asan -Action check: 48/48 passed.
- tools/windows-check.ps1 -Preset release -Action format: passed.
- Five independent Release process workflows passed; medians/ranges and exact
accounting/payload sizes are in WORLD_PROOF.md. Measurements preceded only the
equivalent size_t constant and error-reporting correction; no success-path
workload, snapshot or algorithm changed.
- git diff --check: passed.
CTest integrates the same frozen expectations into existing Linux CI, but this
turn did not execute Linux, UBSan, physical power-cut, target-server or cluster
checks. ASan is not UBSan. Process restarts are graceful; M5.7 fault injection
remains separate coverage. No production durability or planet capacity claim.
No Argo or World0 operation occurred. Next: M6.1 units and environmental coupling.
## M6.1 environmental units and coupling - 2026-09-27
Added ADR-0050, ENVIRONMENTAL_UNITS.md, checked local habitat geometry, exact mass
concentration/water saturation and rational pJ/tick integration helpers. No field
storage, density converter, source scheduler, gameplay or snapshot format changed.
Terrain volume still cannot be credited as biological mass.
Executed on Windows x64 / MSVC 19.44:
- `tools/windows-check.ps1 -Preset release -Action check`: 49/49 passed.
- `tools/windows-check.ps1 -Preset analysis -Action check`: clang-tidy 19 and
49/49 passed.
- `tools/windows-check.ps1 -Preset asan -Action check`: 49/49 passed.
- After strengthening independent balance sums, open-boundary and rollback tests,
reconfigured each preset, rebuilt `evo-environment-units-tests`, and ran
`ctest --test-dir build/<preset> -R '^environment-units-contracts$' --output-on-failure`:
passed in Release, analysis and ASan. Final code-only adjustment was an explicit
cstddef include; CTest registration was grouped with other world tests.
- `tools/windows-check.ps1 -Preset release -Action format`: passed.
- `git diff --check`: passed.
Coverage includes each invalid box axis, zero/maximum inputs, exact dimensional
examples, concentration/type separation, dry/saturated/excess water, multiplication
and addition overflow, invalid denominators/residuals, and 78,660 repeated-addition
oracle comparisons of split versus combined intervals. The closed/open accounting
example uses a finite boundary reservoir, independently sums balances and verifies
rejected overdraft leaves all balances unchanged. Documentation's large terrain
mass example was checked using exact BigInteger division (1e21 / 65535).
Linux/UBSan and target-cluster checks were not executed locally; existing CI includes
the new CTest. There is no new concurrency requiring TSan work. Helpers do not enforce
owner identity, interval deduplication or residual persistence; those remain explicit
M6.2-6.7 obligations. No deployment or World0 operation occurred. Next: M6.2.
## M6.2 optional environmental fields - 2026-09-27
Added ADR-0051, FieldStore, field snapshot v1 and its byte schema. Optional water,
N/P, organic matter/energy, incident-energy and heat channels have explicit finite
openings. Material defaults specify water capacity only. Uniform baselines with
optional 64-site pages preserve absence versus zero. Same-channel local transfers
stage replacements before commit and persist a strictly increasing command watermark.
There is no resource setter, site regeneration, cross-compartment or organism credit.
Executed on Windows x64 / MSVC 19.44:
- `tools/windows-check.ps1 -Preset release -Action check`: 50/50 passed on final code.
- `tools/windows-check.ps1 -Preset analysis -Action check`: clang-tidy 19 and
50/50 passed on final code.
- `tools/windows-check.ps1 -Preset asan -Action check`: 50/50 passed on final code.
- `tools/windows-check.ps1 -Preset release -Action format`: passed.
- `git diff --check`: passed.
Initial build diagnostics corrected explicit aggregate initialization and bit widths.
The hostile fixed-underlying-type enum test has a narrowly scoped analyzer suppression
with rationale; actual channel validation remains exercised.
Tests cover uniform/sparse/varying memory, page seams and release, exhausted sources,
source/capacity/budget/stale-command rollback, absent/present-zero restore, cross-field
conserved-channel overflow, independent dense transfer continuation and all ten fields
at 4096 sites. Maximum logical snapshot is 630880 bytes. A separately constructed
170-byte absent-field fixture checks canonical schema bytes. Every truncation of a
260-byte fixture, invalid/old versions, identity, malformed/resealed mappings and
quantities, oversized snapshots and 256 repaired-checksum mutations are exercised.
Memory observations are in ENVIRONMENTAL_FIELDS.md; accounted bytes are not RSS.
Existing terrain/simulation formats and goldens are unchanged. Field snapshot v1
is a buffer codec, not LocalWorldFiles publication or combined terrain/biology
durability. Rate sources and their interval/residual persistence are not enabled.
Linux/UBSan, target-host performance, allocator-failure injection and cluster recovery
were not executed locally. Existing Linux CTest CI includes the new test. No deployment
or World0 operation occurred. Next: M6.3 read-only environment views.
## TSan world-proof scratch mount correction - 2026-09-27
The supplied Linux TSan log failed world-process-restart at CMake MAKE_DIRECTORY,
before launching any proof phase. Its build-local output path had no writable mount
in the read-only test container; Testing and cli-integration already had mounts.
Added the matching 16 MiB, mode-1777, noexec/nosuid tmpfs for
`/source/build/tsan/world-proof`. Container user, rootfs, network, capabilities and
seccomp restrictions are unchanged. Documented all test scratch mounts in deploy/README.md.
Executed `ctest --test-dir build/release -R '^world-process-restart$' --output-on-failure`
after configuring the Release preset: passed on Windows. `git diff --check` passed.
The workflow mount path was checked against the CMake test's TEST_DIR. No C++ changes
or new sanitizer claims. Local Docker CLI exists, but docker version could not connect
because the dockerDesktopLinuxEngine named pipe is absent. Consequently the corrected
Linux TSan container invocation requires confirmation in Gitea CI; it was not run
locally. No application deployment or World0 changes.
## M6.3 read-only environment views - 2026-09-27
Added ADR-0052 and a bounded canonical footprint query with typed mass/energy,
exact concentration/moisture, habitat resolution, explicit caller tick, field
watermark and sampled chunk revisions. FieldStore gained read-only identity and
capacity accessors. Views own copied samples and retain private non-owning store/
chunk tokens for freshness checks. Existing formats, versions and goldens are unchanged.
Executed on Windows x64 / MSVC 19.44:
- `tools/windows-check.ps1 -Preset release -Action check`: 51/51 passed on final code.
- `tools/windows-check.ps1 -Preset analysis -Action check`: clang-tidy 19 and
51/51 passed. Removed an ineffective move of a trivially copyable sample after
the initial analysis diagnostic.
- `tools/windows-check.ps1 -Preset asan -Action check`: 51/51 passed on final code.
- `tools/windows-check.ps1 -Preset release -Action format`: passed.
- `git diff --check`: passed.
Tests cover negative/positive chunk seams, input permutation, missing site/chunk,
absent versus zero/unrequested fields, all moisture statuses, invalid masks and
footprints, exact quantities/ratios, maximum 256-site footprint with all ten fields,
and unchanged terrain/field snapshots after queries. Freshness checks cover changed
tick, restored foreign stores, successful versus rejected field transfers, global
field invalidation, unrelated terrain activity, sampled dirty marking, excavation
and eviction/reload. The query never loads missing chunks or authorizes consumption.
