Wave P drafted at 97% and then cost a dozen clean rebuilds to bank, and not one of those rebuilds
failed on a matching problem. Banks the whole failure surface:
A. The SEVEN under-reporting holes in gate_main, all the same shape (R32): the checker never read
the destination TU, shared headers, a draft's own definition, lines with trailing comments,
typedef aliases, the build's own error text, or file order. Law: audit a batch-integration tool
for what it DOESN'T look at -- its verdicts can be correct on the inputs it reads and still be
worthless, because the compiler reads more.
B. Typedef handling, with the two wrong strategies that both look right: blanket STRIP (assumes
the surviving definition sits above the insertion point -- src/800.c defines Rec14 at 7336
while stubs wanting it sit at 7272), blanket RENAME (breaks drafts sharing an identical
typedef, because their externs stop agreeing -- my regression, three drafts at once), and the
rescan loop that deletes the definition it just renamed. The survivor is body-aware +
position-aware in a single pass over a snapshot.
C. The remaining limit: conflict detection compares spelled type NAMES, so three drafts each
defining their own Slot54 with different layouts all declare func_80032A74(Slot54*) and compare
equal. Comparing struct LAYOUTS is the real fix.
D. The measured cost shape -- drafting cheap and solved, integration expensive -- and therefore the
next lever: a STATIC pre-gate check over the substituted text, no make at all. Plus the R39
lesson that negative controls apply to the tool you are FIXING, not just the one you ship.
A wave is now sized by INSTRUCTION MASS, not card count. The metric is instruction-weighted, so a
wave is worth what its instructions are worth: the 12-42-ins card lanes carried ~1,400 ins/wave
(~0.011pp, ~440 waves to finish) while wave O carried 6,266 ins at the same gate cost and the same
draft rate.
build_wave_atlas --target-ins draws cards until the instruction budget is met (still capped by n)
and refuses to under-fill silently. Standard recipe: --target-ins 6500 --min-ins 60 --max-ins 200
--max-bins 4, levers now including UNKNOWN.
THE MEASUREMENT BEHIND IT: draft rate barely decays with size -- wave M 98% at avg 51 ins, wave N
92% at avg 65, wave O 96% at avg 128. Mass is nearly free.
THE UNKNOWN UNLOCK: UNKNOWN is not a difficulty label, it means the atlas could not name a lever,
and it had been routed as needing its own bespoke lane. Wave O's 22-card R37 probe drafted it like
any other lane -- reclassifying ~138k ins (a quarter of everything open) as ordinary wave fuel.
With UNKNOWN in, 9,224 fns / 417,325 ins = 70% of all open instructions are agent-draftable; the
60-200-ins mass band alone is 164,357 ins ~= 27 waves, and is the band to work first.
Also banks the five-step PRE-GATE PROTOCOL (independent re-verify -> reloc_identity -> dry-run to
0-dropped -> reconcile declarations toward the form the match needs -> gate), cookbook §176g.
R37 probe: 20 shape-verified AND symbol-verified stored drafts, 5 gate groups -> 1 banked (5%),
statistically the same as the project's A10 stored-verdict law (~0-8%; T1 measured 0/23 on the same
kind of pile earlier this phase). The null is the finding: a stored draft's rejection is almost
never symbol identity, it is TU plumbing (§176d) or staleness. reloc_identity's real home is a
PRE-GATE check on FRESH drafts, not a backlog resurrection tool. The remaining 30 groups are not
worth 30 rebuilds -- lane closed rather than pursued because the tooling was new and interesting.
Also names the SYMBOL-COLLAPSE class in the fixer's refusal (one draft extern standing in for N
distinct globals; a textual rename moves every occurrence together, so it needs one extern per
site) and records the R38 self-note: the 0/23 prior was already in the phase log.
- 48 atlas mass cards on ov_SC02_000 with --min-ins 40: 45/48 shape-verified, 42 banked,
ONE gate group. stubs 11,549 -> 11,477. distinct-code 90.4%.
- the band question is answered: 94% draft at avg 65 ins (up to 119), after 98% at avg 51 in
wave M. The mass lane is NOT size-limited in the band the instruction-weighted metric tracks.
- 2 NEARs enqueued with unusually deep analyses, both reusable beyond their own functions:
* func_80189C6C (close=2): magic 0x66666667 + mfhi-shift 5 DECODES to a plain /80 — write
'(x<<12)/80' and let gcc synthesize its own magic multiply. Residual root-caused against
REAL cc1 -da RTL dumps to loop.c move_movables desirability (threshold 29 vs measured
insn_count 24-26). Matches the func_80015A74 hard-tail class.
* func_80185840 (close=3): 'register u32 zr __asm__("$0"); c = val + zr;' reproduces the
non-coalescing addu-zero copy that cookbook §52a had classified as a WALL; plus 'r = K;'
before a SINGLE-armed if is what lets reorg's backward scan steal the li into the delay
slot (any two-arm spelling needs the eager target-thread steal, which never fires).
- 44 atlas mass cards on ov_SC04_011, avg 51 ins (up to 112) -> 43/44 shape-verified,
40 banked, ONE gate group. stubs 11,589 -> 11,549.
- data point that matters for the endgame: the mass lane holds at ~98% draft on the LARGER
band (51 avg vs the 12-42 cousins the night started with). Since the public metric is
instruction-weighted, that is the band that moves it — and it is in reach of haiku/sonnet,
not only the frontier tier.
- 1 near enqueued (func_80188A30, close=9: gcc reorders a beqz + delay-slot nop; core logic
verified correct) as grinder fuel.
- 44 atlas mass cards -> 44/44 shape-verified, 42 banked (2 dropped: a duplicate SVECTOR
typedef and a u8[] vs char[] decl conflict). main stubs 955 -> 913. fleet stubs -> 11,589.
- first wave carrying law 1c (match_one masks relocations => verifies SHAPE not SYMBOL
IDENTITY; re-check every symbol against the target .s relocation lines after MATCH).
- the compile-error shortcut paid for itself: named 'SVECTOR at src/800.c:94' and the exact
draft in seconds, where the old bisect path burned 28 minutes producing nothing.
- KNOWN NEXT IMPROVEMENT: gate_main should strip duplicate typedefs on substitution the way
harvest_verify already does — src/800.c carries a local SVECTOR from a previously banked
function, so any later draft defining its own collides. Mechanical, recurring, cheap.
