Skip to content

[M2-AC-01] AC-4.x acceptance suite (projection-mode independence, 60 FPS sort gate) #99

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
offdev merged 14 commits into from
Oct 14, 2026
Merged

[M2-AC-01] AC-4.x acceptance suite (projection-mode independence, 60 FPS sort gate) #99

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
14 commits
Select commit Hold shift + click to select a range
cb40697
[M2-AC-01] AC-4.x acceptance suite (projection-mode independence, 60 …
offdev Oct 14, 2026
e387be3
Fix MSVC C4457: rename the ac4 local config to ac4Cfg
offdev Oct 14, 2026
d43871c
TEMPORARY: gdb backtrace entry for the ac4 exit-time segfault
offdev Oct 14, 2026
4b49fa7
TEMPORARY: install gdb in the diagnostic entry (not on ubuntu-24.04)
offdev Oct 14, 2026
911a9b2
TEMPORARY: fix the gdb diagnostic command + add an ASan diagnostic entry
offdev Oct 14, 2026
6c77a75
TEMPORARY: one single-string bash -c in the gdb diagnostic
offdev Oct 14, 2026
6b5795a
TEMPORARY: absolute binary path in the gdb diagnostic
offdev Oct 14, 2026
f9483a6
TEMPORARY: run the gdb diagnostic at the canonical 3000/1000 shape
offdev Oct 14, 2026
920dc7a
TEMPORARY: in-process crash capture for the ac4 exit segfault
offdev Oct 14, 2026
6c30638
TEMPORARY: identify the worker threads that segfault at ac4 exit
offdev Oct 14, 2026
0a1c0f3
TEMPORARY: fail-fast census for the ac4 and render-50k worker threads
offdev Oct 14, 2026
a11d5c8
Fix the ac4 exit segfault: drain pending GL work before context destroy
offdev Oct 14, 2026
1bcf3a1
Stop dlclose'ing the EGL driver (live worker threads)
offdev Oct 14, 2026
fd04a1f
Record the M2-AC-01 acceptance: baseline, budgets, roadmap, reference…
offdev Oct 14, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
51 changes: 29 additions & 22 deletions .github/workflows/ci-pull.yml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -27,11 +27,16 @@
# ci:slow:particle particle_render gcc clang tsan
# ci:slow:zero-alloc zero_alloc asan tsan
# ci:slow:ecs-stress ecs_stress asan tsan
# ci:slow:bench-50k laige_bench_render_50k_fpx16 / _fp32 gcc clang
# ci:slow:bench-50k laige_bench_render_50k_fpx16 / _fp32 and
# laige_bench_ac4_fpx16 / _fp32 (M2-AC-01) gcc clang
#
# Evidence (run 38312355303, per-entry times in s, threshold 30 s):
# laige_bench_render_50k_fpx16 gcc 577.05 clang 578.42
# laige_bench_render_50k_fp32 gcc 593.82 clang 566.52
# laige_bench_ac4_fpx16 / _fp32 NEW in M2-AC-01 — estimated ~60 s
# from the render-50k evidence (577 s over 4000 frames of 116k
# quads vs the ac4 frame's 10k quads); first-run times replace this
# estimate in the header once the entry has a master-run measurement
# laige-sim_tests gcc 59.97 clang 53.79 asan 124.03 tsan 159.52
# laige-render_tests gcc 53.83 clang 44.45 asan 43.20 tsan 60.41
# particle_render gcc 38.76 clang 34.67 asan 28.84 tsan 39.56
Expand Down Expand Up @@ -148,7 +153,7 @@ jobs:
# gated run.
run: |
ctest --test-dir build --output-on-failure \
-E '^(laige-sim_tests|laige-render_tests|particle_render|laige_bench_render_50k_fpx16|laige_bench_render_50k_fp32)$'
-E '^(laige-sim_tests|laige-render_tests|particle_render|laige_bench_render_50k_fpx16|laige_bench_render_50k_fp32|laige_bench_ac4_fpx16|laige_bench_ac4_fp32)$'

