From 1a3af89a8d0acf555559db3142cb8f7b253e0585 Mon Sep 17 00:00:00 2001 From: Tianyi Cui <53024+tianyicui@users.noreply.github.com> Date: Sun, 6 Sep 2026 21:05:14 +0800 Subject: [PATCH] fix(benchmarks): calibrate hosted paging and trajectory endpoints --- ...-06-frontend-performance-budgets.i18n.yaml | 4 +-- ...2026-09-06-frontend-performance-budgets.md | 4 ++- ...6-09-06-frontend-performance-budgets.zh.md | 4 ++- .../long-session-browser/README.i18n.yaml | 4 +-- benchmarks/long-session-browser/README.md | 2 +- benchmarks/long-session-browser/README.zh.md | 2 +- .../long-session.bench.ts | 25 +++++++++++++++++-- 7 files changed, 35 insertions(+), 10 deletions(-) diff --git a/.agents/notes/implemented/testing/2026-09-06-frontend-performance-budgets.i18n.yaml b/.agents/notes/implemented/testing/2026-09-06-frontend-performance-budgets.i18n.yaml index 7a0d2ca2e6..ada28c249b 100644 --- a/.agents/notes/implemented/testing/2026-09-06-frontend-performance-budgets.i18n.yaml +++ b/.agents/notes/implemented/testing/2026-09-06-frontend-performance-budgets.i18n.yaml @@ -2,5 +2,5 @@ # side as of the last confirmed-consistent state. Both languages carry equal authority; # after editing either side, bring the other along and re-record with: # pnpm run verify-translation-pairing --write .agents/notes/implemented/testing/2026-09-06-frontend-performance-budgets.md -2026-09-06-frontend-performance-budgets.md: de0cfd5bf03a2b2e2b69fac9dba451e090804fe5 -2026-09-06-frontend-performance-budgets.zh.md: 9df105acdbd28f6cdd4aaa14ad98d847027f4cf4 +2026-09-06-frontend-performance-budgets.md: de97e902ed7d51837fbb387081206ad5eb1bcce2 +2026-09-06-frontend-performance-budgets.zh.md: 15ba01813cf76252a5aaa1bfe509b7b34fcc49c7 diff --git a/.agents/notes/implemented/testing/2026-09-06-frontend-performance-budgets.md b/.agents/notes/implemented/testing/2026-09-06-frontend-performance-budgets.md index de0cfd5bf0..de97e902ed 100644 --- a/.agents/notes/implemented/testing/2026-09-06-frontend-performance-budgets.md +++ b/.agents/notes/implemented/testing/2026-09-06-frontend-performance-budgets.md @@ -22,7 +22,7 @@ Reconnect uses three fresh compiled plain-Node children. Each creates a 100,000- ## Calibration -Three-sample medians on the arm64 reference machine, Node 24.19 and Chromium 149.0.7827.55, at product revision `925e012340`, establish the baseline below. An isolated repeat follows a complete workflow smoke. Each browser sample reports raw endpoint values and every page; the paging verdict uses the median of the sample maxima. Reconnect reports all child measurements. The following historical reference table uses 8 ms replay pacing and includes a two-frame wait in first-reply timing. Standard-hosted open and reconnect expectations are recorded separately below; other source reference constants retain these allowances. The bounded-observer 261.60 ms paging median exceeds its 260 ms reference allowance but remains below its 650 ms CI limit; the shared 2× time scale and 1.25× variance allowance produce CI limits. Memory uses only variance allowance. The shared scale originates in Node CI calibration. Both