- 44 atlas mass cards on main -> 44/44 standalone, 41 banked after 3 in-TU decl-conflict drops.
gate_main clean rebuild -> 143dbb89 BYTE-IDENTICAL, then full-fleet R22 213/213.
main stubs 996 -> 955; main REAL 92 -> 133 (incl. propagated/dedup credit).
- took 5 attempts; each exposed a real defect, the last one substantive:
* conflict checker too coarse (u8 D_x == u8 D_x[]) -> fixed + NC'd
* my pkill pattern matched its own shell -> stop shell process-matching
* FALSE PASS: sha() read a stale binary when the build failed -> rm output + check returncode
* bisect burned clean rebuilds; the compiler names the culprit -> read the error instead
* THE REAL BUG: func_8002A234 stored to the WRONG GLOBALS (v1->D_80078EE8/0->D_80078EE4,
target is the reverse). 2 bytes, both at %lo relocation offsets.
- STANDING CAVEAT (write this into the wave prompt): match_one MASKS jal/HI16/LO16, so it
verifies INSTRUCTION SHAPE, NOT SYMBOL IDENTITY. A draft that calls the wrong function or
stores to the wrong global passes standalone every time. Only the whole-binary gate sees it
— same class as the PsyQ symbol-name errors in waves F/G.
- first full wave against the main EXE: 40 atlas mass cards -> 39/40 standalone (98%), ALL haiku.
main drafts exactly like an overlay; the only special handling is the gate path.
- gated via the clean-rebuild batch path: substitute -> make extract BINARY=main -> make build
BINARY=main -> 143dbb89 BYTE-IDENTICAL; then full-fleet R22 213/213. main stubs 1030 -> 996.
- NEW CLASS: in-TU cross-draft declaration conflicts. Batching N drafts into ONE .c means their
externs must agree with EACH OTHER (D_800A4ED4 s16-vs-u16; func_8001C9D0 void/void*/s32).
Resolved greedily (keep-in-order, drop incompatible) at a cost of 5 recoverable drafts.
- the recovery lever, proven on func_80037368: adopt the shared header's decl VERBATIM
(extern u8 D_80076251;) and adapt at the USE site ((&D_80076251)[i]) instead of redeclaring.
- NOTE on my own tooling: my first conflict detector compared parameter NAMES and wrongly
dropped 2 good drafts ((s32 *_) vs (s32 *)); comparing type signatures only recovered them.
Second time tonight a refusal check of mine discarded good work (R39).
- wave H: 40 atlas mass cards on ov_SC02_005 -> 38/40 standalone (95%), 34 banked, ONE gate
group. R22 213/213. stubs 11,788 -> 11,751. fleet 95.4%.
- §175 (NEW): a register pin to a CALLER-SAVED reg is not a scheduling hint, it changes program
meaning. func_80182EB0: value written before a jal and read after it; the $2 pin let gcc treat
the pre-call store as dead across the call and SILENTLY DELETE addiu v0,zero,-1 (49 vs 50 ins),
post-call read = garbage. Fix was to DROP the pin and kill the cross-call live range in C.
Rule: never pin a caller-saved reg to a value whose live range crosses a jal — use a
callee-saved $s0-$s7 (safe by ABI, the §17 lever) or restructure. A one-instruction count
mismatch on a caller-saved-pinned draft is this bug until proven otherwise.
- 2 NEARs enqueued with full diagnoses (prologue $ra-save scheduling; S3 chain-priority).
- wave G: 36 atlas mass cards on ov_SC03_006 -> 36/36 standalone (100%, independently
re-verified), 32 banked of 33 gated, ONE gate group. R22 213/213. stubs -> 11,788.
- two TU-packed waves now confirm the shape: ~1 rebuild per wave instead of 23.
- law 1b added to the wave prompt: agents reconstruct CODE at 91-100% but INVENT PsyQ symbol
names (S80131E00->Square0, Blk20_...->RotMatrixY, SRM_...->RotTransSV). Dangerous because
match_one MASKS relocations -> a wrong callee name still reports MATCH; only the
whole-binary gate + symbol audit catch it, after a wasted rebuild.
- 3 wave-G drafts held by that audit (would previously have crashed it pre-commit:2330).
- wave F: 60 atlas 'mass' cards (fresh crack, NO seed body) -> 55/60 standalone pre-repair
(59/60 post-repair), 50 BANKED of 53 gated, in a SINGLE gate group. R22 213/213.
- the throughput thesis is confirmed: wave D banked 45 across 23 whole-binary rebuilds;
wave F banked 50 across 1. gate cost scales with (binary,TU) groups, not drafts.
- the fresh-crack lane converts like the seeded lanes (~91%) => the atlas's ~7k draftable
candidates are all reachable, not just those resembling a prior match.
- grinder: 0 banked across 8 seeds / 11 ILS cycles (band exhausted); re-exposed the Phase-22
split-file blindness (no .s under md_MAIN_027).
- KNOWN DEFECT logged: aprop_symfix crashes on non-hex symbols (int('Square0',16)) — one
PsyQ-named callee aborts the whole audit; needs 1:1-rename handling + per-pair isolation.
- stubs 11,876 -> 11,826
- wave D (48 adapt cards): 47/48 standalone, 45 BANKED across two gates (40 + 5 late-repair).
Operational lesson: build the gate slate AFTER the repair stage lands — the first slate was
built early and 5 rescued drafts needed a second gate.
- wave C (35: 11 tell + 24 weak): 32 banked, 91% gate. Weak lane proven 24/24 on haiku.
- MAIN BLOCKER DIAGNOSED (the night's most valuable finding): main drafts are byte-correct yet
gate 0/4. Byte-diff of the built EXE = exactly 2 bytes in 413,696, NOT in the drafted fn: a
jal at 0x80060E74 retargets func_80061FA8 (game code, 800c2.o) -> firstfile (PsyQ libapi
A66.o). Adding one C fn perturbs symbol resolution between game code and the LINKED PsyQ
archives. main is an INTEGRATION wall, not a matching wall -> its own lane; excluded from
build_wave_atlas by default.