Linux/UBSan/TSan and cluster execution were not run locally; existing CI includes
the new CTest without additional scratch mounts (this test writes no files).
Host tick/store pairing and single-owner execution are explicit preconditions.
ChunkStore does not yet bind a world identity, so coherent field/terrain pairing
cannot be independently authenticated by this view. Combined ownership, reservation,
atomic inventory transfer and retry outcomes remain M6.4. No deployment or World0
operation occurred.
## M6.4 atomic environmental transfers - 2026-09-27
Executed on Windows x64 with the installed Visual Studio toolchain:
- `tools/windows-check.ps1 -Preset release -Action check`: build and 52/52 CTests passed (NDEBUG).
- `tools/windows-check.ps1 -Preset analysis -Action check`: clang-tidy/warnings-as-errors build and 52/52 CTests passed.
- `tools/windows-check.ps1 -Preset asan -Action check`: AddressSanitizer build and 52/52 CTests passed.
- `tools/windows-check.ps1 -Preset release -Action format`: passed.
New environment-transfers-contracts exercises last-stock contention, fresh versus
stale/foreign views, exact retries, changed-payload refusal, expired command IDs,
command exhaustion, late-leg rollback, account and water capacities, admission
overflow, separate owner/field allocation budgets, and 100 batches checked against
independent expected balances across all ten compartments. No golden changed.
This evidence covers local in-memory atomicity. Coupled snapshots, allocation-failure
injection, process crash recovery and biological lifecycle integration are not claimed.
Linux/UBSan/TSan were not run locally; their existing Gitea jobs remain required CI
evidence. No cluster workload or World0 state was changed.
## M6.5 prerequisite review - 2026-09-27
Documentation-only review identified contradictory source-persistence sequencing in
ADRs 0050/0051 and 0053. ADR-0054 proposes a resolution; it is not accepted or
implemented. `git diff --check` passed. No source, public API, test, schema or build
configuration changed, so builds/sanitizers were not rerun for this proposal. The
M6.4 validation above remains historical evidence, not M6.5 completion evidence.
## M6.5a coupled resource codec - 2026-09-27
ADR-0054 accepted by the maintainer. Added resource envelope v1 without changing
nested terrain/field formats or existing simulation goldens. Rates are still disabled;
source configuration/residuals and biology join the later M6.5c fixture envelope.
Executed Windows checks:
- Release check: 52/52 CTests passed with NDEBUG.
- Analysis check: clang-tidy/warnings-as-errors build and 52/52 CTests passed.
- ASan check: 52/52 CTests passed.
- After tightening restore to require all mapped chunks already in the snapshot,
Release and analysis rebuilt and environment-transfers-contracts passed again;
the ASan full run includes this final adjustment.
- Release format target passed.
Codec tests cover empty history, 66 transfers crossing the retry-ring boundary,
canonical restored bytes, exact/conflicting/expired retries, equivalent continuation,
all truncation lengths, checksum errors, recomputed-checksum invalid version/counts/
amounts, identity mismatch and each budget. Initial compilation caught missing codec
error enum entries; corrected before the passing runs above.
No Linux/UBSan/TSan, disk crash, allocation-failure injection or cluster validation
was performed locally. Buffer round trips are not durable publication. M6.5b-d remain.
## M6.5b process contract - 2026-09-27
Documentation-only segment: accepted ADR-0055 and ENVIRONMENTAL_PROCESS.md define
versioned fixture conversions and the narrow process/world ownership boundary.
Inspected M2 producer/feeding implementations, their tests and ADRs, M3 metabolic
budget contract, and M6 resource APIs before specifying the adapter.
Executed independent PowerShell arithmetic checks for the documented fractional
capture/efficiency sequence, nutrient-limited growth and per-channel feeding split;
all passed. `git diff --check` passed. These are specification checks, not runtime
integration tests. No C++/CMake/schema change; builds, CTest and sanitizers were not
rerun for this documentation segment. M6.5c/d must supply executable evidence for
rollback, source exhaustion, temperature limits, lifecycle and restored replay.
## M6.5c persisted environmental community - 2026-09-27
Implemented the one-habitat/two-actor fixture and envelope v1, preserving source
identity/rate, residuals, last funded interval, lifecycle, resources and 64 exact
input/report receipts. Restore replays receipts from a historical anchor, validates
reports and compares the final state. Scalar candidate staging commits without
allocation; the world layer checks capacities and conserved totals before publication.
Executed Windows x64 checks:
- Release check: 53/53 CTests passed, NDEBUG enabled.
- Analysis check: clang-tidy/warnings-as-errors build and 53/53 CTests passed.
- ASan check: 53/53 CTests passed.
- After final tight-budget and continuous/restored comparison additions, analysis
ran all 53 tests again; Release and ASan rebuilt and the two environmental
transfer/community tests passed again.
- Release format check passed. No existing golden changed.
Coverage includes hand-calculated capture/growth/feeding, source disabled/unfunded,
frozen residuals, inclusive temperature gates, terminal producer, whole-tick rollback,
exact/conflicting/expired retries, eight deterministic input sequences through 90
ticks, continuous versus restored bytes/reports, every truncation length, corruption,
recomputed-checksum invalid config/anchor/report/count/state, identity and budgets.
M6.5d remains the broader adversarial, long-run and cost qualification gate.
The supplied Gitea log shows GCC 14 Release rejecting transfer_codec.cpp with
-Werror=free-nonheap-object in the vector reserve/push_back path. Changed that writer
and the new process writer to exactly sized buffers, checked writes and final cursor
validation. No warning suppression or resource format change. The next Gitea run must
confirm Linux: Docker's Linux engine was unavailable locally; WSL Ubuntu exists but
has no g++/g++-14, CMake or Ninja. No toolchain installation was attempted. No local
Linux/UBSan/TSan or cluster result is claimed.
Initial native static analysis flagged an unspecified enum base and integer widening
in a size constant; both corrected before the passing analysis runs. No disk/host-loss
recovery, biological maintenance, reproduction or player/offline progression added.
## Environment-view optional guards - 2026-09-27
The supplied Linux clang-tidy-19 log rejected five optional field accesses in
`tests/environment_views_tests.cpp` with bugprone-unchecked-optional-access. Three
had no presence check; two relied on CHECK, whose throwing helper is not recognized
as a control-flow guard by this analysis. All five now use required_field, which
explicitly throws on absence before returning a reference. The rule remains enabled;
production code, formats, expected values and NDEBUG behavior are unchanged.
Executed native analysis, Release and ASan builds plus the targeted
`environment-views-contracts` CTest in each preset; all passed. Release format check
and git diff --check passed. Full suites were not repeated for this test-only fix.
Linux clang-tidy/libstdc++ confirmation remains with the next Gitea run; native
analysis is not claimed as a reproduction of the Linux diagnostics.
## M6.5d environmental qualification - 2026-09-27
Added the environmental-qualification CTest and optional `--long` driver. It tests
admission/capacity/overflow boundaries, each missing recipe resource, efficiency
endpoints, consumer temperature gates, recomputed-checksum mutations, a u64-max
trace and deterministic paired continuous/restored simulations. No runtime change.
