Repository navigation
[PR] Revert "[PR] feat(fastpath): v6 composite-identity incremental fastpath lane (full rework, #5576 line voided)" - #5854
Conversation
…th lane (full rework, #5576 line voided)"
📦 Built Package ArtifactBranch: Latest build — install this exact versionnpm install -g billion-context@0.0.8446-pr.5854.be2b6b9This exact version is immutable and reproducible. The rolling Recent builds (commit → version)
Download artifact
tar xzf billion-context-pr5854.tgz
npm install -g packageInstall instructions are refreshed on each push. |
📦 Merged Build PublishedThis merge to master was published as npm install -g billion-context@0.0.1063-master.bebc239 # this exact build
npm install -g billion-context@master # always the newest merged build
|
…f the fastpath-v6 lane Upstream c45ac21..bebc239 is a single revert: ranxianglei#5854 undoes ranxianglei#5652 (66aadbd "Revert feat(fastpath): v6 composite-identity incremental fastpath lane"). 38 files, -5838/+684 on net: src/server/fastpath-v6.ts, src/server/fastpath-v6-render.ts and their two test files are gone, and the sqlite unified store drops back from schema v13 to v11. That matters here for one specific reason. The previous sync resolved a derivation conflict in tests/issue4666-migration-dedup.test.ts by keeping the STACKED form, and the same merge added an export of MYSQL_SCHEMA_VERSION so the mysql-side stamp assertions stop borrowing the sqlite number. The revert collapses both numbers back to 11, which would have silently re-hidden the bug that fix exists to prevent, so the split is kept: the sqlite user_version pin keeps deriving from STORAGE_SCHEMA_VERSION, the four mysql stamp pins keep using MYSQL_SCHEMA_VERSION. Both currently equal 11, but they are different versions owned by different backends and will diverge again on the next plane. Only one conflict this time, same line 159 as before: upstream reverts the literal back to 11, this lane derives. Kept the derived form. src/plugin.ts auto-merged (upstream removes the fastpath hooks, this lane owns the strip entry points). The two CONFIGURATION files and the rollout/env-only seed yaml are generated and were regenerated. Verification on the merge tree: issue5175 58/0, tag-echo 79/0, issue2190-degen-close 32/0, golden-storage-render 5/0, storage-engine-proofs 12/0, issue4666-migration-dedup 2/0. tsc --noEmit clean, four gates exit 0, generated docs byte-identical. Lane symbols all survived the revert: stripDriftResidue 6, isDraftDump 4, containsRenderTagText 8, mayStartDegenerateRenderTag 6, DRIFT_BARE_MISMATCH 3, DRAFT_CLOSE_TAIL 3.
(ai: vllm-qwen/qwen3.8-27b) 查重:不重复 —— 本条是已合并 revert PR #5854 的镜像条目;相关条目:#5652(源 PR,已合并后被整体回退)、#5855(revert 的精确逆操作 re-land,open,#5643 下 snapshot-centric 重构的工作基座)。无同主题并行修复项。 标签:dsh、scope/wiring、p3(pr 为系统既有)—— 该车道属请求管道机制(wiring),diff 涉及 dsh 车道文件;两条 rollout 均默认 0%、终端用户零影响故 p3。不打 bug:这是主动回滚,不是缺陷报告。 结论:revert 精确且完整,master 状态干净,CI 无红,无需任何动作。 这是「revert → 逆操作 re-land」两步棋的第一步(第二步 = #5855,待审);master 当前不含 v6 车道是设计中的中间态,不是终态。 八问:
CI 状态:revert 后 master HEAD(bebc239)单次拉取的 check 全绿(release-kernel / release / build-artifact,0 red);全量测试矩阵由定时 backstop 覆盖(11-01 起 master 不随 push 触发),#4599 今日无红报。时间线附注:第三方 PR #5856(watchdog 可调 + 服务端自持水印摘要,同属水印快路径域)当天 06:04 UTC 开、06:05 UTC 未合关闭,与本 revert 无代码交集,仅记录。另供 #5855 审核留意:其正文称「revert 窗口内无 master 提交触碰清单文件」,但 #5808(04:42 UTC 合入)实际改过 src/persist.ts(v6 亦改此文件)——本 revert 对此处理正确(第 4 问内容级证明),该表述请审核时按实际 diff 复核。 判定:revert 精确、完整、无回归、不破坏缓存,本侧无需动作;下一步等 #5855 的 re-land 评审。 |
ranxianglei#5719 pin realignment + the ranxianglei#5854 revert Upstream moved bebc239 -> 21c950e (14 files). Relevant to this lane: - 151efca test(ranxianglei#5719): realign two stale pins after ranxianglei#5734 changed the re-anchor rung order (30a8bc2). This is the fix for the red I reported on ranxianglei#5176: 30a8bc2 added the E5 self-checkpoint rung without updating T1/T8 of tests/issue5719-msgid-lane-identity.test.ts, so every branch (master included) was red on exactly those two pins. The suite is 8/0 on the merged tree. - 66aadbd + bebc239: the ranxianglei#5854 revert of the fastpath-v6 lane, adopted earlier in 42ece0f6a and still absent from this merge's diff (already present in the lane). - src/cache-keepalive.ts / cache-ledger.ts / served-markers.ts / src/server.ts / upstream-throttle.ts + their tests: no overlap. Overlap: src/server/relay.ts only, and the merge was automatic (my ranxianglei#5176 non-streaming egress at :2062 and the upstream hold path do not cross). Verification on the merged tree (kernel rebuilt, treeless farm): issue5175-drift-lane-leak 58/0, tag-echo 79/0, issue2190-degen-close 32/0, golden-storage-render 5/0, storage-engine-proofs 12/0, native-compact-terminal 1/0, issue4666-migration-dedup 2/0, issue5719-msgid-lane-identity 8/0 (was 6/2). tsc --noEmit 0; four gates 0; config docs fixed point 0.