- build_wave_atlas.py now packs by (binary, TU) — the REAL gate-group key, since each group is
one whole-binary rebuild (wave D: 42 drafts / 23 groups = the throughput ceiling).
- 3 NEARs enqueued as grinder fuel incl. func_80183578 at close=1 DELAY-SLOT (§60a precedent).
- wave C: 35 cards (11 tell + 24 weak) -> 35/35 standalone (re-verified independently, R14)
-> 32 banked / 3 near, 91% gate, 0 symbol failures (Law 4 prevention worked)
- weak lane proven for the first time: 24/24 on haiku; 890 candidates remain
- reach measured: 32 exemplars, 8 with sharers, x2 each => ~1.25x effective (the x134
era ended in P25/29/30) -> throughput, not leverage, is now the lever
- tools/build_wave.py (pool=adapt|weak, corpus-derived open-stub filter, R35 gate guard)
- 3 self-inflicted instrument defects found+fixed+NC'd (P9, recorded not buried):
pgrep self-match via shell=True; corpus.stubs() is addr->Stub not names (nearly
declared both card pools spent); a wave fired on hand-typed placeholder cards (stopped)
- STRATEGIC: card lanes are ~0.23% of open ins/wave; the Atlas's head-crack bucket is
1,276 groups / 186k ins with high-reach groups up to 265 instances -> retarget waves
at atlas groups next
- probe 3 tell-cards: 1 MATCH, 2 NEAR (both genuine compiler residuals -> grinder)
- func_80181724 gated 0/1 at standalone MATCH: the TU already declared it (s32,s32) while
the matching def wanted s16 -> conflicting prototype. Canonical sig + cast-at-use
((s16)a0) = identical 13 bytes, re-gated 1/1
- cookbook §174 Law 4 + measured lane economics (tell ~100k tok/card @33% vs adapt
64-88k/bank @95% gate) + 'a standalone MATCH that gates 0 is a declaration fact'
- 2 NEARs enqueued as permuter fuel (func_8017DAEC count-exact 113=113, role swap)
- tools/plumbing_groups.py: derives the honest still-open pool from the classified
ledgers (R38) — '1,217 PLUMBING' collapsed to 237 (SELF 109 / CALLEE 48 / OTHER
48 / DATA 32)
- recover_integration: PER-GROUP ISOLATION (git-checkout binary TUs between groups
— one TU-stage edit was poisoning every other group's whole-binary gate with a
phantom shared error; per-group banked_from_source capture) + new stages
'macro-externs' (§121 draft-tier, via family_sweep.macro_def_sig_map, R33) and
'tu-scope' (§103 STU binary-tier, the sweep-only lever)
- the probe (ov_SC03_107): raw 0/14 -> root-caused (poisoning + stale seed
symbols; rtu_match MATCHes them — blind to reloc names, R34) -> symfix-first
-> 9/14 BANKED (64%)
- sweep finding (Law 3): the no-draft majority (ov_SC02_037 44/44, most of
ov_MAIN_012) had verdicts from transient sweep remaps never persisted — family-
lane fuel, not recovery fuel; the stored-draft class is consumed
- cookbook §173 (symfix-first / per-group isolation / verdicts-without-drafts);
index 518 green; R22 clean fleet 213/213; phase total 17 banked @ 0 agent tokens
- decision-log: the P31 re-charter entry (organize-before-grind; R37/R38/R39
ratified at gate-1) per R31
- harvest_verify.py: import guard — a bare import now RAISES loud instead of
running a full gate (CLI unchanged, verified both directions)
- sig-resident: bootstrap boundary artifacts fixed (fused +0 data word with
func_800CEDFC; func_800D33E0 dropped past a glued tail) -> ELF-seeded per the
S45 pattern, exactly 145 fns; true denominator confirmed 145 (progress was
right); audit-corpus 0 PHANTOM + 0 TRUNCATED; all three oracles agree
- family maps regenerated at HEAD commit:2161: 11,025 open non-main members
reconciles EXACTLY with 12,059 - main 1,034 (102 stale phantoms cleared);
adapt cards 704, aprop cards 204 (full emission)
- main fuel-gap finding: 2,001/2,002 main stubs already have cached Ghidra-C
(only func_80049600 missing) — the roadmap '0/2,096' note was stale
- tools-health OK (dedup 2,063/0; C1 254,521/254,521; audit-digest green)
The §172a/§172b tells + repaired instruments swept over all 892 open near-misses:
- 33/95 stored drafts re-verified MATCH and banked through the whole-binary gate
(aprop_symfix caught 40/108 carrying stale seed symbols before gating — §171 at scale)
- 19/20 hand/mech fixes banked: four pure lhu<->lh s16 flips; the lhu+sltiu->lh+slti
shared-global quadruplet (D_80126B5E/B66/CB0, D_80126CB0 are s16 FLEET-WIDE); one xor-eq
rewrite; 11 per-location literal swaps (mask/threshold constants from sibling binaries)
- 1 refusal (func_8017EE78) stays as redraft fuel
Stubs 12,111 -> 12,059. Fleet 95.3% instr / 90.0% distinct / 96.68% fn-count.
Veins mapped for next waves: ~400 LEN+N drafts, 13 ambiguous-symbol, 7 multi-literal.
Audit ledger: .run/c294/audit_results.json (classifier derives from match_one's own sig).
Three carries a mechanical seed-body draft needs beyond the symbol rebase (cookbook §171b),
each found from one compiler verdict:
- DATA DEFINED INSIDE THE MEMBER'S OWN .s must be DEFINED, not externed — it vanishes with
the stub it lives in. Re-initialised with THIS member's bytes (the 0x801F1CD8 family carries
4 distinct 8-byte patterns across 42 members). Flat-byte-list initializers only; refuse the
rest rather than mis-initialise something the gate would reject unexplainably.
- SHARED TYPES the destination cannot see: MATRIX/SVECTOR live in engine_types.h, which md_*
TUs never include. `parse error before 'm1'` was the only thing between 4 of 9 members and a
bank. Carried brace-matched, vetoed by the destination.
- A POSITIONAL LITERAL MAP where imm_map_tier1 gives up: it refuses a value that also appears
at a non-differing slot (0x10 collides with the struct offsets), but the 8 differing slots
map 1:1 onto the C's call sites in order. Asserted ([C literals] == [seed slot values])
before substituting — that assert is the whole safety argument. 10/10 refused -> 9/9 banked.