linux-clang:
name: Linux x64 (clang++)
Expand Down Expand Up @@ -182,7 +187,7 @@ jobs:
# gated run.
run: |
ctest --test-dir build --output-on-failure \
-E '^(laige-sim_tests|laige-render_tests|particle_render|laige_bench_render_50k_fpx16|laige_bench_render_50k_fp32)$'
-E '^(laige-sim_tests|laige-render_tests|particle_render|laige_bench_render_50k_fpx16|laige_bench_render_50k_fp32|laige_bench_ac4_fpx16|laige_bench_ac4_fp32)$'

linux-asan:
name: Linux x64 ASan+UBSan (clang++)
Expand Down Expand Up @@ -352,11 +357,11 @@ jobs:
run: ctest --test-dir build --output-on-failure -R '^particle_render$'

linux-gcc-bench-fpx16:
name: "Linux x64 (g++) slow bench: laige_bench_render_50k_fpx16"
name: "Linux x64 (g++) slow bench: laige_bench_render_50k_fpx16 + laige_bench_ac4_fpx16"
runs-on: ubuntu-24.04
# 30: build (~1.5 min) + the render-50k bench (577.05 s in the evidence
# run) plus runner variance. The pre-split 45-min exception minus
# the rest of the suite.
# run) + the ac4 bench (estimated ~60 s, M2-AC-01) plus runner
# variance. The pre-split 45-min exception minus the rest of the suite.
timeout-minutes: 30
if: contains(github.event.pull_request.labels.*.name, 'ci:slow:bench-50k')
steps:
Expand All @@ -373,14 +378,14 @@ jobs:
run: cmake -S . -B build -DCMAKE_BUILD_TYPE=Debug -DCMAKE_CXX_COMPILER=g++
- name: Build
run: cmake --build build -j
- name: "Test (slow bench: laige_bench_render_50k_fpx16)"
run: ctest --test-dir build --output-on-failure -R '^laige_bench_render_50k_fpx16$'
- name: "Test (slow bench: render-50k + ac4, fpx16)"
run: ctest --test-dir build --output-on-failure -R '^(laige_bench_render_50k_fpx16|laige_bench_ac4_fpx16)$'

linux-gcc-bench-fp32:
name: "Linux x64 (g++) slow bench: laige_bench_render_50k_fp32"
name: "Linux x64 (g++) slow bench: laige_bench_render_50k_fp32 + laige_bench_ac4_fp32"
runs-on: ubuntu-24.04
# 30: build (~1.5 min) + the render-50k bench (593.82 s in the evidence
# run) plus runner variance.
# run) + the ac4 bench (estimated ~60 s, M2-AC-01) plus runner variance.
timeout-minutes: 30
if: contains(github.event.pull_request.labels.*.name, 'ci:slow:bench-50k')
steps:
Expand All @@ -397,8 +402,8 @@ jobs:
run: cmake -S . -B build -DCMAKE_BUILD_TYPE=Debug -DCMAKE_CXX_COMPILER=g++
- name: Build
run: cmake --build build -j
- name: "Test (slow bench: laige_bench_render_50k_fp32)"
run: ctest --test-dir build --output-on-failure -R '^laige_bench_render_50k_fp32$'
- name: "Test (slow bench: render-50k + ac4, fp32)"
run: ctest --test-dir build --output-on-failure -R '^(laige_bench_render_50k_fp32|laige_bench_ac4_fp32)$'

linux-clang-sim:
name: "Linux x64 (clang++) slow: laige-sim_tests"
Expand Down Expand Up @@ -473,10 +478,10 @@ jobs:
run: ctest --test-dir build --output-on-failure -R '^particle_render$'

linux-clang-bench-fpx16:
name: "Linux x64 (clang++) slow bench: laige_bench_render_50k_fpx16"
name: "Linux x64 (clang++) slow bench: laige_bench_render_50k_fpx16 + laige_bench_ac4_fpx16"
runs-on: ubuntu-24.04
# 30: build (~1.5 min) + the render-50k bench (578.42 s in the evidence
# run) plus runner variance.
# run) + the ac4 bench (estimated ~60 s, M2-AC-01) plus runner variance.
timeout-minutes: 30
if: contains(github.event.pull_request.labels.*.name, 'ci:slow:bench-50k')
steps:
Expand All @@ -493,14 +498,14 @@ jobs:
run: cmake -S . -B build -DCMAKE_BUILD_TYPE=Debug -DCMAKE_CXX_COMPILER=clang++
- name: Build
run: cmake --build build -j
- name: "Test (slow bench: laige_bench_render_50k_fpx16)"
run: ctest --test-dir build --output-on-failure -R '^laige_bench_render_50k_fpx16$'
- name: "Test (slow bench: render-50k + ac4, fpx16)"
run: ctest --test-dir build --output-on-failure -R '^(laige_bench_render_50k_fpx16|laige_bench_ac4_fpx16)$'