See [ENVIRONMENTAL_QUALIFICATION.md](ENVIRONMENTAL_QUALIFICATION.md) for detailed
coverage, exact commands, long-run output, machine and measured memory/timing.
Executed Windows checks:
- Release configure/build/check: 54/54 CTests passed.
- Analysis configure/build/check: clang-tidy/warnings-as-errors and 54/54 passed.
- ASan configure/build/check: 54/54 passed, including tick exhaustion.
- After adding the final tick-exhaustion case, Release and analysis rebuilt and
environmental-qualification passed again. A constant-expression integer-width
diagnostic in the new test was corrected before that final analysis run.
- Release format and git diff --check passed.
- Final Release `evo-environmental-qualification.exe --long`: 16 x 65536 logical
ticks and paired continuous branches passed; 2097152 advances, 329231 bites,
matching snapshots and conserved totals. Reported fingerprint 93a9f9b296caac36.
Native M6.5 qualification is complete; no new Linux/UBSan/TSan or R730xd/cluster
run was available locally. CI must confirm the committed test on its toolchains.
No source-rule, schema, golden or deployment change.
## M6.6 retained corpse resources - 2026-09-27
Rules/envelope v2 now support trusted terminal facts for either fixture actor,
retained corpse inventory and bounded tagged decay. This is an in-memory process
checkpoint, not durable region publication or gameplay mortality. Rule v1 and its
canonical bytes are preserved; no golden was replaced. Config/input extensions
are rejected under v1 rather than silently omitted.
New tests cover pending death plus feeding rollback, terminal actor rejection,
shared per-channel caps in ID order, exact transfer destinations, partial decay,
zero caps, water headroom, restored duplicate/conflicting/stale requests, permanent
terminal state, consumer-corpse nutrients feeding a surviving producer next tick,
every v2 byte truncation, and checksum-repaired invalid lifecycle/history fields.
Eight 90-tick cases compare retained balances against independent cumulative-quota
formulas, with restore at ticks 17, 65 and 90 and exact conservation every tick.
Commands: tools/windows-check.ps1 -Preset release/analysis/asan -Action check;
-Preset release -Action format; git diff --check. The initial suites found the
older negative test's rules=2 was now valid; it now rejects unsupported rules=3.
Final results are recorded below after all processes complete.
Release long qualification again ran 16 x 65536 logical ticks, 2097152 paired
advances, and produced unchanged v1 fingerprint 93a9f9b296caac36. Snapshot size
remains 30615 bytes in that v1 case. Owner size is now 20152 bytes (formerly13400),
including bounded extension receipts; budget admission uses sizeof dynamically.
No new Linux/UBSan/TSan or R730xd measurement is claimed. Linux CI remains a
separate verification step; local Docker Linux engine/toolchain is unavailable.
Final native validation: Release 54/54 (3.66s), clang-tidy analysis 54/54 (13.37s),
ASan 54/54 (46.72s), format check and git diff --check passed. M6.6 is natively
qualified; the next gate is M6.7 interval scheduling. No deployment was changed.
## M6.7 local environmental intervals - 2026-09-27
ADR-0057 and ENVIRONMENTAL_SCHEDULING.md define the stateless interval adapter.
Admission validates world/frame, fixed quantum, completed process/field revision,
ordered endpoints, at most 64 ticks and an exact input count. Each process tick is
atomic; execution failure explicitly returns the committed prefix and failing tick.
No new scheduler state, snapshot format, biology rule, dependency or deployment.
Tests compare 130-tick streams under rules 1 and 2 against direct execution using
batch widths 1/3/17/64, flattened report equality and final canonical snapshots.
Every batch pauses without mutation and restores before continuing. Admission
cases include foreign/stale boundaries, quantum mismatch, missing/extra input,
65-tick backlog, u64-max backlog and reversed intervals. Failure at first, middle
and final tick preserves precisely the committed prefix, then restores and resumes
to the uninterrupted result. The existing valid u64-max checkpoint now also tests
scheduler pause and wrap refusal. Partial success is intentionally not a whole-batch
transaction; callers must inspect failure and committed count.
Executed tools/windows-check.ps1 -Preset release/analysis/asan -Action check:
Release 54/54 (6.44s); clang-tidy analysis 54/54 (14.20s); ASan 54/54.
After making the optional failure access explicitly guarded, rebuilt Release and
reran both environmental CTests: 2/2 passed. Analysis/ASan include that final guard.
Release format-check and git diff --check passed. Tests retain NDEBUG checks.
Linux/UBSan/TSan CI confirmation remains pending; native validation does not claim
a Linux or target-server run. Next is M6.8 integrated ecological qualification.
## M6.8 integrated environmental qualification - 2026-09-27
Extended the existing qualification executable; no runtime, schema or golden
changed. Complete evidence and reproduction commands are in
[ENVIRONMENTAL_SLICE_QUALIFICATION.md](ENVIRONMENTAL_SLICE_QUALIFICATION.md).
- tools/windows-check.ps1 -Preset release -Action check: 54/54 passed (4.06s).
- -Preset analysis -Action check: clang-tidy and 54/54 passed (18.57s).
- -Preset asan -Action check: 54/54 passed (60.83s); qualification itself 12.83s.
- -Preset release -Action format and git diff --check passed.
- Release evo-environmental-qualification.exe --long passed: existing v1 long
fingerprint unchanged, plus 5120 v2 episodes/655360 logical ticks/1310720 paired
advances. Full-cycle assertions, checked E/C/N/P/other totals, separate water,
residual ranges, stable resource residency and restored retry equality passed.
- Three integrated execution/snapshot/restore cost samples were collected on the
developer machine. Concurrent analysis/ASan work is disclosed in the report.
M6 is natively qualified for the bounded synthetic scope. Linux CI/UBSan/TSan,
R730xd measurements and persistent server recovery are not claimed. The local
Linux toolchain/Docker engine remains unavailable. Next: M7.1 durability contract.
## M7.1 region durability contract - 2026-09-27
Documentation-only gate: ADR-0058, REGION_DURABILITY.md and logical region-commit
schema. Reviewed existing M6 process/resource codecs and scheduling contracts,
LineageArchive public API and retention tests, individual genome codec, separate
population/particle interfaces, persistence architecture and ADRs 0013/0016.
The audit identifies complete M6 state and missing archive/binding/host state.
Full checkpoint batches are selected; semantic events alone are not recovery.
Verified new contract relative links (2 resolved) and all eight ADR sections;
git diff --check passed. Reviewed the crash matrix against upload, fenced DB
publication, ambiguous commit, ACK, recovery and clone boundaries. Those rows
are future executable acceptance tests, not fault-injection results from this gate.
No C++/CMake/schema codec implementation changed, so build/sanitizer suites were
not rerun. M6.8's recorded native checks remain the latest executable evidence.
No database, object storage, Kubernetes access or deployment was performed.
M7.1 design gate is complete. M7.2 must evaluate/pin dependencies and exercise real
migrations; M7.4 must implement additional ancestry state closure before M7.9 can
claim lineage recovery. No retained-history or World0 requirement was relaxed.