- ROOT-CAUSE FIX: body_text matched `extern void func_X(...);` at column 0 and returned the
NEXT function's body — silently shipping wrong seed bodies, visible only as "no definition
after rename" skips. A definition is now confirmed by a `{` with no `;` before it.
R22 clean rebuild: check-all 213 passed, 0 failed of 213. Stubs 12,161 -> 12,111.
Fleet 95.3% instr / 90.0% distinct / 96.66% fn-count.
Final S50 state: 307 instances banked, stubs 12,468 -> 12,161, fleet 95.3% instr / 90.0%
distinct / 96.65% fn-count. R22 clean rebuild 4x, check-all 213/213 every time.
- tools/aprop_autodraft.py + tools/draft_prechecks.py: seed body + symbol_map + a MINIMAL
synthesized preamble. The seed's decl layer never travels — that layer is family_sweep's
dominant failure (331 of 458 S49 verdicts). 256 banked at zero agent tokens, against the
~20M the same work would have cost as a wave.
- Macro seeds (567 of 1196 members, all 3737 de-macroize) take the DEFINITION only; the block
stays the decl source. Pasting it whole measured 28% vs inline's 68% — func_8016AB6C's macro
is 1,891 lines of which 108 are the function.
- IMM is a second engine, not a wall: T2a's imm_map_tier1 resolves a per-location LITERAL like
symbol_map resolves a per-location SYMBOL. 131 of 275 IMM members resolve.
- draft_prechecks negative-controlled against ALL 205 banked drafts: zero false positives,
catches 39 of 67 known failures. That control found two bugs in the checks themselves —
C89 `f()` declares UNSPECIFIED parameters (not zero), and a member's own definition read as
a call to itself. Conservative by design: a pre-check that discards good drafts is worse
than one that lets a few builds fail.
- The A-prop pool is now priced exactly: PURE 437/37,376 ins, IMM 275/8,849, STRUCT 238/4,259.
- Cookbook §171a; SETUP rows; CURRENT_PHASE S50 FINAL checkpoint.
- The blocker was carried as "one missing file-scope extern gates 83 PURE members". Both
halves were wrong (R14): corpus.stubs says 4 open members, and D_801ED98C is a DEFINED
const Blk8 whose rodata lives inside the member's own nonmatchings .s — replacing the stub
deletes the data with it. gather_externs can carry an extern DECL, never a DEFINITION,
which is why it reported "no file-scope decl" for a symbol md_SC05_023 defines on line 114.
- Fix: paste typedef + const definition + body per sibling (data bytes verified identical
across md_SC05_024/025/028/029). 4/4 banked.
- aprop_symfix: new `local-only` class — draft-DEFINED identifiers that merely carry a
vram-looking suffix (Blk8_…, S8_…, L_call_…) are not stale symbols. Measured: that is every
non-clean case in the whole wave-7a/7b stored-draft residue, which holds ZERO stale-symbol
recoveries (a clean negative result — the defect was A-prop-specific).
- cookbook index regenerated (tools-health fails closed on a stale index — it caught §171).
- R22 clean rebuild: check-all 213 passed, 0 failed of 213. Stubs 12,445 -> 12,441.
- REFUTES §170's open hypothesis (batched cards concentrate members into one TU ⇒ §169
collision): 5-draft groups banked 5/5; 11 of 35 unbanked drafts were already one-per-TU;
and the two "concentrated" groups banked 12/12 and 10/10 once the real defect was fixed.
- The cause: a per-location data symbol carried out of the seed body unrebased. match_one
compares instruction ENCODINGS and is blind to a relocation's target NAME, so it scores
MATCH standalone and dies at link in the host TU. 24 of 24 concentrated failures, all 1:1
rewritable at one constant vram delta (0x4128).
- tools/aprop_symfix.py: audit + --fix, emits a gate_lane-shaped slate; deterministic and
build-free, so it runs BEFORE the gate. The R34 second oracle for the class match_one
cannot see.
- family_cousins.py --aprop-cards: members now carry sym_map, the explicit {seed -> member}
renames, read from the seed's C BODY (a matched seed has no .s of its own) vs the member's
.s. Two case-mismatch defects fixed while wiring it (sig lowercase vs splat uppercase).
- 23/24 banked. Stubs 12,468 -> 12,445. Fleet 95.2% instr / 89.9% distinct / 96.57% fn.
R22 clean rebuild: check-all 213 passed, 0 failed of 213. dedup 2,043/0.
- A-prop's true conversion is 87% (79/91); the 320 batched members are unblocked.
- Cookbook §171 + §170 struck in place; SETUP row; decision-log (R31).
- NEW family_cousins.py --aprop-cards + tools/wave/aprop_wave.js: lane A (1,700 open fns /
76,419 ins) had NO card type — cousin diffs are empty for h_seq-identical members, so the card
is a positional WORD diff vs the matched sibling, grouped BY FAMILY (one agent, N drafts).
Head cards: 13 families / 433 members, median TWO differing words each.
- calibration 9 batches / 108 members: 98 agent-MATCH (91%, best of any wave) -> 56 BANKED (57%),
~80k tok/banked fn vs 157k (cousin card) vs 400k+ (crack wave). R22 213/213 BYTE-IDENTICAL.
- HONEST GAP (R14): 91% agent -> 57% gate is the worst conversion measured; 14 groups banked 0.
Hypothesis TESTABLE not proven — family batching concentrates members per destination TU, the
§169 collision. Re-gate unbanked ONE PER TU before scaling the remaining 320.
- >=16 head diagnosed: 3 of 4 blockers are plumbing — the --band substantial default hid 5 of 13
families from every prior sweep; one missing file-scope extern (D_801ED98C) gates 56 PURE
members; dedup_extend is macro-only. Only func_8017C294 is a genuine crack.
- fleet 96.56% fn / 95.2% instr / 89.9% distinct; stubs 12,535 -> 12,468; dedup 2,043/0.
- cookbook §170.
- thresholds relaxed to <=6 blocks/<=16 tokens UNION edit-fraction <=0.20: cards 518 -> 721,
MIXED 310 -> 50 skeletons; the 753-ins func_8017BEBC (0.987 sim) became reachable.