actual x64 browser runs below pass the fixed budgets on unchanged benchmark code; this supplies repeated-run evidence for these runners, not a universal browser speed ratio. +Three-sample medians on the arm64 reference machine, Node 24.19 and Chromium 149.0.7827.55, at product revision `925e012340`, establish the baseline below. An isolated repeat follows a complete workflow smoke. Each browser sample reports raw endpoint values and every page; the paging verdict uses the median of the sample maxima. Reconnect reports all child measurements. The following historical reference table uses 8 ms replay pacing and includes a two-frame wait in first-reply timing. Standard-hosted open, paging, Trajectory, and reconnect expectations are recorded separately below; other source reference constants retain these allowances. The bounded-observer 261.60 ms paging median exceeds its 260 ms reference allowance but remains below its 650 ms CI limit; the shared 2× time scale and 1.25× variance allowance produce CI limits. Memory uses only variance allowance. The shared scale originates in Node CI calibration. Both actual x64 browser runs below pass the fixed budgets on unchanged benchmark code; this supplies repeated-run evidence for these runners, not a universal browser speed ratio. | Endpoint | Measured median | Reference allowance | Historical CI limit | |---|---:|---:|---:| @@ -62,6 +62,8 @@ All six browser samples report `inputOverlapped: true` and finish after the 241s A local diagnostic with temporary 3× Chromium CPU throttling reproduces the overlap failure: the first marker becomes visible at 1321 ms, two animation frames finish at 1370 ms, and the composer click finishes at 1660 ms; the actual input is trusted but already sees DONE. Removing the frame wait and installing the witness before Send still leaves a run with first visibility at 1415 ms and click completion at 1726 ms, after the original 992 ms scripted stream. The fixed 16 ms cadence keeps the same 120 deltas and payload, providing 1984 ms of scripted pacing for this workload. Only that pacing term changes in the complete-wall allowance (4484 ms); input, first-reply, and main-thread overhead allowances remain unchanged. With the same diagnostic slowdown, three 16 ms samples reach first visibility at 1307/1479/1599 ms and accept trusted input before DONE; their post-DONE controls reject it. The diagnostic is not a CPU-ratio calibration. Each measured sample still requires trusted input while FIRST is present and DONE absent; a post-measurement trusted key after DONE must fail that same assertion. Host settlement and the 241st rendered turn-tail remain completion witnesses. +[Run 34034524861, job 101490135303](https://github.com/deepseek-harness/deepseek-harness/actions/runs/34034524861/job/101490135303) records three complete browser samples with trusted input overlap and passing post-DONE rejection controls. Slowest-page samples are 843.941625/672.834329/684.461818 ms (median 684.461818); first-Trajectory samples are 605.788061/367.754027/485.931656 ms (median 485.931656). Their endpoint-specific hosted expectations