linux-clang-bench-fp32:
name: "Linux x64 (clang++) slow bench: laige_bench_render_50k_fp32"
name: "Linux x64 (clang++) slow bench: laige_bench_render_50k_fp32 + laige_bench_ac4_fp32"
runs-on: ubuntu-24.04
# 30: build (~1.5 min) + the render-50k bench (566.52 s in the evidence
# run) plus runner variance.
# run) + the ac4 bench (estimated ~60 s, M2-AC-01) plus runner variance.
timeout-minutes: 30
if: contains(github.event.pull_request.labels.*.name, 'ci:slow:bench-50k')
steps:
Expand All @@ -517,8 +522,8 @@ jobs:
run: cmake -S . -B build -DCMAKE_BUILD_TYPE=Debug -DCMAKE_CXX_COMPILER=clang++
- name: Build
run: cmake --build build -j
- name: "Test (slow bench: laige_bench_render_50k_fp32)"
run: ctest --test-dir build --output-on-failure -R '^laige_bench_render_50k_fp32$'
- name: "Test (slow bench: render-50k + ac4, fp32)"
run: ctest --test-dir build --output-on-failure -R '^(laige_bench_render_50k_fp32|laige_bench_ac4_fp32)$'

linux-asan-sim:
name: "Linux x64 ASan+UBSan (clang++) slow: laige-sim_tests"
Expand Down Expand Up @@ -790,8 +795,10 @@ jobs:
golden-references:
name: Golden references (label-gated, ubuntu-24.04, g++, llvmpipe)
runs-on: ubuntu-24.04
# 30: build (~1.5 min) + two 1280x720 software-GL renders + font
# rasterization, plus runner variance.
# 30: build (~1.5 min) + four 1280x720 software-GL renders (the
# sprite + UI fixtures, and the two M2-AC-01 ac4 fixtures — two
# passes each, small scenes) + font rasterization, plus runner
# variance.
timeout-minutes: 30
if: contains(github.event.pull_request.labels.*.name, 'ci:golden-update')
steps:
Expand Down
54 changes: 31 additions & 23 deletions .github/workflows/ci.yml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -29,11 +29,18 @@
# slow-zero-alloc / ci:slow:zero-alloc zero_alloc asan tsan
# slow-ecs-stress / ci:slow:ecs-stress ecs_stress asan tsan
# slow-bench-50k / ci:slow:bench-50k laige_bench_render_50k_fpx16,
# laige_bench_render_50k_fp32 gcc clang
# laige_bench_render_50k_fp32,
# laige_bench_ac4_fpx16,
# laige_bench_ac4_fp32 (M2-AC-01) gcc clang
#
# Evidence (run 38312355303, per-entry times in s, threshold 30 s):
# laige_bench_render_50k_fpx16 gcc 577.05 clang 578.42
# laige_bench_render_50k_fp32 gcc 593.82 clang 566.52
# laige_bench_ac4_fpx16 / _fp32 NEW in M2-AC-01 — estimated ~60 s
# from the render-50k evidence (577 s over 4000 frames of 116k
# quads vs the ac4 frame's 10k quads); first-run times replace
# this estimate in the header once the entry has a master-run
# measurement
# laige-sim_tests gcc 59.97 clang 53.79 asan 124.03 tsan 159.52
# laige-render_tests gcc 53.83 clang 44.45 asan 43.20 tsan 60.41
# particle_render gcc 38.76 clang 34.67 asan 28.84 tsan 39.56
Expand Down Expand Up @@ -184,8 +191,9 @@
# (laige_bench_sim_tick_smoke): it exercises the workload and asserts
# its bit-identical fpx16_16 final_hash on every platform, without
# checking timing. The render-50k budget entries
# (laige_bench_render_50k_fpx16 / _fp32) are the SLOW bench jobs above —
# reference-platform-scoped, gated by ci:slow:bench-50k / the
# (laige_bench_render_50k_fpx16 / _fp32) and the AC-4 budget entries
# (laige_bench_ac4_fpx16 / _fp32, M2-AC-01) are the SLOW bench jobs
# above — reference-platform-scoped, gated by ci:slow:bench-50k / the
# slow-bench-50k checkbox, and run at every milestone gate (AGENTS.md
# §5).
#
Expand Down Expand Up @@ -223,7 +231,7 @@ on:
type: boolean
default: false
slow-bench-50k:
description: "Run the slow laige_bench_render_50k jobs (gcc, clang)"
description: "Run the slow bench jobs (laige_bench_render_50k + laige_bench_ac4, gcc, clang)"
type: boolean
default: false