- 59 cards -> 48 agent-MATCH (81%) -> 44 BANKED (92% MATCH->bank, 75% end-to-end), 6.9M tok.
- FINDING (the actionable one): 7b's bank rate crushed 7a's because it SPREAD 48 drafts over 35
destination TUs; 7a's failures were per-TU declaration collisions between sibling drafts.
Cookbook §169 updated with the spread law.
- R22 213/213 BYTE-IDENTICAL from clean; fleet 96.55% fn / 95.2% instr / 89.9% distinct;
stubs 12,584 -> 12,535; dedup 2,035/0.
- incidents 3 & 4 recorded: an agent wrote a TRACKED header (guard caught it, prose is not
enforcement); my own gate_lane filtered on the wrong key and printed 'gating 0 drafts' as a
result (R32 silent skip) — fixed with a coverage assertion that refuses to report 0.
- pilot 30 cards -> 25 agent-MATCH (0 refuted) -> 16 banked; 29 instances banked tonight
(89 incl. propagation); 2.7M tokens haiku-tier ~= 30k/banked instance vs a crack wave's ~75k.
- R22 213/213 BYTE-IDENTICAL from clean; fleet 96.53% fn / 95.1% instr / 89.8% distinct;
stubs 12,613 -> 12,584; dedup 2,029/0.
- R14 CORRECTION: a banked cousin usually does NOT propagate (2 of 8; cousins are byte-variant).
The card 'reach' column is cousin fuel, not dedup copies — priced wrong in my earlier framing.
- FINDING: the 9 gate failures are per-TU INTEGRATION (standalone-MATCH, host-TU-rejected),
clustered 5+2 in two binaries — the reconcile-ladder class, not codegen.
- TWO INCIDENTS (mine): an outer timeout tighter than gate_stage's own scaled timeout killed a
healthy 5-bank group mid-write AND orphaned its dedup_propagate child, which kept rewriting
src/ through a git checkout. Killed, inspected, reverted; the same 5 drafts banked 5/5 untimed.
Law: never wrap a self-timing tool in a tighter cap; kill process GROUPS, not pids.
- cookbook §169 (the lane + the three laws + the threshold sizing table).
Wave 6 added 110 (24 cracks + 23/24 families propagated). Fleet 95.1% instr /
89.7% distinct / 96.51% fn-count. R22 clean-fleet run 9x this session, 213/213
every time. Bank rate across six waves: 67/79/69/68/73/60%.
Records §166a (the destination-TU oracle) and the four-instance pattern it
completes: a tool asserting a conclusion it never reached. A confident wrong
label costs more than a missing one.
Wave 5 added 152 (29 cracks + 28/29 families propagated) — the session's
largest. Fleet 95.0% instr / 89.6% distinct / 96.48% fn-count; stubs
13,345 -> 12,771. R22 clean-fleet run 8x this session, 213/213 every time.
P30's milestone is '>=95% instr fleet, or every remaining overlay stub on a
named ledger'. THE FIRST HALF IS NOW MET — T5 (phase close) is a live option.
Bank rate across five waves: 67% -> 79% -> 69% -> 68% -> 73%.
Wave 4 added 105 (27 cracks + 27/27 families propagated). Fleet 94.9% instr /
89.4% distinct / 96.44% fn-count. R22 clean-fleet run 7x this session, 213/213
every time.
Bank rate now measured four times: 67% -> 79% -> 69% -> 68%. Prior-notes
seeding 10/12 (was 7/9). func_8017C294 — the x16 family, largest item on the
board — is NEAR at 2 ins after three seeded attempts (18 -> 11 -> 2).
Also records the 4th comment-blindness defect and its blast radius (one draft
comment refused a binary's stub oracle, failed 5 later binaries, and left
drafts spliced in src/ so 17 re-gates read a poisoned tree as 0/17), and that
the wave harness now lives in tools/wave/ with its contracts written down.
Wave 3 added 84 (27 cracks + 21 propagated families). Fleet 94.8% instr /
89.2% distinct / 96.41% fn-count. R22 clean-fleet run 6x this session, 213/213
every time. 96 commits.
Bank rate measured three times: 67% -> 79% -> 69%. The dip is the cost curve
(wave 3's tier was 29 Opus-band / 14 jr vs wave 2's 8 / 5, median reach x6 ->
x3-4), not a regression.
Two levers proved out and belong in every future wave: the hardened harness
contract (0 drafts lost vs 21) and prior-notes seeding (7 of 9 previously
failed targets converted, incl. both long-standing NEARs and all three wave-2
gate misses). NEAR is a resumable state, not a write-off.
Session close state. Three parts: stage 0b (91, zero decompilation), wave 1
(26), wave 2 (116). Fleet 94.4% -> 94.7% instr, 88.3% -> 88.9% distinct,
13,345 -> 13,112 stubs. R22 clean-fleet run 5x, 213/213 every time.
The campaign now has a MEASURED rate, twice: 67% (wave 1, all-Opus) then 79%
(wave 2, 20 of 28 Sonnet) of cracks survive the whole-binary gate. The Sonnet
band beating the all-Opus wave is the session's most useful economic finding
and sets wave 3's routing.
Resume order changed on evidence, twice over:
- harden the wave harness FIRST (per-agent dirs, sha1-last verifier, and a
tools/recover_drafts.py built from the transcript-replay method that
recovered 21/21 today);
- then wave 3, sized on 79%, not on the reach-15 prior.
Error ledger grew to 6. The two that matter: I wrote off 21 verified cracks as
lost when the run transcripts held every one of them, and my first two
recovery passes both failed by reading a single tool record instead of
replaying the file's mutation history.
Stage 0b closed (91, zero decompilation) + Stage-1 wave 1 (8 cracks -> 26
instances). Fleet 94.6% instr / 88.8% distinct / 96.36% fn-count.
Resume order changed on measured evidence: FIX THE md_ MODULE LANE FIRST.
16 of the wave's 42 member slots were unreachable for tooling reasons, not
matching reasons — 12 on a carve that assumes raw data lives in
<binary>/data/*.data.s (modules do not), 4 on an uncarried extern
(`D_8011511A' undeclared). Both are named with verbatim errors; probe one of
each before pricing (R37). Precedent: 0b's three repairs banked 91 for ~0
agent tokens; the wave spent 3.36M for 26.