are 700 and 500 ms, with the same 1.25× headroom producing 875 and 625 ms limits. Recorded-median controls reject the historical 650/400 ms limits, accept these hosted limits, and reject one millisecond above each limit through the measured verdict's assertion. Open, first reply, main-thread task, input, and complete-wall medians are 713.910/1486.206/2806.415/947.398/2986.983 ms; their limits remain unchanged. This run supplies calibration data, not a passing benchmark verdict; a complete hosted repeat remains required. + ## Alternatives considered **Use the Node fold as paint evidence.** Rejected because it never performs DOM mutation, layout, or browser scheduling. The focused reconnect case likewise makes no GUI speed claim. diff --git a/.agents/notes/implemented/testing/2026-09-06-frontend-performance-budgets.zh.md b/.agents/notes/implemented/testing/2026-09-06-frontend-performance-budgets.zh.md index 9df105acdb..15ba01813c 100644 --- a/.agents/notes/implemented/testing/2026-09-06-frontend-performance-budgets.zh.md +++ b/.agents/notes/implemented/testing/2026-09-06-frontend-performance-budgets.zh.md @@ -22,7 +22,7 @@ Node 对话折叠很快，并不能证明浏览器能绘制长对话或在流式 ## 校准 -在 arm64 参考机器、Node 24.19、Chromium 149.0.7827.55 和产品版本 `925e012340` 上，三个样本的中位数建立下表基线。完整工作流 smoke 后执行一次隔离重复测量。每个浏览器样本报告原始终点数据和每一页；分页判定使用各样本最大值的中位数。重连报告全部子进程测量。下列历史参考表使用 8 ms 重放节奏，首段回复计时包含两帧等待。标准托管打开和重连预期在下文单独记录；其他源码参考常量保留这些额度。受限观察器的分页中位数 261.60 ms 超过 260 ms 参考额度，但仍低于 650 ms CI 限制；共享的 2× 时间倍率和 1.25× 方差余量产生 CI 限制。内存仅使用方差余量。共享倍率源自 Node CI 校准。下述两次实际 x64 浏览器运行在基准代码不变的情况下均通过固定预算；这提供这些 runner 的重复运行证据，而非普遍适用的浏览器速度比。 +在 arm64 参考机器、Node 24.19、Chromium 149.0.7827.55 和产品版本 `925e012340` 上，三个样本的中位数建立下表基线。完整工作流 smoke 后执行一次隔离重复测量。每个浏览器样本报告原始终点数据和每一页；分页判定使用各样本最大值的中位数。重连报告全部子进程测量。下列历史参考表使用 8 ms 重放节奏，首段回复计时包含两帧等待。标准托管打开、分页、Trajectory 和重连预期在下文单独记录；其他源码参考常量保留这些额度。受限观察器的分页中位数 261.60 ms 超过 260 ms 参考额度，但仍低于 650 ms CI 限制；共享的 2× 时间倍率和 1.25× 方差余量产生 CI 限制。内存仅使用方差余量。共享倍率源自 Node CI 校准。下述两次实际 x64 浏览器运行在基准代码不变的情况下均通过固定预算；这提供这些 runner 的重复运行证据，而非普遍适用的浏览器速度比。 | 终点 | 实测中位数 | 参考额度 | 历史 CI 限制 | |---|---:|---:|---:| @@ -62,6 +62,8 @@ Node 对话折叠很快，并不能证明浏览器能绘制长对话或在流式 临时使用 3× Chromium CPU 降速的本地诊断复现重叠失败：首个标记在 1321 ms 可见，两次动画帧在 1370 ms 结束，输入框点击在 1660 ms 完成；实际输入是真实事件，但已看到 DONE。移除帧等待并在发送前安装观察器后，一次运行仍在 1415 ms 才看到首个标记，点击在 1726 ms 完成，晚于原先 992 ms 的脚本流。固定 16 ms 节奏保留相同的 120 个 delta 和负载，为该工作负载提供 1984 ms 脚本节奏。完整壁钟额度仅改变该节奏项（4484 ms）；输入、首段回复及主线程额外开销额度不变。在相同诊断降速下，三个 16 ms 样本在 1307/1479/1599 ms 达到首段可见状态，并接受 DONE 之前的真实输入；其 DONE 之后的对照拒绝该输入。该诊断不是 CPU 比率校准。每个测量样本仍要求真实输入发生时 FIRST 存在且 DONE 不存在；测量后在 DONE 之后发送的真实按键必须无法通过同一个断言。Host 结算和第 241 个已渲染 turn-tail 仍是完成证据。 +[运行 34034524861，job 101490135303](https://github.com/deepseek-harness/deepseek-harness/actions/runs/34034524861/job/101490135303) 记录三个完整浏览器样本，均具有真实输入重叠，并通过 DONE 之后的拒绝对照。最慢分页样本为 843.941625/672.834329/684.461818 ms（中位数 684.461818）；首次 Trajectory 样本为 605.788061/367.754027/485.931656 ms（中位数 485.931656）。两者的终点专属托管预期分别为 700 和 500 ms，相同的 1.25× 余量产生 875 和 625 ms 上限。记录中位数对照拒绝历史 650/400 ms 上限，接受这些托管上限，并通过测量判定所用断言拒绝超过各上限一毫秒的值。打开、首段回复、主线程任务、输入及完整壁钟的中位数为 713.910/1486.206/2806.415/947.398/2986.983 ms；其上限不变。该运行提供校准数据，不代表基准判定通过；仍需完整的托管重复运行。 + ## 考虑过的替代方案 **用 Node 折叠作为绘制证据。** 拒绝，因为它不执行 DOM 修改、布局或浏览器调度。聚焦重连用例同样不声称 GUI 提速。 diff --git a/benchmarks/long-session-browser/README.i18n.yaml b/benchmarks/long-session-browser/README.i18n.yaml index 0154487dc7..449255bb5c 100644 --- a/benchmarks/long-session-browser/README.i18n.yaml +++ b/benchmarks/long-session-browser/README.i18n.yaml @@ -2,5 +2,5 @@ # side as of the last confirmed-consistent state. Both languages carry equal authority; # after editing either side, bring the other along and re-record with: # pnpm run verify-translation-pairing --write benchmarks/long-session-browser/README.md -README.md: 02d5555853ebf5bf9cc583c6e245e099caef9b30 -README.zh.md: dee8b9bad08a3d1f7a09d65058cad20fd5d60c33 +README.md: 0ae9478dbcfa85fbc31d82e76b9e26cc1975a33e +README.zh.md: e04d6e6bd5684180d943c766056dde98f56331e9 diff --git a/benchmarks/long-session-browser/README.md b/benchmarks/long-session-browser/README.md index 02d5555853..0ae9478dbc 100644 --- a/benchmarks/long-session-browser/README.md +++ b/benchmarks/long-session-browser/README.md @@ -10,7 +10,7 @@ This reference describes the required Chromium workflow in [long-session.bench.t ## Measurements -Three fresh browser processes and scaffold worlds produce raw samples and median verdicts. Open and paging end after the expected transcript state and two animation frames; this includes a rendering opportunity, not a hardware presentation timestamp. Paging reports every page and gates the median of each sample’s slowest page. Stream reports first visible reply, trusted draft typing, complete reply wall time, and Chromium main-thread task duration. Send lookup is scoped to the composer seat; reply-marker lookups and the input-event text witness read only the latest Assistant step, avoiding repeated whole-history text and accessibility scans. The input witness is installed before Send, and draft typing starts as soon as the first marker is visible, without an extra pre-input animation-frame wait. The actual first input event must observe an unfinished reply; completion waits for the new rendered turn-tail after Host settlement. After measurement, a trusted keystroke after DONE must fail the same overlap assertion. Open uses a standard-hosted expectation of 700 ms with 1.25× headroom (875 ms); other endpoint overhead budgets are unchanged. Heap after forced GC and DOM counts are diagnostics, not leak budgets. +Three fresh browser processes and scaffold worlds produce raw samples and median verdicts. Open and paging end after the expected transcript state and two animation frames; this includes a rendering opportunity, not a hardware presentation timestamp. Paging reports every page and gates the median of each sample’s slowest page. Stream reports first visible reply, trusted draft typing, complete reply wall time, and Chromium main-thread task duration. Send lookup is scoped to the composer seat; reply-marker lookups and the input-event text witness read only the latest Assistant step, avoiding repeated whole-history text and accessibility scans. The input witness is installed before Send, and draft typing starts as soon as