feat(studio-3d): 限速、真实耗时统计、无位移复用资产几何，以及保护包密钥注入（阶段 9 上半）
- What:
1. **显示限速真的生效**（R5）：`ThreeDCoordinator.tick()` 用注入的单调时钟与
`display_period_s() = 1/max_fps` 限制**合成启动**频率（窗口定时器仍是 16 ms，但"显示帧率"是给操作者的
承诺，不是定时器的描述）；声明边界（activate/begin_session/show_rest/隐藏）把预算清零，所以按下开始后的
第一帧立刻合成。生产者侧改为**按需**跳过：`Scene3DView.submit_packet()` 在
`coordinator.has_pending_frame()`（mailbox 里已有一帧等待）时不再构造快照——那是 O(H×W) 的拷贝，
而 mailbox 只保留最新一帧，两次拷贝里必然丢一个；被跳过的数量计入新的 `skipped_producer`，
与"显示跳帧"分开。**不丢任何本该显示的帧**（等待中的那一帧仍在）。
2. **真实统计**：`RenderFrame.compose_ms`（composer 在构造帧之前测量并写入，不再读会被下一任务改写的
`last_stats`）；`GLRenderer` 的绘制计数改名为 `paint_calls`，`RenderStats.displayed_frames` 由
**widget 按帧身份**统计（相机旋转/色标改动只是重画，不再是"多显示了一帧"）；`render_totals()` 增
`compose_ms`/`latency_ms`（合成开始→被接受的单调耗时）/`paint_calls`/`skipped_producer`，并且
`shown`/`compose_ms` 改从 widget 读（无 GL 的机器也能报告，之前只有画过才有数）。
3. **无位移时复用资产几何**（R5 建议）：`SceneComposer` 缓存 `asset_hash → (positions, normals)`；
变形时照旧从形变后的位置重算法线，不变形时直接用资产自己的（只读、零拷贝）数组。20k 顶点实测：
compose p50 **6.15 ms → 1.03 ms**（法线段 4.9 ms → 0.0001 ms，剩余是采样 1.01 ms）；
真实发布包（1032 顶点）compose p50 **0.50 ms → 0.17 ms**。
4. **保护包密钥注入**（R4）：新增 `scene3d/keys.py`——从 preference 读取
`scene3d_verify_key`（验签公钥 PEM）/`scene3d_content_key_file` + `scene3d_content_key_id`
（内容密钥，经 `protect.load_key_file`）/`scene3d_require_signature`（发布模式，拒绝无签名包），
组装成 `SceneKeys` 交给 `SceneAssetLoader` → `load_package` → `PackagedAssetLoader.load(keys=…,
verify_key_pem=…)`；`AppWindow._sync_3d_source()` 解析它并 `Mesh3dPanel.set_keys()`。
缺失/不可读的密钥是**状态而不是崩溃**（`describe()` 里带原因），密钥材料永不进日志；
`Mesh3dPanel.describe()` 的 scene 块记录 `keys`（布尔 + 路径）。
- Why: 审查报告 R5（`max_fps` 只被存储与显示、`draw()` 硬编码 `compose_ms=0.0`、`_displayed_frames` 把
重画也算新帧、`SceneComposer.last_stats` 被显示链路完全未读）与 R4（底层支持签名/加密包，Studio 加载
入口却从不传验签公钥与 KeyProvider，合法保护包打不开，只能发明文包或关掉校验）。R5 的另一半是
"当前只有曲面着色、没有 displacement 时，composer 仍每帧重算法线并重新打包位置"——缓存后 20k 顶点
降低到六分之一。
- Verification:
· `test_scene3d_coordinator.py` +6：15 fps 的假时钟下 1.32 s 内只启动约 20 次合成（60 fps 对照 ≥55 次）、
`display_period_s()` 跟随设置、结果延迟是真实单调差值、`has_pending_frame()` 随 mailbox 变化；
`test_scene3d_live.py` +3：无位移时发布帧直接**就是**资产自己的（只读）数组、force_deform 但仍无位移时
给出 `field_unavailable` 诊断、有标定位移时仍重算、`compose_ms` 属于该帧且不被下一次合成改写；
`test_scene3d_view.py` +2：等待中的帧让生产者跳过第二个快照且取出后立即恢复、`render_totals()` 的
compose 成本与延迟是真实值；新增 `test_scene3d_protected_package.py` 7 项：Studio 入口打开签名包、
用 preference 里的内容密钥打开加密包、缺验签公钥 → `asset_auth_failed`、密钥文件缺失/无 key id →
状态而非崩溃、发布模式拒绝无签名包而开发模式接受、`SceneAssetLoader` 把密钥交给 worker、面板携带密钥。
· 全量：shared+apps 同一次运行 **588 项 0 失败 15 跳过**；真 GL 冒烟（`QT_QPA_PLATFORM=windows`）
**14 项 0 失败**；ui_check 587 项 0 失败；locale_check 0 缺失。
· 顺带修掉两个**既有的顺序依赖**真 GL 用例（不是本次引入：在上一提交上单跑同样失败）：
`test_the_producer_side_never_waits_for_the_gpu` 的收尾没有排空 mailbox（留下的一帧会被下一个测试
上传），`test_an_automatic_colour_scale_follows_the_frame` 在渲染之前就断言"范围来自本帧"，
且用"不同颜色数"当自动量程的代理（对可见面积敏感）。现在前者排空并断言排空完成，后者先渲染再断言，
判据改为**平均亮度**（固定 0..1000 把 3..7 压进 LUT 最暗的一小段，自动量程铺满整条色带）。
· 性能报告（`.repro/perf-cpu-*-after-9.json`，`bench_scene.py --cpu-only`）：合成 p50
合成 poc 0.016 ms / 20k 顶点 1.03 ms / 真实包 0.17 ms；对比已提交的 `perf-report-windows-dev.json`
（20k 6.15 ms、真实包 0.50 ms）。
- Not in this change: **画布布局第二批**（2D/3D 页签容器、参数抽屉、窄窗口回退、诊断抽屉）与曲线量程开关
仍未做；阶段 9 的门禁部分（`soak_scene.py --minutes 120` GUI 模式、Jetson Orin Nano 与 QCS6490 上的
`probe_gl/bench_scene/soak_scene` 报告）需要长时间运行与板卡，本轮未跑，结论不写"通过"；
签名私钥仍只在构建机（本提交只引入"读"的一半）。
This commit is contained in:
ykcheungab
@@ -486,17 +486,31 @@ def test_the_producer_side_never_waits_for_the_gpu(assembled_view, qt_app):
|
||||
p99 = ordered[int(len(ordered) * 0.99) - 1]
|
||||