…stpath-v6 code was reverted, ranxianglei#5854) ranxianglei#5826 commit 33b2f82 re-pinned two msgid-rule-parity rows to the v6 F1/F2 rulings «after master merge»; revert ranxianglei#5854 then removed the fastpath-v6 code (ranxianglei#5652), leaving those pins asserting behavior that no longer exists — master failed the same two rows on pure fb18e7e (verified via detached worktree). This restores tests/msgid-rule-parity.test.ts to its bc44f77 state, i.e. exactly undoes the 33b2f82 hunks: PF_SLOT_UNKNOWN import back to pf-channel.js, per-slot "?" rides-again semantics, K-sliced tail window, and the removed OVER_CAP row (v6 MAX_IDS 16384 no longer applies).
…nel wiring + ranxianglei#5862 parity hygiene + ranxianglei#5860 phantom-pressure guard Master moved twice since the last sync (295ba58 -> 591e940 -> df0a402). - 591e940 (folded in here): ranxianglei#5862 "master hygiene" restored the msgid-rule-parity pins to their pre-v6 verdicts (undoing 33b2f82, which had pinned behaviour that the ranxianglei#5854 revert deleted), plus the ranxianglei#5860 phantom-pressure guard (a7cbfd1) and the kernel 0.0.184 bump. - df0a402 (this merge): ranxianglei#5842 identity-channel wiring - one unified consume seam for every client lane (opencode V2 stamped header, goose/ qwen adapters, openai row-pin parity, decoupled cohort gate), the identity DETERMINATION docs, and the unused-exports gate fixes. The merge itself was conflict-free: the only shared files are the two generated CONFIGURATION docs plus the website config-reference/web artifacts, all of which are regenerated from the committed seeds rather than hand-merged. One inherited red needed a golden refresh: tests/golden/storage-render/ *.json digest drift, pre-existing on pristine master (verified on a detached df0a402 worktree before the merge). The metadata section digest and its derived stateDigest/stateDigestAligned moved with a7cbfd1, which added compress.requireUsageSampleForPressure (default true) to the config echo written into session.metadata - but that commit did not regenerate tests/golden/storage-render/. Regenerated all four under the documented pinned environment (node --import tsx scripts/update-storage-render-goldens.ts); the diff is exactly 3 lines per wire (stateDigest, stateDigestAligned, sectionDigests.metadata) and every other section digest is byte-identical.
…F2 ruling governs again Master's 613d290 restored these pins to pre-v6 verdicts on the explicit condition that 'fastpath-v6 code was reverted (#5854)'. This PR re-lands v6, so per KDD #17 (deliberate divergences re-pinned consciously, with stated rationale) the pins follow the code again: - pf-channel e2e: F2 owner ruling (2026-11-04, PR #5652 review follow-up) — any hole/duplicate clears the WHOLE captured table; '?' never travels from the client. Pre-v6 '?'-slice verdicts replaced by count-only verdicts. - pf-channel e2e: F1 fixed tail window (last <=64 ids) replaces K-based slicing; capture cap raised 2048 -> 16384 (cap still pinned via a 16385-id row so coverage does not shrink). - dsh pfSlotIdsOf: same F2 whole-table-clear (was '?' slot list). - PF_SLOT_UNKNOWN import moves from agent/pf-channel.js (constant removed client-side by F2) to server/fastpath.js (wire decoder still accepts legacy '?' segments from v5 traffic).
…nel wiring + ranxianglei#5862 parity hygiene + ranxianglei#5860 phantom-pressure guard Master moved twice since the last sync (295ba58 -> 591e940 -> df0a402). - 591e940 (folded in here): ranxianglei#5862 "master hygiene" restored the msgid-rule-parity pins to their pre-v6 verdicts (undoing 33b2f82, which had pinned behaviour that the ranxianglei#5854 revert deleted), the ranxianglei#5860 phantom-pressure guard (a7cbfd1), and the kernel 0.0.184 bump. - df0a402 (this merge): ranxianglei#5842 identity-channel wiring - one unified consume seam for every client lane (opencode V2 stamped header, goose/ qwen adapters, openai row-pin parity, decoupled cohort gate), the identity DETERMINATION docs, and the unused-exports gate fixes. The merge itself was conflict-free: the only shared files are the two generated CONFIGURATION docs plus the website config-reference/web artifacts, all of which are regenerated from the committed seeds rather than hand-merged. Two golden refreshes ride along, both caused by a7cbfd1 and both already red on pristine master before this merge (verified on detached worktrees at 591e940 and df0a402): - tests/golden/storage-render/*.json - the new compress.requireUsageSampleForPressure (default true) lands in the config echo written into session.metadata, so only the metadata section digest and its derived stateDigest/stateDigestAligned moved. Three lines per wire; every other section digest is byte-identical. - tests/golden/wire-contract/kernel.json - the same knob appears in the pinned acp-kernel NudgeConfig export snapshot: exactly one added line, "requireUsageSampleForPressure": true. Both regenerated with the documented scripts (update-storage-render-goldens, update-wire-contract-goldens) under the pinned suite environment. Not fixed here (upstream behaviour, needs an owner ruling): a7cbfd1 defaults the pressure band to usage-grade readings only, but tests/fork-nudge-trigger.test.ts and tests/silent-backend-nudge.test.ts still assert that a local-estimate reading fires the pressure nudge. Both are byte-identical to master and fail there too, so this merge does not touch them.
Reverts #5652