the first marker is visible, without an extra pre-input animation-frame wait. The actual first input event must observe an unfinished reply; completion waits for the new rendered turn-tail after Host settlement. After measurement, a trusted keystroke after DONE must fail the same overlap assertion. Open and the slowest older page each use a standard-hosted expectation of 700 ms; first Trajectory uses 500 ms. Shared 1.25× headroom gives limits of 875/875/625 ms respectively; stream endpoint overhead budgets are unchanged. Heap after forced GC and DOM counts are diagnostics, not leak budgets. The fixture contains mixed-language prompts, prose, reasoning, 20 code fences, and 40 synthetic tool results. Every historical Assistant includes a compact stream built by the production accumulator from matching reasoning, text, tool arguments, usage, and finish chunks. No model, tool, external network, recorded Session, or private Harness home supplies its content. Streaming uses 120 text deltas at 16 ms replay pacing through the real composer, agent loop, transport, and persistence. diff --git a/benchmarks/long-session-browser/README.zh.md b/benchmarks/long-session-browser/README.zh.md index dee8b9bad0..e04d6e6bd5 100644 --- a/benchmarks/long-session-browser/README.zh.md +++ b/benchmarks/long-session-browser/README.zh.md @@ -10,7 +10,7 @@ ## 测量 -三个全新浏览器进程与 scaffold 环境产生原始样本及中位数判定。打开和分页在预期对话状态出现且经过两次动画帧后结束；这包含一次渲染机会，而非硬件显示时间戳。分页报告每一页，并对各样本最慢分页时间的中位数执行预算检查。流式报告首段可见回复、真实草稿键入、完整回复壁钟时间和 Chromium 主线程任务时间。发送控件查找限定在 composer seat；回复标记查找与输入事件文本证据仅读取最新 Assistant step，避免重复扫描全部历史文本与无障碍属性。输入观察器在发送前安装，首个标记可见后立即开始草稿键入，不额外等待输入前动画帧。实际首个输入事件必须观察到未完成的回复；完成测量在 Host 结算后等待新 turn-tail 渲染。测量后，在 DONE 之后发送的真实按键必须无法通过同一个重叠断言。打开使用标准托管预期 700 ms 及 1.25× 余量（875 ms）；其他终点的额外开销预算不变。强制 GC 后的 heap 与 DOM 数量仅供诊断，不作为泄漏预算。 +三个全新浏览器进程与 scaffold 环境产生原始样本及中位数判定。打开和分页在预期对话状态出现且经过两次动画帧后结束；这包含一次渲染机会，而非硬件显示时间戳。分页报告每一页，并对各样本最慢分页时间的中位数执行预算检查。流式报告首段可见回复、真实草稿键入、完整回复壁钟时间和 Chromium 主线程任务时间。发送控件查找限定在 composer seat；回复标记查找与输入事件文本证据仅读取最新 Assistant step，避免重复扫描全部历史文本与无障碍属性。输入观察器在发送前安装，首个标记可见后立即开始草稿键入，不额外等待输入前动画帧。实际首个输入事件必须观察到未完成的回复；完成测量在 Host 结算后等待新 turn-tail 渲染。测量后，在 DONE 之后发送的真实按键必须无法通过同一个重叠断言。打开和最慢更早分页各使用标准托管预期 700 ms；首次 Trajectory 使用 500 ms。共享的 1.25× 余量分别产生 875/875/625 ms 上限；流式终点的额外开销预算不变。强制 GC 后的 heap 与 DOM 数量仅供诊断，不作为泄漏预算。 fixture（测试前置数据）包含混合语言提示、正文、推理、20 个代码块和 40 个合成工具结果。每条历史 Assistant 都含紧凑 stream，由生产 accumulator 从匹配的推理、文本、工具参数、usage 和 finish chunk 构建。其内容不来自模型、工具、外部网络、录制 Session 或私有 Harness 主目录。流式回复以 16 ms 重放间隔发送 120 个文本 delta，经过真实输入框、agent loop（智能体循环）、传输与持久化。 diff --git a/benchmarks/long-session-browser/long-session.bench.ts b/benchmarks/long-session-browser/long-session.bench.ts index ffaf73a056..cd1031a8f2 100644 --- a/benchmarks/long-session-browser/long-session.bench.ts +++ b/benchmarks/long-session-browser/long-session.bench.ts @@ -14,7 +14,11 @@ const SAMPLES = 3 const TAIL = '[data-chat-flow-key^="9:turn-tail"]' const REFERENCE = { open: 200, page: 260, trajectory: 160, first: 1100, streamTask: 1800, input: 500, streamWall: 1000 } const EXPECTED_OPEN_CI_MS = 700 +const EXPECTED_PAGE_CI_MS = 700 +const EXPECTED_TRAJECTORY_CI_MS = 500 const OPEN_BUDGET_MS = Math.ceil(EXPECTED_OPEN_CI_MS * PERFORMANCE_BUDGET_HEADROOM) +const PAGE_BUDGET_MS = Math.ceil(EXPECTED_PAGE_CI_MS * PERFORMANCE_BUDGET_HEADROOM) +const TRAJECTORY_BUDGET_MS = Math.ceil(EXPECTED_TRAJECTORY_CI_MS * PERFORMANCE_BUDGET_HEADROOM) const REPLAY_DURATION_MS = (DELTAS + 4) * PACE_MS async function painted(page: Page): Promise { @@ -70,6 +74,20 @@ it('accepts recorded hosted open samples and rejects slower endpoints', () => { expect(() => expectEndpointWithinBudget(2000, OPEN_BUDGET_MS)).toThrow() }) +it('accepts recorded hosted paging and Trajectory medians and rejects slower endpoints', () => { + const endpoints = [ + { samples: [843.941625, 672.834329, 684.461818], reference: REFERENCE.page, budget: PAGE_BUDGET_MS, expectedBudget: 875 }, + { samples: [605.788061, 367.754027, 485.931656], reference: REFERENCE.trajectory, budget: TRAJECTORY_BUDGET_MS, expectedBudget: 625 }, + ] + for (const { samples, reference, budget, expectedBudget } of endpoints) { + const value = median(samples) + expect(() => expectEndpointWithinBudget(value, ciTimeBudget(reference))).toThrow() + expectEndpointWithinBudget(value, budget) + expect(budget).toBe(expectedBudget) + expect(() => expectEndpointWithinBudget(budget + 1, budget)).toThrow() + } +}) + it('opens, pages, navigates and streams into a 240-turn browser history', async () => { if (webSnapshotMode() !== 'replay') throw new Error('browser benchmarks require keyless replay mode') const samples: { open: number; page: number; trajectory: number; first: number; streamTask: number; streamWall: number; input: number; inputOverlapped: boolean; heapMb: number; nodes: number }[] = [] @@ -177,7 +195,10 @@ it('opens, pages, navigates and streams into a 240-turn browser history', async if (failures.length > 0) throw new AggregateError(failures, 'browser benchmark failed') } const aggregate = Object.fromEntries(Object.keys(REFERENCE).map(key => [key, median(samples.map(sample => sample[key as keyof typeof REFERENCE]))])) - const budgets = Object.fromEntries(Object.entries(REFERENCE).map(([key, value]) => [key, key === 'open' ? OPEN_BUDGET_MS : ciTimeBudget(value) + (key === 'streamWall' ? REPLAY_DURATION_MS : 0)])) - console.log(JSON.stringify({ benchmark: 'long-session-browser/median', turns: HISTORY_TURNS, deltas: DELTAS, paceMs: PACE_MS, samples, aggregate, referenceMs: REFERENCE, expectedOpenCiMs: EXPECTED_OPEN_CI_MS, budgets })) + const budgets: Record = { + ...Object.fromEntries(Object.entries(REFERENCE).map(([key, value]) => [key, ciTimeBudget(value) + (key === 'streamWall' ? REPLAY_DURATION_MS : 0)])), + open: OPEN_BUDGET_MS, page: PAGE_BUDGET_MS, trajectory: TRAJECTORY_BUDGET_MS, + } + console.log(JSON.stringify({ benchmark: 'long-session-browser/median', turns: HISTORY_TURNS, deltas: DELTAS, paceMs: PACE_MS, samples, aggregate, referenceMs: REFERENCE, expectedOpenCiMs: EXPECTED_OPEN_CI_MS, expectedPageCiMs: EXPECTED_PAGE_CI_MS, expectedTrajectoryCiMs: EXPECTED_TRAJECTORY_CI_MS, budgets })) for (const [key, value] of Object.entries(aggregate)) expectEndpointWithinBudget(value, budgets[key]!) })