Also recorded: the frontier re-derivation (1,955 zero-crack families /
330,622 templatable ins), the tier-ordering correction (ins-per-crack is flat
across x5-x8, so rank by templatable weight, not by tier), and the §162
harvest with its two in-place cookbook corrections.
Stage 0a's first defect, and the largest single zero-token bank of the session.
family_remap's kind test asked ONE question — is this address a function in the SIBLING'S OWN sig?
— and defaulted to `D_` on "no". But a body calls outside its image constantly: an overlay calls
resident helpers, an md_* module calls the overlay-range engine. Those addresses are absent from
the sibling's sig, so the test fell through and emitted a DATA NAME FOR A FUNCTION —
`D_800183E0`, `D_800D1EBC`, `D_80171A1C`. None exist anywhere in src/ or config/symbols.us.txt,
while `func_80171A1C` alone has 1,061 references. Measured: 611 member-rows across 45 symbols,
the largest named residue class. "Not in MY sig" means "not mine", not "is data".
Fix — three oracles, strongest first, never a blanket fallback:
1. the sibling's own sig (authoritative for its image; this is what preserves the Phase-29 T82
case where a slot is a function in the exemplar and DATA in the member — unioning every sig
would have re-broken the 251 members T82 fixed),
2. the always-linked images via extern_fn_addrs() — resident + main, 2,146 addresses whose ranges
cannot collide with an overlay's,
3. the exemplar reached it by `jal` — a call target is a function BY DEFINITION, which covers an
external address neither sig claims (0x80171A1C from an md_* module, 112 rows).
Only a non-call reloc no oracle claims still falls to `D_`.
Result: BANKED 205 member-matches, failures 670 -> 575, derived net = report = 205.
R22 clean-fleet 213 passed / 0 failed of 213.
Fleet 94.4% instr / 88.3% distinct / 96.27 -> 96.33% fn-count; stubs 13,563 -> 13,345.
THE RESIDUE HAS CROSSED OVER: DIFF is now the LARGEST class at 143 of 575 — real byte divergence
outranks plumbing for the first time this session (undefined-ref 611 -> ~8, PLUMBING-other
231 -> 81). The ~5:1 plumbing:DIFF ratio that justified "tooling beats volume" has inverted in this
queue, exactly as the frontier analysis predicted: the declaration-axis vein was one-time.
The new head class is `conflicting types for func_80175414` (27) — the same addresses this fix
started naming correctly, now surfacing the NEXT layer (the symbol resolves; its declared signature
disagrees). That is the conform axis, not the remap axis.
Note for anyone auditing this class: rtu_match MASKS HI16/LO16, so a wrong %hi/%lo symbol still
reports MATCH (the T82 comment records `MATCH (10 ins)` on a member the fleet gate refused). This
defect is invisible to the per-function tool by construction — only the whole-binary gate sees it.
W1b — the 3 targets whose agents died on API rate limiting, retried with cookbook §160 in the
prompt: func_801EFBF4 (reach 12), func_801EFDC8 (12), func_8018CC40 (10, jr). 3/3 confirmed by an
independent verifier, all banked, R22 clean-fleet 213 passed / 0 failed of 213.
func_8018CC40 failed the first gate with `too many arguments to function func_80178970` — which its
own crack agent had PREDICTED in its report, naming the §17a-1 remedy. Dropped the draft's
empty-paren externs and cast 6 call sites instead; banked. Read the agent's integration notes
before diagnosing a gate failure — it has already seen the TU.
Cookbook §161a-c (index 469 sections):
§161a case 0: break; is LOAD-BEARING when a jump table is indexed from zero. The natural
case 1..5 makes gcc-2.7.2 pick minval=1, emit `addiu $v1,-1`, and shift every table index —
58 of 77 mismatched on a byte-perfect body. Tell: the table's FIRST entry points at the
function's own end address. Family-wide (10 members).
§161b aliasing a parameter into a local can force a SECOND callee-saved register (+8 frame,
+3 ins) even when uses are mutually exclusive. Suspect it before reaching for register pins.
§161c loose-prototype engine helpers: don't fight the TU's (void) decl, cast at the call site.
G2 — THE MAIN EXPERIMENT. family_hseq excludes main as "structurally barren — zero h_exact
overlap". True and irrelevant: an h_exact claim guarding an h_seq tool. There is not even a
sig-main target — main had never been signed for this pipeline. Signed it (2,002 fns, seeded from
splat boundaries via corpus.stubs rather than --bootstrap, which glues functions around jtbl
dispatch and would have corrupted the hashes under test).
Result: main is ~85% singleton work, not 100%.
internal h_seq families (>=2): 207 families / 748 fns / 11,537 ins (13.7%)
shapes shared with the fleet: 161 fns / 1,346 ins (1.6%)
genuine x1 remainder: ~71,034 ins (84.6%)
IMMEDIATELY ACTIONABLE: 44 classes / 151 main functions / 1,239 ins already have a matched exemplar
in the fleet — free propagation, invisible only because main is not in the map.
Long-term: 748 of main's 2,002 functions (37%) are templatable once one exemplar per family is
cracked, which refutes "2,002 independent cracks" as the planning assumption for the 79k-ins tail.
OPEN, deliberately not done unilaterally: adding a sig-main target and dropping main's exclusion
from family_hseq.load() changes a fleet-shared oracle every targeting tool reads. Needs Drew's call.
func_801EDC18 (md_SC05_023) is the largest multiplier remaining — 57 members. The wave agent
abandoned it at "closeness 6" with class SIZE-MISMATCH [redraft]. It was two lines from correct.
THE CODE (cookbook §160a): the target copies 8 bytes with lwl/lwr + swl/swr — gcc-2.7.2's
emit_block_move for a type with ALIGNMENT 1. The draft used a u32 copy (aligned lw/sw), which is
wrong by construction. `typedef struct { char c[8]; } Blk8; buffer = D_801ED98C;` reproduces it.
Six spellings were tried in parallel; two independent agents converged on the same one.