Expand Down Expand Up @@ -271,7 +279,7 @@ jobs:
# gated run.
run: |
ctest --test-dir build --output-on-failure \
-E '^(laige-sim_tests|laige-render_tests|particle_render|laige_bench_render_50k_fpx16|laige_bench_render_50k_fp32)$'
-E '^(laige-sim_tests|laige-render_tests|particle_render|laige_bench_render_50k_fpx16|laige_bench_render_50k_fp32|laige_bench_ac4_fpx16|laige_bench_ac4_fp32)$'

linux-clang:
name: Linux x64 (clang++)
Expand Down Expand Up @@ -306,7 +314,7 @@ jobs:
# gated run.
run: |
ctest --test-dir build --output-on-failure \
-E '^(laige-sim_tests|laige-render_tests|particle_render|laige_bench_render_50k_fpx16|laige_bench_render_50k_fp32)$'
-E '^(laige-sim_tests|laige-render_tests|particle_render|laige_bench_render_50k_fpx16|laige_bench_render_50k_fp32|laige_bench_ac4_fpx16|laige_bench_ac4_fp32)$'

linux-asan:
name: Linux x64 ASan+UBSan (clang++)
Expand Down Expand Up @@ -505,11 +513,11 @@ jobs:
run: ctest --test-dir build --output-on-failure -R '^particle_render$'

linux-gcc-bench-fpx16:
name: "Linux x64 (g++) slow bench: laige_bench_render_50k_fpx16"
name: "Linux x64 (g++) slow bench: laige_bench_render_50k_fpx16 + laige_bench_ac4_fpx16"
runs-on: ubuntu-24.04
# 30: build (~1.5 min) + the render-50k bench (577.05 s in the evidence
# run) plus runner variance. The pre-split 45-min exception minus
# the rest of the suite.
# run) + the ac4 bench (estimated ~60 s, M2-AC-01) plus runner
# variance. The pre-split 45-min exception minus the rest of the suite.
timeout-minutes: 30
if: github.event.inputs.slow-bench-50k == 'true'
steps:
Expand All @@ -526,14 +534,14 @@ jobs:
run: cmake -S . -B build -DCMAKE_BUILD_TYPE=Debug -DCMAKE_CXX_COMPILER=g++
- name: Build
run: cmake --build build -j
- name: "Test (slow bench: laige_bench_render_50k_fpx16)"
run: ctest --test-dir build --output-on-failure -R '^laige_bench_render_50k_fpx16$'
- name: "Test (slow bench: render-50k + ac4, fpx16)"
run: ctest --test-dir build --output-on-failure -R '^(laige_bench_render_50k_fpx16|laige_bench_ac4_fpx16)$'