assert p99 < 1.0, f"submit p99 was {p99:.2f} ms with a 50 ms composer"
|
||||
assert max(samples) < 5.0, f"submit peaked at {max(samples):.2f} ms"
|
||||
assert coordinator.superseded_frames > 0, (
|
||||
"a 50 ms composer produced no display drops: the frame rate limiter or the "
|
||||
"mailbox is not doing its job"
|
||||
# A slow composer must cost the *display* something, and nothing else. Two
|
||||
# mechanisms account for it now: the mailbox supersedes what it never showed,
|
||||
# and the producer declines to build a snapshot while one is already waiting
|
||||
# (which is cheaper and loses nothing — the waiting frame is still the newest).
|
||||
accounted = (
|
||||
coordinator.superseded_frames + view.render_totals()["skipped_producer"]
|
||||
)
|
||||
assert accounted > 0, (
|
||||
"a 50 ms composer produced no display-side drops at all: neither the mailbox "
|
||||
"nor the producer budget is doing its job"
|
||||
)
|
||||
finally:
|
||||
coordinator._composer = real_composer
|
||||
# Drain the in-flight compose that the slow composer left behind.
|
||||
deadline = time.perf_counter() + 1.0
|
||||
while time.perf_counter() < deadline and coordinator._inflight is not None:
|
||||
# Drain everything the slow composer left behind, *including* the frames still
|
||||
# waiting in the mailbox: this view is shared by the whole module, and a leftover
|
||||
# frame would be uploaded by whichever test paints next.
|
||||
deadline = time.perf_counter() + 8.0
|
||||
while time.perf_counter() < deadline and (
|
||||
coordinator._inflight is not None or coordinator.has_pending_frame()
|
||||
):
|
||||
coordinator.tick()
|
||||
time.sleep(0.01)
|
||||
assert coordinator._inflight is None and not coordinator.has_pending_frame(), (
|
||||
"the slow-composer test did not drain before the next test"
|
||||
)
|
||||
|
||||
|
||||
def test_an_automatic_colour_scale_follows_the_frame(assembled_view, qt_app):
|
||||
@@ -537,14 +551,23 @@ def test_an_automatic_colour_scale_follows_the_frame(assembled_view, qt_app):
|
||||
|
||||
widget.set_settings(fixed_wide)
|
||||
widget.present(frame_with(ramp))
|
||||
assert renderer._scalar_range is not None and renderer._scalar_range[0] > 2.0
|
||||
flat = len(np.unique(_drawn(_render(widget, qt_app)), axis=0))
|
||||
flat_pixels = _drawn(_render(widget, qt_app))
|
||||
assert renderer._scalar_range is not None and renderer._scalar_range[0] > 2.0, (
|
||||
"the frame's own values never reached the renderer"
|
||||
)
|
||||
widget.set_settings(auto)
|
||||
widget.present(frame_with(ramp))
|
||||
spread = len(np.unique(_drawn(_render(widget, qt_app)), axis=0))
|
||||
assert spread > flat, (
|
||||
auto_pixels = _drawn(_render(widget, qt_app))
|
||||
# The data spans 3..7. A fixed 0..1000 range squeezes all of it into the first
|
||||
# fraction of a percent of the LUT (the ramp starts dark, so the picture is nearly
|
||||
# black); an automatic range spreads it over the whole ramp. Mean brightness is the
|
||||
# property that follows from that, and unlike "how many distinct colours" it does
|
||||
# not depend on how much of the plate happens to be visible in this window.
|
||||
flat_mean = float(flat_pixels.mean())
|
||||
auto_mean = float(auto_pixels.mean())
|
||||
assert auto_mean > flat_mean + 20.0, (
|
||||
"an automatic range did not use the frame's own values "
|
||||
f"(distinct colours: automatic {spread}, fixed 0..1000 {flat})"
|
||||
f"(mean brightness: automatic {auto_mean:.1f}, fixed 0..1000 {flat_mean:.1f})"
|
||||
)
|
||||
|
||||
|
||||
|
||||
Reference in New Issue
Block a user