## M7.2 PostgreSQL bootstrap - 2026-09-28
- Windows Release configure/build/CTest passed 54/54 (24.17s); format passed.
- Built checksum-verified PostgreSQL/libpq 18.6 from source in WSL Ubuntu.
- GCC14 Release, clang19 analysis (clang-tidy warnings-as-errors), and clang19
ASan/UBSan optional adapter builds and real private-database integration passed.
Tests cover parameter separation/NULL, rollback, duplicate SQLSTATE, u64 range
and fractional rejection, runtime-role denial, deadline closure, result bounds,
backend termination, clean/repeated migrations, failed migration rollback,
unknown versions and changed checksums. Expected SQL errors are negative tests.
- Fixed analysis enum-size warning and migration rejection exit status during
qualification; reran all three final configurations successfully.
- clang-format19 dry-run passed for all three new C++ files; bash -n passed for
preparation, migration, integration and publish scripts. ldd found all psql
dependencies: libpq, libc/libm and OpenSSL (matching tooling image packages).
- Docker Linux engine is unavailable locally. Container build/smoke and Gitea
remote checks are configured, not claimed as executed. No cluster, external DB,
TLS endpoint or Kubernetes Secret was accessed. TSan was not rerun: adapter has
one caller-owned I/O thread and introduces no concurrent execution.
The adapter is separate from the headless kernel. Region fencing, publication,
ambiguous durable commit reconciliation and crash recovery are later M7 gates.
Only clean-to-v1 and exact-v1 restart are supported migration paths at this gate.
## M7.3 fenced region ownership - 2026-09-28
- Windows MSVC Release configure/build/CTest passed 55/55, including the new
portable `region-lease-contracts` test. Repository clang-format 19 check passed.
- WSL Ubuntu (GCC 14.2.0; clang 19.1.1; PostgreSQL/libpq 18.6 source build):
GCC14 Release, clang19 analysis (clang-tidy warnings-as-errors) and clang19
ASan/UBSan builds of `evo-postgres-tests` each passed `tools/ci/postgres-check.sh`
against a fresh private cluster. The first analysis build flagged two
unchecked optional accesses and two unreserved vectors. These were fixed and
all three configurations were rerun after formatting.
- Real-database coverage, as the runtime login rather than SET ROLE:
- Denied (42501): table read/write, provisioning, the internal check function,
and object creation. A hostile search_path is ignored.
- Busy, forged-holder, stale-epoch and released-lease rejections; epoch
non-reuse.
- 1..64 tick delta; predecessor mismatch; exact-retry detection via the head.
- 1 s database-time expiry and successor acquisition; paused stale writer fenced
locally and by the database.
- A writer queued 1.3 s on the lease row lock is rejected after expiry.
- A 100 ms renewal deadline under lock (unknown) fences locally before a
successor acquires.
- 64 two-thread acquisition rounds with exactly one winner and consecutive
epochs; 32 racing head advances with exactly one winner.
- Epoch and sequence exhaustion.
- Migrations tested: clean->v2, repeated v2 no-op, v1->v2 upgrade, and failed
0001 rollback. Superuser and missing runtime roles are refused with no schema
left. Unknown v3, changed v2 checksum and changed v1 checksum are refused.
- Mutation sensitivity (scratch copies only): replacing post-lock
`clock_timestamp()` with `now()`, removing the lease check from head advance,
and disabling the busy check each made the suite fail at the expected test.
- Local Unix-socket latency, 200 samples per operation in each final run:
- p50 about 0.61-0.74 ms and p99 about 0.8-1.15 ms for acquire, renew,
advance and release.
- Max up to 2.8 ms under analysis/ASan builds.
These are local numbers, not remote TLS or cluster measurements.
- Not executed: Docker/Gitea CI, remote TLS endpoints, Kubernetes Secrets, real pod
kill, real network partition (a lock wait plus client deadline stands in for
it), VM freeze and TSan. The contention threads each own a separate
connection; libpq is uninstrumented.
### M7.3 review fixes - 2026-09-28
- P1: an attribute-only runtime role check let a login inheriting
`pg_write_all_data` update `evolution.region_leases` directly. The CI script
reproduces that UPDATE first. `apply.sql` now audits effective privileges on
every run (the v2 migration file and its checksum are unchanged). Rejected, each
with a rollback:
- a runtime group inheriting `pg_write_all_data` on a fresh database (no
schema left);
- that grant on the login, and on the group WITH INHERIT FALSE;
- a PUBLIC table grant, EXECUTE on the internal check function, and a
CREATEDB login.
The restored configuration passes again.
- P2: `valid(LeasePolicy)` now bounds each term by the duration before any
arithmetic. `acquired` validates the duration before computing a deadline.
`renewal_unknown` ignores durations above 60 s and fences on non-positive ones.
New tests use `milliseconds::max()` and `min()` for every field. Built against
the previous source under clang19 UBSan, they reproduce the reported signed
overflow (`9223372036854775807 + 2000`).
- Reran Windows CTest 55/55, the repository format check, and GCC14 Release,
clang-tidy analysis and ASan/UBSan private-PostgreSQL suites: all passed.
## M7.4a checkpoint publication - 2026-09-28
- Windows MSVC Release configure/build/CTest passed 56/56, including the new
portable `region-checkpoint-contracts`. It covers:
- golden byte offsets and round trips;
- malformed and unsupported manifests, including u32 values above 255 that
must not truncate onto supported enums;
- chain rules;
- payload identity, tick, quantum, corruption and budget checks;
- 24-tick continuation equality after restore.
The repository clang-format 19 check passed.
- WSL Ubuntu, with GCC 14.2.0, clang 19.1.1, PostgreSQL/libpq 18.6 and OpenSSL
3.0.13-0ubuntu3.15, running as uid 1000 so permission faults were really
exercised. GCC14 Release, clang19 analysis (clang-tidy warnings-as-errors) and
clang19 ASan/UBSan each passed `tools/ci/postgres-check.sh`.
- Analysis initially flagged `performance-enum-size` on the u32-based
profile/durability enums. They are now u8 with explicit decoding of values
above 255 as unknown. All three configurations were rerun.
- New real-database/filesystem coverage:
- SHA-256 vectors (NIST "", "abc" and the 448-bit message) and agreement with
PostgreSQL `sha256()`.
- Object store: rejects root symlinks, creates 0700 namespaces, puts are
idempotent, never overwrites differing content, ignores temporary leftovers,
and detects byte flips, symlinks, truncation and growth. Size bounds and
write-permission failure are also covered.
- Genesis provisioning and duplicate refusal; three batches (16/64/16 ticks);
a 65-tick batch refused before upload.
- Recovery from a new store handle is byte-identical and continues 44 ticks
identically to an uninterrupted reference.
- The build allowlist is enforced by the host and by the database.
- Upload failure leaves the head unchanged. A fenced publication leaves a
verified orphan, and recovery ignores it.
- Missing and corrupt payloads quarantine; restoring the object recovers.
- Owner edits to a manifest (23514) and head repointing (23503) are refused.
With the constraint dropped, recovery still detects the edit, and the
constraint is restored afterwards.
- Unknown regions are never created.
- Lease tests were ported to manifest publication. They additionally cover
malformed manifests, a quantum change, an unretained build, a future rules
version and a globally reused commit ID.