linux-gcc-bench-fp32:
name: "Linux x64 (g++) slow bench: laige_bench_render_50k_fp32"
name: "Linux x64 (g++) slow bench: laige_bench_render_50k_fp32 + laige_bench_ac4_fp32"
runs-on: ubuntu-24.04
# 30: build (~1.5 min) + the render-50k bench (593.82 s in the evidence
# run) plus runner variance.
# run) + the ac4 bench (estimated ~60 s, M2-AC-01) plus runner variance.
timeout-minutes: 30
if: github.event.inputs.slow-bench-50k == 'true'
steps:
Expand All @@ -550,8 +558,8 @@ jobs:
run: cmake -S . -B build -DCMAKE_BUILD_TYPE=Debug -DCMAKE_CXX_COMPILER=g++
- name: Build
run: cmake --build build -j
- name: "Test (slow bench: laige_bench_render_50k_fp32)"
run: ctest --test-dir build --output-on-failure -R '^laige_bench_render_50k_fp32$'
- name: "Test (slow bench: render-50k + ac4, fp32)"
run: ctest --test-dir build --output-on-failure -R '^(laige_bench_render_50k_fp32|laige_bench_ac4_fp32)$'

linux-clang-sim:
name: "Linux x64 (clang++) slow: laige-sim_tests"
Expand Down Expand Up @@ -626,10 +634,10 @@ jobs:
run: ctest --test-dir build --output-on-failure -R '^particle_render$'

linux-clang-bench-fpx16:
name: "Linux x64 (clang++) slow bench: laige_bench_render_50k_fpx16"
name: "Linux x64 (clang++) slow bench: laige_bench_render_50k_fpx16 + laige_bench_ac4_fpx16"
runs-on: ubuntu-24.04
# 30: build (~1.5 min) + the render-50k bench (578.42 s in the evidence
# run) plus runner variance.
# run) + the ac4 bench (estimated ~60 s, M2-AC-01) plus runner variance.
timeout-minutes: 30
if: github.event.inputs.slow-bench-50k == 'true'
steps:
Expand All @@ -646,14 +654,14 @@ jobs:
run: cmake -S . -B build -DCMAKE_BUILD_TYPE=Debug -DCMAKE_CXX_COMPILER=clang++
- name: Build
run: cmake --build build -j
- name: "Test (slow bench: laige_bench_render_50k_fpx16)"
run: ctest --test-dir build --output-on-failure -R '^laige_bench_render_50k_fpx16$'
- name: "Test (slow bench: render-50k + ac4, fpx16)"
run: ctest --test-dir build --output-on-failure -R '^(laige_bench_render_50k_fpx16|laige_bench_ac4_fpx16)$'

linux-clang-bench-fp32:
name: "Linux x64 (clang++) slow bench: laige_bench_render_50k_fp32"
name: "Linux x64 (clang++) slow bench: laige_bench_render_50k_fp32 + laige_bench_ac4_fp32"
runs-on: ubuntu-24.04
# 30: build (~1.5 min) + the render-50k bench (566.52 s in the evidence
# run) plus runner variance.
# run) + the ac4 bench (estimated ~60 s, M2-AC-01) plus runner variance.
timeout-minutes: 30
if: github.event.inputs.slow-bench-50k == 'true'
steps:
Expand All @@ -670,8 +678,8 @@ jobs:
run: cmake -S . -B build -DCMAKE_BUILD_TYPE=Debug -DCMAKE_CXX_COMPILER=clang++
- name: Build
run: cmake --build build -j
- name: "Test (slow bench: laige_bench_render_50k_fp32)"
run: ctest --test-dir build --output-on-failure -R '^laige_bench_render_50k_fp32$'
- name: "Test (slow bench: render-50k + ac4, fp32)"
run: ctest --test-dir build --output-on-failure -R '^(laige_bench_render_50k_fp32|laige_bench_ac4_fp32)$'

linux-asan-sim:
name: "Linux x64 ASan+UBSan (clang++) slow: laige-sim_tests"
Expand Down
8 changes: 8 additions & 0 deletions budgets.json
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -42,6 +42,14 @@
"measured": 60.4450,
"workload": "50k visible sprites (worst-case isometric overlap), 3 parallax layers, UI (PRD 8.1)"
},
{
"name": "sprites_10k_frame",
"metric": "mean",
"unit": "ms",
"target": 16.7,
"measured": 12.1453,
"workload": "10 000 overlapping sprites (100 x 10 grid, 10 stacked per cell), whole frame sorted + drawn offscreen at 60 FPS (M2-AC-01 AC-4.3, PRD 4)"
},
{
"name": "iso_depthkey_rebuild",
"metric": "mean",
Expand Down
Loading
Loading