THE INSTRUMENT (§160b) — this is the part worth more than the function. The target .s bundles a
leading `.section .rodata` block (D_801ED98C as two .word) ahead of .text. Those lines carry the
same `/* off vaddr HEX */` shape as instructions, so masked_diff.insns_from_s counted them as TARGET
instructions, while insns_from_object (objdump -j .text) can never emit them. A byte-perfect draft
therefore read `mine=26, target=28, 26 mismatched` — every position shifted by a constant +2 — and
got classified as needing a redraft. 116 of 12,583 .s files in the corpus have this shape, one at
-29 instructions. Every one of them would report a false wall to any agent that tried it.
Fixed: insns_from_s tracks .section and counts only .text. Full-corpus control: 12,467 unchanged,
116 corrected, 0 regressions. Same artifact class as §129a (post-carve jtbl inflation).
THE OWNERSHIP LAW (§160c) — my own error, corrected by the gate. Four sites declare
`extern short D_801ED98C;` and nothing in src/ defines it, so I shipped an extern-only draft. The
gate refuted it: `undefined reference`. The .s block the draft REPLACED was the definition. The
variant emitting `const Blk8 D_801ED98C = {{...}}` banks clean. Never infer ownership from externs.
R22 clean-fleet: check-all 213 passed / 0 failed of 213.
ALSO BANKED — the wave's idiom harvest, which had been sitting unwritten in workflow transcripts
(R16/R30 debt): §160d the ASYMMETRIC INDEX RELOAD (a just-stored narrow field read twice emits
reuse-then-reload; the C is deliberately asymmetric), §160e a stack-layout scheduling rule now
byte-proven on a SECOND independent function (promoting it from coincidence to rule), §160f the
address-only global store via array decl, §160g sibling-search keyed on the CALLEE SET as step 0 of
every wave prompt (one grep turned a 126-instruction crack into a copy-edit).
Cookbook index regenerated: 468 sections.
The 10 exemplars from W1 flipped modal -> matched in the regenerated map, so family_sweep could
template them. 9 non-jr families swept: BANKED 140 member-matches / 32 failed across 50 overlays.
Derived net = report = 140 (no untracked carve files this time, so the two counts agree).
R22 clean-fleet: check-all 213 passed / 0 failed of 213.
Fleet 94.3 -> 94.4% instr / 88.2 -> 88.3% distinct / 96.22 -> 96.27% fn-count; stubs 13,713 -> 13,563.
WAVE ONE, FULLY ACCOUNTED: 10 agent cracks + 140 templated members = 150 functions for 1.36M
tokens (~9k tokens/function). Still owed from this wave: 73 member-slots in 2 NEAR families,
34 in 3 rate-limited targets, 9 in the jr family (routes to jtbl_family_bank, §53).
TWO MEASUREMENTS THAT CORRECT MY OWN FORECASTS (R14):
1. Conversion was 81%, not the 58% I projected from this morning's propagation run. Today's
plumbing fixes (alias-drop, cpp-derived TU type map, group-level draft-vs-draft aliasing) are
paying off in a population they were not tuned for.
2. The effective multiplier was 15x, not the 2-3.5x I predicted. That estimate used the MEAN
family size across the whole zero-crack pool (3.55); this wave deliberately targeted the TOP of
the reach distribution, where families run 10-28 members. Ordering waves by reach is what
produced the difference — the mean was the wrong statistic for a wave that selects on the tail.
The regen step is load-bearing and now byte-proven twice: a fresh crack reads as `modal` until sigs
+ family_hseq are rebuilt, and family_sweep templates only from `matched`. Skipping it sweeps a
stale map and the multiplier evaporates (the Phase-26 finding, whose surviving qualifier is that
remap works BEHIND a fresh crack).
The one-line fix committed ahead of this run (CdFileLoc_80128C98 aliasing CdFileLoc) cleared the
largest remaining propagation-sweep class. Re-sweep: 138 member-matches banked, failures 875 -> 737,
`conflicting types for cdFileLocTable` gone entirely (136 -> 0).
Derived net (138 INCLUDE_ASM removed, 0 re-added) equals the report's 138 — they agree.
R22 clean-fleet: check-all 213 passed / 0 failed of 213.
Fleet 94.2 -> 94.3% instr / 87.9 -> 88.1% distinct / 96.15 -> 96.21% fn-count; stubs 13,780.
Residue reclassified — no symbol dominates any more: 227 PLUMBING-other, 125 DIFF (real byte
divergence, 17%), 93 CC1-FAIL(no-diagnostic), 26 memcpy, then a tail of small data-symbol
conflicts (D_80114F24 12, D_800AE620 11, D_800183E0 9, D_80126B58 6, D_80078EB4 6).
CC1-FAIL rose 77 -> 93 and that is NOT a regression: members that previously died earlier on the
cdFileLocTable conflict now reach a different compile error. Those 93 are hard gcc errors whose
text the sweep's classifier discards because it greps for `error:`, which gcc-2.7.2 never emits on
hard errors. That classifier is now the highest-value instrument fix left — three times today a
no-diagnostic verdict concealed something cheap.
scope_data_externs §8d drops the draft's decl of any symbol the TU already declares at file scope.
It keys on the SYMBOL, but a §37 asm-label ALIAS binds a DIFFERENT C identifier to that symbol:
the TU declares `D_801851BC`, it does NOT declare `tbl_D_80187044`. Dropping the alias left the
body referencing an undeclared name, which cc1 reports with no `error:` prefix — so the sweep
classified all 132 siblings as CC1-FAIL(no-diagnostic), i.e. as a codegen wall.
The bitter part: the alias exists PRECISELY BECAUSE the TU declares that symbol with a conflicting
type (a `void (*[])(void)` dispatch table vs this function's 20-byte-stride view). The drop rule
fired on exactly the declarations written to survive it. Why 1 of 2 died was fully determined:
tbl_D_80187048's symbol is not in the TU, so it demoted normally.
Fix: is_asm_alias() — an alias is demoted into the body, never dropped (the identifiers differ, so
it cannot collide with the TU's decl). Control-tested 6 ways incl. self-labels and plain externs.
Measured: func_80132018 3/135 -> 135/135; full re-sweep +16 more. Total +148 members.
R22 clean-fleet 213 passed / 0 failed of 213. tools-health OK, dedup-check 1949/0.
Fleet 96.11 -> 96.15% fn-count, 87.8 -> 87.9% distinct; stubs 14,120 -> 13,972 = -148 (2nd oracle).