- Migrations: clean→v3, repeated no-op, v1→v3. v2 with existing heads fails
closed and stays at v2. The audit rejects EXECUTE on provisioning and SELECT on
commits. Tampered v2/v3 checksums and an unknown v4 are refused.
- Mutations, run on scratch copies with the source restored afterwards: removing
the predecessor-digest check, the lease check, the version/quantum continuity
check or the retained-build check in SQL, or the host manifest re-digest, each
failed at the expected assertion.
- Local fixture measurements: payloads of 22775 and 37367 bytes.
- Release publication took 3.6-4.8 ms per batch; analysis 5.5-6.7 ms; ASan
about 10.5 ms.
- Manifest publish p50 was about 0.74-0.91 ms.
These are local socket and local disk only.
- Not executed: Docker/Gitea CI, a remote object store or PVC, power-loss fsync
semantics, read-back verification fault injection, process kill, and TSan.
## M7.4b lineage checkpoint profile - 2026-09-28
- Windows MSVC Release CTest passed 57/57, including the new `lineage-region-contracts`
and extended lineage and region-checkpoint contracts. The repository clang-format
19 check passed.
- WSL Ubuntu (GCC 14.2.0, clang 19.1.1, PostgreSQL 18.6, OpenSSL 3.0.13):
- The Linux `checks` build set (separate build dirs) passed CTest 57/57 in each
of GCC14 Debug, GCC14 Release, clang19 analysis and clang19 ASan/UBSan.
- The PostgreSQL integration suite passed 2× Release, 2× analysis and 5×
ASan/UBSan.
- Archive codec coverage:
- canonical encoding regardless of admission order, including a child with a
smaller ID than its parent;
- checksum, magic, version, count cap, length and organism-count framing;
- genome damage, unknown genome, invalid alive/died forms, death before birth,
reordered organisms (non-canonical) and a child encoded before its parent;
- a full 4096-genome/65536-organism round trip at 2,543,648 bytes.
- `LineageRegion` coverage:
- closure refusals: missing, identical or dead binding; low high-waters; stray
living organism; future birth; archive death while the community is active;
- the death recorded at tick 20 in the same advance; a refused tick changes
nothing;
- restore plus 36 continuation ticks identical to the original;
- container, binding, envelope and archive damage refused.
- Profile 2 coverage:
- the 4,268,144-byte cap is asserted in C++ and enforced by SQL (the maximum is
accepted, +1 refused); the profile-1 1 MiB+1 is refused;
- mismatched profile/history is EVL06;
- lineage publish (death included), recover and continue, and a second
publication;
- the environmental path refuses a lineage predecessor before upload;
- the database refuses a profile switch;
- a host of the other profile reports unsupported.
- Migrations: clean→v4; v1→v4; v3 with a committed genesis upgraded to v4 with
the commit retained; v2 heads still block. Tampered v2/v3/v4 checksums and an
unknown v5 are refused.
- Mutations (restored afterwards), each failing at the expected line: removing
the SQL profile/history continuity check (store test 392), the living-set
closure check (lineage region 58), or death recording (lineage region 78).
- Flake found and fixed. The ASan PostgreSQL suite failed intermittently: 1 of 3,
then 1 of 5 runs, at the queued-writer and expired-renewal assertions.
- Cause: a 4-minute probe measured WSL2 stepping CLOCK_REALTIME back 470-596 ms
every ~32 s (3.66 s total, roughly 1.5% versus CLOCK_MONOTONIC).
- PostgreSQL `clock_timestamp()` follows that wall clock, so a 1.2 s local sleep
could be under 1 s of database time.
- Tests now poll `clock_timestamp() >= expires_at`, holding the row lock where
relevant. Afterwards the suite ran 5/5 under ASan.
- The fencing logic itself was unchanged; the requirement is recorded in
ADR-0060.
- Local measurements, Release:
- manifest publish p50 0.72 ms and p99 1.24 ms;
- checkpoint publication 3.5-5.0 ms for 23-37 KiB payloads.
- Not executed: Docker/Gitea CI, remote storage, process kill, reproduction within
a batch (the fixture has none), durable history rows (M7.5) and TSan.
## M7.5 durable batch outcomes - 2026-09-28
- Windows MSVC Release CTest passed 58/58, including the new `region-batch-contracts`:
- recorder sealing at a death or at 64 ticks, early flush, confirm matching, and
a refused tick recording nothing;
- trace round trip, size (19,552 bytes at 64 ticks), range, contiguity and
header binding;
- decision order, uniqueness, range, and refusal of unknown values.
The manifest v2 golden layout, v1 decoding and batch-evidence structural rules
were also covered.
- WSL Linux: the `checks` build set (GCC14 Debug and Release, clang19 analysis,
clang19 ASan/UBSan) passed CTest 58/58 each, and the repository format check
passed.
- The first PostgreSQL clang-tidy build flagged a returned const-reference
parameter and two unchecked optional accesses; all three were fixed. Final
PostgreSQL suites: Release 2 runs, analysis 2, ASan/UBSan 4, all passing.
- Real-database coverage:
- Durable death at tick 20, with decisions and outbox rows in the same commit.
- Outbox: redelivered, acknowledged true then false, then empty; invalid limit
and unknown rows refused.
- A lost response retried with identical bytes returns the same receipt and
records no second death; resolution says committed.
- A changed manifest or changed decisions conflict (EVL07/22023). Other regions
and unknown IDs resolve as conflict and absent.
- A repeated death is EVL08. A forged same-size block, a partial block and a
reordered block are 22023. A gap is EVL03. None of these moved the head or
decisions.
- A 300 ms client timeout while queued on the lease row became unknown; once the
lock cleared the server committed, resolution returned committed and the retry
was idempotent.
- A terminated orphan request resolved as absent, and the same bytes were
retried and committed.
- A missing predecessor payload is `predecessor`. State advanced outside the
batch fails replay. Batches from the wrong predecessor and unsealed batches are
refused before upload.
- Recovery reconciles decisions. An owner-inserted death the state does not
show quarantines with `history`, and cannot be deleted (EVL09).
- Commit and decision UPDATE and DELETE are EVL09 even for the owner; the
defense-in-depth test disables and restores the trigger explicitly.
- Lineage profile: a death is recorded with the archived organism id and
reconciled against the archive.
- Migrations: clean→v5, v1→v5 and v3→v5 with an existing v1-manifest commit;
tampered checksums for v2 through v5 and an unknown v6 are refused.
- Mutations (scratch SQL copies; host source restored and rerun green), each
caught: removing idempotent retry, outcome finality, the decision-digest check,
decision recording, or recovery reconciliation.
- Local measurements, including predecessor replay: Release 6.6-8.1 ms per batch;
analysis about 20-23 ms; ASan about 19-21 ms. Manifest publish p50 about
0.75-0.86 ms.
- Not executed: Docker/Gitea CI, remote object store, process kill, a broker
consumer, an injected failure between commit insert and statement end, and TSan.
### M7.5 review fixes - 2026-09-28
- P1 object durability:
- `put()` now re-fsyncs the file and the directory before reusing an existing
object, on both the found-existing and the lost link-race paths.
