mirror of
https://github.com/Druthulu/BFM-decomp
synced 2026-10-03 08:07:25 -04:00
Drew T
d53c4bf796
Closes the gap that made main red for nine hours. R42 ('commit a dirty tree rather than
revert it') is correct for a per-binary gate that leaves PROVEN banks uncommitted, and WRONG
for gate_main, whose substitution is unverified by construction until the SHA matches.
Two guards, defense in depth:
1. gate_main installs atexit + SIGTERM/SIGINT/SIGHUP handlers that revert its own substitution
unless a bank actually succeeded. Killed mid-run, it now cleans up after itself.
2. ox_campaign's dirty-tree commit REFUSES top-level src/*.c (main's sources), reverting those
and committing the rest. Verified: src/800c.c and src/800.c refused, src/ov_*/... and
src/shared/engine_core.h still commit.
Also versions the autonomous lane scripts under tools/lanes/ — they lived only in gitignored
.run/, so a fresh clone had no drafter, gater, maintenance or stallguard at all.
39 lines
2.4 KiB
Bash
39 lines
2.4 KiB
Bash
#!/usr/bin/env bash
|
|
# S58 dual-pool campaign. Two provider pools drafting at once, ONE serial gate.
|
|
#
|
|
# DOUBLED 2026-08-23 (Drew): every 429 ever recorded here is `upstream_provider_shared_pool` — the
|
|
# PROVIDER's pool, never OpenRouter refusing this account. A shared pool saturates because of
|
|
# everyone on it, so those 429s measure the POOL's total load and say nothing about OUR ceiling,
|
|
# which has therefore never been tested. 128 workers produced 1,836 requests and ZERO 429s.
|
|
#
|
|
# 640 workers (ox 512 / deepseek 128) is the MEASURED ceiling for this box, not a round number:
|
|
# * one shard's real RSS is 31 MB (import chain + the full card file), and WSL sees 30 GB
|
|
# available -> ~685 agents at 70% headroom. The host's spare ~50 GB is NOT ours; WSL2 gets
|
|
# only what .wslconfig allocates (32 GB here). Raising that is the route to 1000+.
|
|
# * --cards-per-wave caps concurrency independently: shards are dealt targets[i::workers], so
|
|
# a 220-card wave feeds exactly 220 shards no matter how many we start. Hence 1400 cards.
|
|
# The 120-400 band holds 442 groups and 400-2000 holds 37, so those waves self-limit.
|
|
# The ledger records free memory per wave: if the box swaps, throughput collapses for a LOCAL
|
|
# reason that looks exactly like a provider ceiling (R40 - exonerate the instrument first).
|
|
#
|
|
# COMMENTS MUST STAY ABOVE THE COMMAND. A `#` line between backslash-continued argument lines
|
|
# swallows every remaining argument and the run falls back to argparse DEFAULTS (8 waves / 24
|
|
# workers / no lanes) while looking completely normal. `bash -n` does NOT catch it — the construct
|
|
# is syntactically valid — and grepping the file finds the arguments even though they never reach
|
|
# the program. The only reliable check is the startup banner, which prints what it actually got.
|
|
set -u
|
|
cd /home/musashi/bfm-decomp
|
|
PY=.venv/bin/python
|
|
say(){ echo "[$(date +%H:%M:%S)] $*"; }
|
|
|
|
say "dual-pool campaign starting"
|
|
$PY tools/ox_campaign.py \
|
|
--waves 40 --workers 640 --max-workers 1024 --step 128 --sustained 3 \
|
|
--models 'stealth/ox-alpha:512,deepseek/deepseek-v4-flash-0731:128:REASON_EFFORT=high' \
|
|
--lanes 'default:,tells:extend-tell;swaprepeat-tell;s16-div-tell' \
|
|
--cards-per-wave 1400 --gate-jobs 12 2>&1
|
|
|
|
say "campaign done; serial idiom lane (jtbl-carve, ov_* only)"
|
|
MAX_429=10 $PY tools/idiom_serial.py --lever jtbl-carve --n 8 --min-ins 60 2>&1
|
|
say "ALL DONE"
|