CORRECTION TO MY OWN CLAIM (R14): after the probe I said the 58% aggregate was concealing a broad
problem. The re-sweep refuted it — only 16 more banks fleet-wide. The alias class really was one
family; the first read ("outlier") was right and the correction was wrong.
875 sweep failures classified: 231 PLUMBING-other, 141 DIFF (real divergence, only 16%),
136 `conflicting types for cdFileLocTable` (ONE symbol — biggest single class left),
77 CC1-FAIL(no-diagnostic), 26 memcpy, 12 D_80114F24, 11 D_800AE620, 9 D_800183E0.
STILL UNFIXED, and the most dangerous instrument left: the sweep's failure classifier greps for
`error:`, which gcc-2.7.2 never emits on hard errors. Every hard error therefore reads
CC1-FAIL(no-diagnostic). That is how a missing declaration looked like a codegen wall across 132
functions. rtu_match was fixed for this at T0(b); this classifier was not.
family_sweep --hseq --band all -j 8 over every matched-exemplar family: 553 families /
203 overlays / 1,419 banked / 1,023 failed (58%). R22 clean-fleet 213 passed / 0 failed of 213.
tools-health OK, dedup-check 1949 validated / 0 failed.
Fleet: 93.9 -> 94.2% instr / 87.2 -> 87.8% distinct / 95.72 -> 96.11% fn-count.
Second oracle (R34): INCLUDE_ASM stubs 15,542 -> 14,120 = -1,422, equal to the diff-derived net
(1,451 removed - 29 re-added = 1,422 = 1,419 sweep + 3 probe). Three independent counts agree.
B -> C -> P IS ONE CHAIN, NOT THREE WINS. 1,102 of the 1,422 landed in ov_SC02_037 (409),
ov_SC03_107 (364), ov_MAIN_012 (329) — the three newly-onboarded binaries from C, which had never
been wired into the shared-body ecosystem, so every matched exemplar was unreachable from them.
B fixed the declarations, C wired the include, P poured through the opening. A repeat sweep will
NOT pay like this; the opening was one-time.
S47 total: 1,481 functions banked with zero agent drafting, all from removing plumbing.
Two findings recorded, neither fixed (deliberate, costed):
- --band defaults to `substantial`: the first probe returned a confident {"families": 0,
"banked": 0} on a real 135-member `mid` family. Always pass --band all.
- The alias-gather defect: probe on 0x80132018 banked 3/135, all 132 failures classified
CC1-FAIL(no-diagnostic) because gcc-2.7.2 emits no `error:` prefix. Real error is
`tbl_D_80187044' undeclared` — the exemplar declares TWO §37 asm-label aliases and uses both,
family_remap carried one. T7-S1's "gather" class. Measured as an OUTLIER (aggregate 58%),
which is why the sweep ran before the fix.
Refused by design, all named: 50 jr families / 183 member-slots (§53 interlock — it printed its
own coverage and reason), 264 STRUCT, 112 unresolved immediates, 3 not-stub.
Regenerated after the S47-B/C banks (family_hseq.py + report): docs/family-hseq.md,
docs/progress.fleet.md, docs/backlog.md. Numbers only — no analysis change.
Frontier at this HEAD (overlays only): 7,085 families / 14,508 instances / 752,073 ins.
with siblings (>=2): 2,429 fams / 9,852 members / 467,634 ins (62.2%)
- matched exemplar (propagate, ~0 tok): 460 fams / 2,861 members / 125,630 ins
- zero-crack (crack 1 -> templates to N): 1,969 fams / 6,991 members / 342,004 ins
singletons: 4,656 fams / 4,656 members / 284,439 ins (37.8%)
- matched exemplar: 143 / 6,603 ins - zero-crack (pays x1): 4,513 / 277,836 ins
Structural: the x138 era is over — 3 fleet-wide families remain and ALL 3 already have matched
exemplars, so no fleet-wide CRACK is left, only propagation. 82% of remaining code now sits in
the two worst cost profiles (x2-9 zero-crack 45.5%, singleton zero-crack 36.9%).
- BANKED: 11 functions at 400-952 ins from the cascade (func_8017D898 952, func_8017CE58 733,
func_801902EC 673, func_8018C2D8 673, func_8018A8D4, func_8017C6F4, func_800CBB38,
func_800CF3A4, +3). check-all 213/213 from a clean tree. 6 near = jr/switch (§53 separate
banking step), 1 failed. The cascade agents wrote 6 new cookbook sections incl. §158.
⚠️ tools-health UNVERIFIED at commit (stale cookbook index fixed, confirming re-run
interrupted) — run it first next session. check-all is the byte oracle and it is green.
- WASTE PREVENTION (Drew: "prevent this from ever happening again, however you need to"):
* tools/validate_targets.py (NEW) — names 5 defect classes (NO-ASM / MID-BODY /
OUT-OF-RANGE / ALREADY-DONE / NO-BOUNDARY), exits non-zero.
* WIRED INTO wave_snapshot so it fails closed — every wave passes through there for its .s
files, so no path from target list to spawned agents bypasses validation. Negative-control:
a 3-target bad list is refused with the exact mid-body offset (+72 bytes of 100).
* The cascade `done()` predicate now short-circuits on SKIPPED as well as MATCH. It tested
only MATCH, so a non-existent target fell Sonnet -> Opus -> Fable and three agents each
proved the same phantom absent: ~29 invalid targets x 3 tiers = 87 of 119 agents, ~9.7M
tokens. A tier that cannot act must END the pipeline, not escalate emptiness.
* docs/accelerators.md A9, including that wave_snapshot's own R32 assertion REFUSED that list
(24 of 57 found) and was routed around — the one instrument warning that was right and ignored.
- B RE-SCOPED (S46-10) and deliberately NOT done: the extend blocker is INTRA-HEADER, not
target-side. engine_core.h declares memcpy FOUR incompatible ways across its DEFINE_ macros;
two in one TU collide. NOT a safe cleanup — the in-tree note at ov_MAIN_012.c:14333 records
that `extern memcpy` disables gcc's builtin and turns an inlined block-move into a CALL, so the
declaration CHANGES CODEGEN. Probe one macro in one binary and byte-gate before any sweep.
- C (dedup_extend over the 129) stays blocked on B. Full context for both in the checkpoint.
Drew T