- Namespace opens always sync the parent directory.
- A test-only `ObjectStoreFaults` hook injects directory-sync failures and
counts successful syncs.
- Tests: a namespace-creation sync failure, then a reopen still failing, then
success with exactly 2 syncs. An install sync failure leaves the object
visible but unacknowledged, three retries while syncs fail all return io with
no extra syncs, then success with +1 sync; every reuse adds a sync.
- Publication with 1, 1 and 2 injected failures returns `upload`/io and sends
no manifest (head stays 0), then publishes once syncs succeed.
- Reverting the reuse sync or the always-sync namespace behavior makes these
tests fail.
- P1 pre-v5 history:
- `pre_v5_guard.sql` (EVL10) runs before 0005 in every upgrade path. 0005 and
its checksum are unchanged.
- CI builds a real v4 database: v1 genesis, a lease, and a sequence-1
publication through the v4 function. The v5 upgrade is refused with the
expected message; history stays 1..4, both commits remain, and no decision
table is created. Replacing the guard with a no-op makes that check fail.
- A committed death published without any decision (the state an accepted
legacy upgrade would produce) recovers as `history`, quarantined and never
served.
- Reran GCC14 Release (3 runs), clang-tidy analysis (1) and ASan/UBSan (2)
PostgreSQL suites, all passing, and the repository format check. The changes
are Linux-only persistence code, SQL and tests; the portable targets and
Windows are unchanged.
## M7.6 region host - 2026-09-28
- Windows MSVC Release CTest passed 58/58; this change touches only the Linux
PostgreSQL CMake section and Linux sources. The repository clang-format 19 check
passed.
- WSL Linux: GCC14 Release, clang19 analysis (clang-tidy) and clang19 ASan/UBSan
builds of `evo-postgres-tests` and `evo-server` passed the full PostgreSQL suite.
Final runs: Release 3, analysis 1, ASan 3 across both verification passes.
- The first analysis build flagged unchecked optional accesses and ignored release
results in the host and server, plus two test issues. These were fixed with
explicit local checks rather than suppressions, then rebuilt and rerun in all
three configurations.
- Host scenarios (real PostgreSQL, scaled scheduling clock, 4 s test lease):
- Admission: not_ready before recovery, 64-input cap (full), duplicate and
invalid refusals, byte accounting.
- Ticks 1–19 are provisional with no acks. Tick 20's death pauses and commits
in the same step: 20 acks and one outcome (consumer, organism 11, sequence 1).
The next batch flushed by interval at tick 41.
- Starvation: debt reached 10 without invented ticks, then catch-up of 8 per step.
- Storage outage: the sealed batch waits with nothing acknowledged and debt
growing, then commits when storage recovers.
- Lease row locked: the commit becomes unknown, the host is not ready and
refuses admission, and resolution adopts the receipt after release.
- Renewals blocked past the local deadline: fence, all 8 unconfirmed inputs
reported `fenced`, then re-acquisition at a newer epoch from the committed head.
- Drain (environmental profile): backlog dropped as drained, 6 pending ticks
committed, lease released, and a successor acquired immediately at epoch 2.
- Unrestorable build and foreign frame quarantine; an invalid configuration is
refused.
- Process tests (real `evo-server` via posix_spawn, stdin and stdout pipes):
- SIGTERM after 80 inputs: exit 0, 80 acks, head 80 at sequence 8, lease
released, readiness file removed, max_queued 64. The outcome line followed
tick 20's final ack.
- SIGKILL during running: every acked tick ≤ the committed head (11/11) and the
lease was left held. The successor was ready after 3.94–3.97 s at epoch 2 with
committed=11, then drained cleanly.
- Mutations (restored and rerun green), each caught: acknowledging at simulation
time, not committing a sealed death, reporting ready while resolving, and
draining without releasing the lease.
- Not executed: Docker/Gitea CI, TSan (the host is single-threaded; libpq is
uninstrumented), remote PostgreSQL/TLS, a production 20 s lease kill test,
durable client deduplication, and a network gateway.
## M7.7 recovery before readiness - 2026-09-28
- WSL Linux: GCC14 Release, clang19 analysis (clang-tidy, clean on the first
build) and clang19 ASan/UBSan builds of `evo-postgres-tests` and `evo-server`
passed the full PostgreSQL suite, including the M7.7 recovery tests. Runs:
Release 2, analysis 1, ASan 2. The repository format check passed after
reordering one fixture include.
- Windows was not rerun. This change touches only Linux persistence and host
code, SQL, tests and docs; M7.6's Windows run (58/58) covers the portable
targets, which are unchanged.
- Kill matrix (real SIGKILL via `--crash-at`, inputs 1..40 with a death at tick
20, 5 s flush):
| Crash point | Head (seq/tick) | Acked up to | Decisions | Objects |
|---|---|---|---|---|
| before-transaction:1 | 0/0 | none | 0 | 3 (verified orphans) |
| before-upload:2 | 1/20 | 20 | 1 | 3 |
| before-transaction:2 | 1/20 | 20 | 1 | 5 (orphans) |
| after-commit:2 | 2/40 | 20 (lost acks are the safe direction) | 1 | 5 |
| after-ack:2 | 2/40 | 40 | 1 | 5 |
The lease stayed held after every kill. Each successor became ready at epoch 2
from the committed head (with tail verification), then drained and released.
- Resumed stale owner: the owner was SIGSTOPped for 4.5 s (4 s lease). The
successor became ready at epoch 2 with committed=10 and processed ticks 11–20.
After SIGCONT the old owner printed `ready 0`, rejected new inputs as
`not_ready`, acknowledged nothing beyond tick 10, and exited 0 on SIGTERM
without touching the successor's chain.
- Damaged heads: a flipped head payload byte and a missing head trace each made
`evo-server` exit 4 (quarantined) without ever printing `ready 1`, and the head
was unchanged. After repair a successor became ready at epoch 4.
- Tail verification (in-process): a missing or corrupt head trace and a missing
predecessor payload are `tail`, and recovery succeeds again after repair. A
published trace of different inputs over the same range (a host bypass the
database cannot detect) is `tail`.
- Mutations (restored and rerun green): skipping tail verification made the
quarantine test time out because the server became ready. Dropping replay
byte-equality failed the forged-trace test (line 262).
- Not executed: Docker/Gitea CI, the production 20 s lease, host loss or VM
freeze, multi-host network partitions, or verification of links older than
the head.
## M7.8 operations endpoints - 2026-09-28
- Windows MSVC Release CTest passed 59/59, including the new portable
`ops-contracts`: exposition format, HELP/TYPE per family, the fixed 30-series
set, bounded labels, one-hot phase and non-wrapping lag. The exposition is
3485 bytes typically and 3836 at worst-case values (limit 8192).
- WSL Linux: the `checks` build set (GCC14 Debug and Release, clang19 analysis,
clang19 ASan/UBSan) passed CTest 59/59 each.
- The PostgreSQL suites (GCC Release, clang-tidy analysis, ASan/UBSan) passed
after fixing the ops test's optional-access and reserve findings. The earlier
PostgreSQL analysis runs used a stale binary and are not counted. The
repository format check passed after formatting the server.
