mirror of
https://github.com/MobileGL-Dev/MobileGL
synced 2026-11-07 18:13:23 +09:00
[CI] (.github/workflows, docs/ci, tools/trace_replay): drop retrace-split's repeated inproc rows
- with build-linux reused, retrace-split runs its spawn rows plus OpenRA's inproc rows (the wire evidence); the other 75 inproc rows are replayed by the retrace job's inproc-acceptance rows on the same library - trace_cases.py: github-split-matrix takes --transports and --keep-inproc; the self-test proves the spawn rows unchanged and every dropped inproc row an inproc-acceptance row; only a narrowed matrix that comes out empty raises, so the workflow's named error for an empty full subset still prints - trace-cases: the spawn-only subset is only a saving, so a failed self-test or an empty result is a warning, the output is not written and retrace-split takes the full subset - run_trace_case.cmake: MGTRACE_EXPECT_IPC (a harness name) requires the client's IPC line to carry the listed tokens; retrace-split legs prove strict=0 role-split-state=1 run-ahead=1 and a changed default reds every leg with the rows to restore; the acceptance rows prove strict=1 role-split-state=1 run-ahead=1 and gain the refusal census; canned tests in test_run_trace_case.py
This commit is contained in:
@@ -2648,6 +2648,7 @@ jobs:
|
||||
names: ${{ steps.trace-cases.outputs.names }}
|
||||
verify-matrix: ${{ steps.trace-cases.outputs.verify-matrix }}
|
||||
split-matrix: ${{ steps.trace-cases.outputs.split-matrix }}
|
||||
split-spawn-matrix: ${{ steps.trace-cases.outputs.split-spawn-matrix }}
|
||||
steps:
|
||||
- name: Checkout repo
|
||||
uses: actions/checkout@v6
|
||||
@@ -2699,6 +2700,22 @@ jobs:
|
||||
exit 1
|
||||
fi
|
||||
echo "split-matrix=${SPLIT_MATRIX}" >> "$GITHUB_OUTPUT"
|
||||
# On the disaggregation branches, the same subset without the inproc rows the
|
||||
# inproc-acceptance retrace rows replay on the same library, which retrace-split uses when
|
||||
# it reuses build-linux's library. OpenRA keeps its inproc rows: they carry the
|
||||
# wire-volume evidence. Its self-test proves every dropped row is an acceptance row. It is
|
||||
# only a saving, so it never fails this job: on a failed self-test or an empty subset it
|
||||
# is not written, and retrace-split takes the full subset, as it does elsewhere.
|
||||
if [ "${MOBILEGL_CI_DISAGGREGATED}" = "ON" ]; then
|
||||
if python3 tools/trace_replay/trace_cases.py --self-test-split-matrix \
|
||||
&& SPLIT_SPAWN_MATRIX=$(python3 tools/trace_replay/trace_cases.py --ci --format github-split-matrix \
|
||||
--transports spawn --keep-inproc OpenRA) \
|
||||
&& [ "$(printf '%s' "${SPLIT_SPAWN_MATRIX}" | python3 -c 'import json,sys; print(len(json.load(sys.stdin)["include"]))' || echo 0)" -ge 1 ]; then
|
||||
echo "split-spawn-matrix=${SPLIT_SPAWN_MATRIX}" >> "$GITHUB_OUTPUT"
|
||||
else
|
||||
echo "::warning::the spawn-only split retrace subset failed its self-test or is empty, so retrace-split replays the full split subset"
|
||||
fi
|
||||
fi
|
||||
|
||||
trace-fixtures:
|
||||
name: trace fixture (${{ matrix.case }})
|
||||
@@ -2873,6 +2890,11 @@ jobs:
|
||||
&& [ '${{ matrix.case }}' = 'improved-transparency-minecraft-26.3' ]; then
|
||||
export MOBILEGL_MAGMA_DISABLE_BLENDED_DEPTH_WRITE=1
|
||||
fi
|
||||
# The inproc rows stand in for retrace-split's dropped inproc rows, so they prove the
|
||||
# configuration they run with from the client's own log.
|
||||
if [ "${MOBILEGL_TRANSPORT}" = inproc ]; then
|
||||
export MGTRACE_EXPECT_IPC='strict=1 role-split-state=1 run-ahead=1'
|
||||
fi
|
||||
ctest -V --no-tests=error -R '^MobileGLTraceReplay\.${{ matrix.case }}\.${{ matrix.backend }}$' --output-junit "${RUNNER_TEMP}/retrace.xml"
