26 Commits

Author SHA1 Message Date
bench
f74fe00aaf hbm/bc 模拟器对比改为 agent 按需开启,只作诊断
阶段配置新增 verification_parameters.hbm_bc_samples(0–16,默认 0)。agent 用
bench stage run hbm --config <yaml> 指定 N 张校准样本,用 x86 模拟器对比 HBM 与 BC,
结果写在 G1.hbm_bc_cos_min,不改 verifier_ok:开了不会被扣分,真正把关的仍是 08-eval 板端实测。
编译配置只取三段,所以同一个 yaml 拿去编译也不受影响。适用任何有 .bc 的后端;跑失败记 None。
s06_hbm 随阶段打包,部署后马上生效(默认不跑);配置校验在 stageworker、说明在 SYSTEM.md,
都要下次重启时一起上线,否则按说明提交会被旧校验拒绝。

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-14 12:48:17 +08:00
bench
f1f44c9d1e fix: 调度程序因读整个 events.jsonl 被 OOM 杀;message_update 不再逐 token 复制整条消息
Pi 每个 token 发一次 message_update,里面 message 和 assistantMessageEvent.partial
各带一份到目前为止的整条消息,events.jsonl 每条轨迹 300–800MB(99.9% 是它)。
budget_used 每次 read_text() 整个文件,调度程序峰值 5GB,10-11 07:08 在 10GB 的
Mac mini 虚拟机里被内核 OOM 杀掉。下游只读 message_update 的 type/delta,
完整文本在 message_end。budget_used 改为逐行读。

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-11 09:47:41 +08:00
5617f57f08 fix: let a long bash command time out alone instead of ending the attempt
Bash has no default timeout, so any command running past the 900 s system
cap ended the whole trajectory (81 of 590 finished rollouts). Give bash a
native timeout one minute before the cap; only that command fails and the
agent continues. The persisted tool call keeps the model's own arguments.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-08 07:34:41 +08:00
106a96ae4a feat: raise the per-trajectory compile cap to five
Historical r20260929-04/05/06 data: a cap of 3 would have lost 19 of 142
PASS trajectories (13%); a cap of 5 loses 2 (1.4%). Prompt states the same
number; agent/VERSION 13 -> 14.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-04 22:45:26 +08:00
4332f43dab feat: cap each trajectory at three started compiles
Compile attempts are now the primary per-trajectory budget: they map to
the real compute cost, unlike turns or tokens, and make the stop reason
obvious. stagerun refuses a 4th 05-compile with COMPILE_BUDGET_EXHAUSTED
before inline calibration or slot acquisition.

- Counted from the stages ledger (05-compile rows that got a slot), so
  resume/recovery cannot reset it; a compile that timed out in the queue
  is free; recompiles after an accuracy miss count.
- Board/eval/review of the last artifact still run after the cap.
- Turn/token/wall budgets, per-phase deadlines, cleanup and API quota
  protection are unchanged as backstops.
- e2e prompt drops "no retry limit"; SYSTEM.md names compiles as the
  primary budget; agent/VERSION 12 -> 13.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-04 22:37:15 +08:00
13f1e99aee fix: distinguish empty documentation searches from tool failures 2026-09-29 18:29:59 +08:00
e1e73816af fix: recover stage registration and reclaim completed local leases 2026-09-29 12:55:35 +08:00
Codex
9af7952ab5 fix: consume tool boundaries under lock and diagnose transport errors 2026-09-23 22:00:17 +08:00
Codex
19e8cabef9 fix: point stage task guidance at managed waits 2026-09-23 21:58:49 +08:00
Codex
b81bc80875 fix: reuse managed waits and drain accepted stage work 2026-09-23 21:57:25 +08:00
f611ef29ab fix: keep managed API requests within the broker lifecycle
Pi requests could hit Node fetch's independent five-minute header deadline,
and queued requests could outlive their local client or terminal rollout.
Set header/body deadlines only for the loopback broker origin, cancel scopes
when clients disconnect or executions terminate, and retain uncertain upstream
reservations instead of retrying after cancellation. Resumable pauses remain open.

Validation: real Pi SDK accepted a response delayed 310 seconds (314.139 s test);
quota, Pi session, acceptance, cancellation, and lifecycle regression coverage.
2026-09-22 20:28:42 +08:00
Codex
068c56043f Merge managed natural-agent runtime into retained benchmark mainline 2026-09-22 15:35:28 +08:00
bench
d309109f2b pre-grind snapshot 2026-09-19: batches 1-7 adapters/probes/recipes + eval additive changes 2026-09-19 17:52:03 +08:00
bench
cd220cc81e fix: preserve deployment outcomes and export completed run data 2026-09-16 07:55:44 +08:00
bench
864c20a087 fix: regenerate configured repairs and use glm-5.3 for collection 2026-09-16 02:32:06 +08:00
bench
4a0e7d521f fix: run S100P board accuracy through the existing pipeline 2026-09-16 02:17:36 +08:00
bench
3dce78e49d sync 2026-09-15_12:38:10 d-robotics@d-roboticsdeMacBook-Pro 2026-09-15 12:38:10 +08:00
bench
5bafe1d7f4 sync 2026-09-15_12:26:22 d-robotics@d-roboticsdeMacBook-Pro 2026-09-15 12:26:22 +08:00
bench
50edd8c2a1 sync 2026-09-15_12:21:38 d-robotics@d-roboticsdeMacBook-Pro 2026-09-15 12:21:38 +08:00
bench
d191b68b88 sync 2026-09-15_11:26:25 d-robotics@d-roboticsdeMacBook-Pro 2026-09-15 11:26:25 +08:00
bench
6b443c5b6f sync 2026-09-15_09:49:07 d-robotics@d-roboticsdeMacBook-Pro 2026-09-15 09:49:07 +08:00
bench
9d1741b9fa sync 2026-09-15_09:42:56 d-robotics@d-roboticsdeMacBook-Pro 2026-09-15 09:42:56 +08:00
bench
469a71dfd6 sync 2026-09-15_09:35:36 d-robotics@d-roboticsdeMacBook-Pro 2026-09-15 09:35:36 +08:00
bench
7ee7ca7c8f sync 2026-09-15_09:01:15 d-robotics@d-roboticsdeMacBook-Pro 2026-09-15 09:01:15 +08:00
bench
833d65f45a sync 2026-09-15_08:21:18 d-robotics@d-roboticsdeMacBook-Pro 2026-09-15 08:21:18 +08:00
bench
29ef645b58 baseline 2026-09-15: P1-P3 landed; first PASS trajectory; collaboration rules (CONTRACT §12) 2026-09-15 08:10:26 +08:00