- Process probe tests (real `evo-server`, loopback, ephemeral port):
- Serving: `/livez` 200 `live`, `/readyz` 200 `ready`. Metrics showed ready 1,
epoch 1, lease remaining above 0, capacities 64/65536 and the running phase.
After 30 inputs: committed tick 30, lag 0, 30 acknowledgements, checkpoint
age under 60 s.
- HTTP bounds: POST 405, unknown path and query string 404, malformed 400,
over 4 KiB 431. Eight idle connections do not block probes; a ninth is
closed within 500 ms; idle ones close at their deadline.
- Database unreachable: `/livez` 200, `/readyz` 503 `not ready: acquiring`,
ready 0, store failures above 0, never `ready 1`, exit 0 on SIGTERM.
- Stale process (SIGSTOP 4.5 s, then SIGCONT): `/livez` 200, `/readyz` 503,
ready 0, fences_total 1, lease remaining 0. The successor answers 200 at
epoch 2.
- Mutations (restored and rerun green): readiness that returns 200 whenever live
was caught. Raising the connection bound to 1000 was not caught at first,
because the check did not distinguish refusal from deadline expiry. The test
now requires the refusal within 500 ms and catches it.
- Image: the Docker engine was not running locally (Docker Desktop stopped), so
the `server-runtime` build and `tools/ci/server-image-smoke.sh` have not
executed. A filesystem emulation (binary plus the pinned libpq, `env -i`)
showed usage exit 2 with no settings and all libraries resolved. It also found
and fixed a bug: without arguments the server had reached object-store opening
instead of refusing with usage. The binary keeps a build-tree RUNPATH, which is
harmless in the image.
- The manifest example parses (Deployment and NetworkPolicy, three probes) and
the workflow YAML parses; `bash -n` passed for the smoke script.
- Not executed: the container build and smoke, the Gitea job, cluster apply,
Prometheus scraping, and TSan.
## M7.9a region clones and retained history - 2026-09-29
- Windows MSVC Release CTest passed 59/59. `lineage-region-contracts` now covers
rebinding: only the new identity restores, continuation matches, and rebinding
back is byte-exact. `region-checkpoint-contracts` covers the clone-genesis
manifest rules.
- WSL Linux `checks` set (GCC14 Debug and Release, clang19 analysis, clang19
ASan/UBSan) passed CTest 59/59 each. The format check passed.
- The PostgreSQL suites passed in GCC Release, clang-tidy analysis and ASan/UBSan.
These runs used a second database, `evo_clone`, with its own runtime group and
login. clang-tidy first flagged branch-clone in the tool's option parsing and
optional access in the new tests; both were fixed, then the suites were rebuilt
and rerun.
- Harness additions:
- A populated v6 database (a v2 genesis) upgrades in place to v7. The replaced
head constraint is `region_heads_published_tick`, and the v7 parser reads the
stored genesis.
- Upgrade paths now end at 1..7.
- Version and checksum drift checks now include v7.
- Clone tests (real `evo-server` and `evo-world-clone` processes):
- Trajectory: world 7 was run to tick 30, with a death at tick 20, then cloned
to world 8 in `evo_clone`. The clone recovered under the clone runtime at
tick 30, with the inherited death and retained ancestry, and the audit found
one clone link naming the source cut. Both worlds then ran inputs 31..90; after
rebinding the clone to world 7, the snapshots were byte-identical.
- Isolation:
- A same-world target was refused as `identity`.
- A world that already holds an original region was refused (22023).
- A forged origin was refused (22023).
- The runtime role cannot provision a clone (42501).
- A clone world refuses an original region.
- Origins are immutable (EVL09).
- The source session is read-only (25006, even through SECURITY DEFINER).
- A clone lease publishing a successor to the source is fenced.
- The clone runtime login cannot reach the source's functions (42501).
- The source head and decisions were unchanged afterwards.
- Failure: a SIGKILL before the target transaction left the source unchanged,
the target unprovisioned with one orphan object, and a target server that
quarantined (exit 4) without becoming ready. The tool refused a same
database or same user (exit 2) and a same world (exit 4). The retry
succeeded and reused the same object. A source with a damaged head payload
was refused (exit 3), and its target stayed untouched.
- Retention: an owner was SIGKILLed after acknowledging tick 100, and its
successor continued to tick 200. The audit verified every link, and the tool
printed the same result. A lower `max_commits` returned `bounds`. Removing the
first link's trace made the audit report `tail` at sequence 1 while head
recovery still succeeded; restoring the trace made the audit pass again.
- Environmental profile: a clone inherited the tick-20 death. Moving that
durable row to tick 19 (triggers disabled as break-glass) made recovery
report `history` and the audit report `history` at sequence 0. Moving it back
made recovery pass.
- Mutations (each restored and rebuilt afterwards). Each was caught by its
intended check:
- dropping the isolated-world refusal;
- dropping the origin check;
- a read-write source session;
- skipping the inherited-outcome digest;
- skipping the audit's per-link tail verification.
The first attempt at the inherited-outcome mutation did not compile, so its
stale-binary run was discarded and the mutation was rerun.
- Not executed: clones across separate PostgreSQL clusters or hosts, Docker, the
Gitea CI, and cluster runs.
## M7.9b upgrade barrier and durability qualification - 2026-09-29
- Every M7.9b change is in the Linux PostgreSQL build or in docs. Windows was not
rerun; its 59/59 from M7.9a still covers the portable code, which is unchanged.
- WSL Linux `checks` set (GCC14 Debug and Release, clang19 analysis, clang19
ASan/UBSan) passed CTest 59/59 each. The format check passed.
- The PostgreSQL suites passed in GCC Release, clang-tidy analysis and ASan/UBSan.
After the mutation run, one host test was corrected: `unrestorable_head_quarantines`
now retains its build, so the quarantine it checks again comes from recovery
rather than acquisition. Release and analysis were rebuilt and rerun green after
that change. The ASan build compiled with it, but its full run predates it.
- Harness: upgrade paths now end at 1..8. The populated v6 database upgrades to v8.
Version and checksum drift checks include v8.
- Migration and barrier test (database `evo_online`, created at v7 from the
checked-in files):
- A pre-v8 writer with a build-less lease published ticks 1..20 (with the
death).
- An `apply.sql` run with an invalid v8 checksum failed. It left history 1..7,
no `retired_at` column and no four-argument acquire. The writer renewed and
published ticks 21..40.
- The real migration then reached v8 while the writer held its lease, and the
writer published ticks 41..60.
- Barrier refusals: the runtime cannot call it (42501), and retiring the last
current build is refused (22023).
- After the barrier (1 lease fenced), the old writer's publication and renewal
were fenced. Build-less acquisition returned EVL06, and acquisition with b1
returned unsupported.
- A b2 lease publishing a link that names b1 was refused as unsupported.
- A b2 `evo-server` became ready at epoch 3 from committed tick 60 and
acknowledged through tick 90. Its final head is built by b2.
- The audit passes with both builds retained, and reports unsupported at
sequence 0 with only b2.
- A b1 `evo-server` quarantined (exit 4) without becoming ready, and the head
was unchanged.