|
||||
# A split row's transport assertions ran on this leg and found their evidence.
|
||||
if [ "${MOBILEGL_TRANSPORT}" != monolith ]; then
|
||||
@@ -2881,6 +2903,33 @@ jobs:
|
||||
--transport "${MOBILEGL_TRANSPORT}" --case '${{ matrix.case }}' --backend '${{ matrix.backend }}'
|
||||
fi
|
||||
|
||||
# The refusal census, recorded rather than gated. run_trace_case.cmake already REDS the case
|
||||
# on any Fatal{, so reaching here means the count is zero - but the number and the distinct
|
||||
# slot names are what MEASUREMENTS wants from every split run, and reading them out of the
|
||||
# library's own log is the only way to get them (ctest -V's transcript is a false zero).
|
||||
- name: Refusal census from the library log
|
||||
if: ${{ always() && matrix.transport == 'inproc' }}
|
||||
working-directory: build-retrace/tools/trace_replay
|
||||
run: |
|
||||
# BOTH ROLES, because under spawn every applier refusal is raised in the SERVER's
|
||||
# process and a census reading only the client's file reports a confident zero for the
|
||||
# half it cannot see.
|
||||
log="${{ matrix.case }}/${{ matrix.backend }}/output/mobilegl.client.log"
|
||||
server_log="${{ matrix.case }}/${{ matrix.backend }}/output/mobilegl.server.log"
|
||||
cat "${{ matrix.case }}/${{ matrix.backend }}/output/split-evidence.txt" 2>/dev/null || echo "no split-evidence.txt"
|
||||
if [ ! -f "${log}" ]; then
|
||||
echo "no ${log} - the replay wrote no library log"
|
||||
exit 0
|
||||
fi
|
||||
census_files="${log}"
|
||||
if [ -f "${server_log}" ]; then census_files="${census_files} ${server_log}"; fi
|
||||
echo "census over: ${census_files}"
|
||||
echo "Fatal{ lines: $(cat ${census_files} | grep -c 'Fatal{' || true)"
|
||||
cat ${census_files} | grep -o 'Fatal{[A-Za-z]*, "[^"]*"' | sort | uniq -c | sort -rn | head -20 || true
|
||||
# The transport line is the CLIENT's: ConfigLoader resolves it there, and under spawn
|
||||
# the server states its own separately.
|
||||
grep -m1 "MOBILEGL_TRANSPORT=" "${log}" || echo "no transport line in ${log}"
|
||||
|
||||
- name: Annotate failed entries
|
||||
if: failure()
|
||||
run: python3 scripts/ci/junit_annotate.py --excerpts 1 "${RUNNER_TEMP}"
|
||||
@@ -3185,7 +3234,10 @@ jobs:
|
||||
strategy:
|
||||
fail-fast: false
|
||||
max-parallel: 8
|
||||
matrix: ${{ fromJSON(needs.trace-cases.outputs.split-matrix) }}
|
||||
# With build-linux reused, the inproc rows (except OpenRA's) are replayed by the retrace job's
|
||||
# inproc-acceptance rows on the same library, strict and with the dual block and run-ahead
|
||||
# on; the spawn rows below prove per leg that those are also the defaults they run with.
|
||||
matrix: ${{ fromJSON(needs.build-linux-split.outputs.reuse-primary == 'true' && needs.trace-cases.outputs.split-spawn-matrix || needs.trace-cases.outputs.split-matrix) }}
|
||||
|
||||
steps:
|
||||
- name: Set swap space
|
||||
@@ -3350,6 +3402,10 @@ jobs:
|
||||
export MGTRACE_WIRE_EVIDENCE=1
|
||||
wire=(--require-wire)
|
||||
fi
|
||||
if [ "${SPLIT_REUSE}" = "true" ]; then
|
||||
export MGTRACE_EXPECT_IPC='strict=0 role-split-state=1 run-ahead=1'
|
||||
export MGTRACE_EXPECT_IPC_REASON="retrace-split dropped its inproc rows because the inproc-acceptance retrace rows replay the same library with strict=1 role-split-state=1 run-ahead=1, which differs from this leg's defaults only in strict while these hold. A default changed: restore the inproc rows (the trace-cases job's split-spawn-matrix without --transports spawn) or update the expectation."
|
||||
fi
|
||||
ctest -V --no-tests=error --timeout 10800 \
|
||||
-R '^MobileGLTraceReplay\.${{ matrix.case }}\.${{ matrix.backend }}$'
|
||||
# The transport assertions ran on this leg and found their evidence.
|
||||
|
||||
Reference in New Issue
Block a user