- Qualification: see engineering/DURABILITY_QUALIFICATION.md. SIGKILL to ready
took 19.5–19.6 s under the production lease in all three builds (the 60 s
target is asserted). Acknowledgement latency p50 was about 549 ms and p99
about 1048 ms. No acknowledged input was lost, and one durable death was
recorded across five owners.
- Mutations (each restored and rebuilt). Each was caught:
- no retired-build trigger;
- no build check in four-argument acquisition;
- no barrier check in build-less acquisition;
- a barrier that does not fence holders;
- a host that retries an unsupported acquisition instead of quarantining.
The migration test caught this one after the host-test correction.
- Not executed: Docker, the Gitea CI, cluster runs, a separate database host, and
host or storage loss (M8).
## M7.6–M7.9 review fixes (schema v9) - 2026-09-29
Review `.cache/reviews/M7_6-M7_9_REVIEW.md` found three issues. All three are
fixed:
- **[P1] Build retirement could race acquisition.** Schema v9 applies one
upgrade-lock order across the barrier, acquisitions, lease checks, provisioning
and the retired-build trigger.
- **[P2] Clone-world classification could race.** Provisioning now takes a
per-world lock and checks classification after it.
- **[P2] The operations listener accepted without bound.** Each `serve()` now
makes at most 4 accept attempts.
Evidence:
- WSL Linux `checks` set (Debug, Release, analysis, ASan/UBSan): CTest 59/59
each. The format check passed.
- PostgreSQL ASan/UBSan: full suite passed. Analysis and Release were rerun green
after fixing two clang-tidy findings in the new tests. Upgrade paths now end at
1..9.
- Race tests (database `evo_race`): each holds a transaction open, confirms the
contender is blocked on a lock via `pg_stat_activity`, then commits.
- A build-less acquisition queued behind a barrier is refused (EVL06).
- A build-aware acquisition of the build being retired is refused, and the
region stays unheld.
- The next build acquires.
- Racing barriers for the last two current builds: one returns 22023 and one
build stays current.
- Different-source clones into one world, and an original region racing a
clone: the second is refused (22023).
- Same-source clones into different regions both succeed.
- The final origins are exactly `2/1<10, 4/1<10, 4/2<10`.
- Accept budget:
- With 16 queued connections, successive `serve()` calls give 4 open, then 8
open with 0 refused, then 8 open with 4 refused.
- Refusals never exceed 4 per call under replenished connections.
- The first accepted probe got 200.
- A real server acknowledged 60 ticks under continuous connect/close churn
without losing readiness.
- Mutations, each reverting one fix (restored and rebuilt). Each was caught:
- an unlocked build-aware acquire;
- an unlocked legacy acquire;
- an unlocked barrier;
- no per-world lock;
- an accept budget of 1000.
- Qualification was unchanged: SIGKILL to ready took 19.55–19.60 s.
- Windows was not rerun. All changes are Linux PostgreSQL, server or doc files.
## M8.1a inventory preparation - 2026-09-29
- No cluster access was available: there is no kubeconfig on the workstation.
Nothing in the cluster was read or changed, and the `aevolution-db` Secret was
not accessed. Only the key names the maintainer supplied are recorded.
- `tools/inventory/collect-cluster.sh` passed `bash -n`. It has not been run
against a cluster. It only reads, and it lists Secrets through the server-side
table, so no data is fetched.
- `deploy/kubernetes/inventory/disk-probe.yaml.example` parses as a PVC,
NetworkPolicy and Job. Its script was run locally in WSL2 on ext4 as a
baseline:
- 4 KiB O_DSYNC writes: mean 1169 µs;
- 40 KiB object publication: mean 4646 µs, max 7983 µs;
- 256 MiB fdatasync: 365 ms.
It has not been applied in any cluster.
- The qualification now records the server's resource profile from /proc. Over
three owners in Release: peak RSS 9,940 KiB, and 60 ms CPU over 16.3 s of
serving. The full PostgreSQL Release suite passed with it. The analysis build
compiled clean, and the format check passed.
## M8.1b cluster inventory - 2026-09-29
- The maintainer ran `tools/inventory/collect-cluster.sh`. It made read requests
only, and listed Secrets as name, type and key count. The output is in
`C:\tmp\cluster` on the workstation and is not committed. The results are
summarized in docs/infrastructure/INVENTORY.md:
- RKE2 v1.32.5 with one untainted control plane (6 vCPU / 11.7 GiB) and three
workers (8 vCPU / 31.3 GiB).
- No kubelet reservations, quotas or LimitRanges.
- Storage is `local-path` only (reclaim Delete, no CSI, no snapshots).
- Canal/Calico, ingress-nginx, MetalLB and cert-manager with Let's Encrypt.
- `aevolution-dev` has no automated sync.
- No PostgreSQL 18.6 in the cluster (Gitea runs 17.6).
- The collector ran as cluster-admin. A read-only identity is an M8.2 item.
- `deploy/kubernetes/inventory/disk-probe-local-path.yaml` parses as one
NetworkPolicy and three PVC/Job pairs, pinned to worker-1..3. It is outside
every Argo path and has **not yet been applied**, because this workstation has
no kubeconfig. The maintainer approved running it. Its results go in
INVENTORY.md.
## M8.1c storage probe and corrections - 2026-09-29
- The maintainer applied `deploy/kubernetes/inventory/disk-probe-local-path.yaml`
by hand on the cluster, and all three Jobs completed. The resources were then
deleted.
- worker-1: 1906 µs dsync, 28.7/60.6 ms object publish, 1003 ms for 256 MiB.
- worker-2: 2511 µs, 36.0/73.4 ms, 2122 ms.
- worker-3: 2239 µs, 32.1/61.0 ms, 2339 ms. Its root filesystem is 87 % full.
Every `local-path` volume is on the node's root LVM ext4 volume. The figures are
provisional until the planned RAID change.
- Corrections from the maintainer:
- Host RAM was increased by about 500 GB, so the earlier "about 20 GiB left"
estimate is withdrawn.
- AEvolution's PostgreSQL is a separate VM outside Kubernetes, reached through
`aevolution-db`.
- Collection is manual, with no automation planned.
- No credential values were read. The 18.6 version, role and TLS requirements are
enforced by the adapter and migrations when M8.3 connects.
## M8.2 environment isolation - 2026-09-29
- `tools/ci/deploy-policy-check.py` (WSL, Python 3 with PyYAML): 35 objects in 27
files, 0 failures.
- Nine mutations were each caught, then restored:
- World0 auto-sync;
- the World0 cascade finalizer;
- the policy set to failurePolicy Ignore;
- the policy no longer covering Secrets;
- an egress hole in default-deny;
- the World0 protected label removed;
- the cleanup tool accepting world0 (rc 127, it reached kubectl);
- service-account token automount;
- a wildcard stage destination.
- `kubectl kustomize` (client v1.31.4) renders the dev, stage and World0 platform
overlays (9 objects each), and still renders the live `deploy/kubernetes/dev`
path unchanged.
- `bash -n` passes for both ops scripts.
- The CI `deploy` suite (a Docker `ci-deploy` stage) has not run yet; Docker is
not running locally.
- Not executed: applying anything to the cluster, the admission policy's CEL
compilation (it needs an API server; the runbook validates it with a server
dry run first), and `tools/ops/verify-isolation.sh`.