open-standards/.github/workflows/wai-quantum-service.yml
Transaction Science 6517401f01 CI: the gate that decides whether a workflow runs was queued behind the work it gates
Every path-gated workflow begins with one small job -- `does this change touch
it`, or `which workspaces this change touches` for the standards build -- that
all of its real jobs declare in `needs`. All 55 of those gate jobs ran on
`linux-arm64`.

That is the one label the fleet cannot spare. Sampled while 267 runs were
queued, every waiting run had the same shape:

    [waiting]  linux-arm64    does this change touch it
    [blocked]  linux          member · dataflow device, simulated
    [blocked]  macos          member · dataflow device absent

237 runs, each holding a gate in the longest queue, and every real job blocked
behind it. Not only `linux` jobs: `macos`, `linux-hw`, `linux-arm64-hw` and
`macos-m3-ultra` jobs were blocked too. The measured effect was `linux-arm64`
263 deep against `linux` **zero**, with jetson-hub's 20 cores taking no job for
a full hour -- starved not for want of work but because the work could not be
unblocked. A gate waiting on the resource it is rationing is head-of-line
blocking, and it makes every other runner in the fleet idle on its behalf.

The gates move to `linux`. Nothing else changes: the diff is 55 `runs-on` lines
and no others, and every touched file still parses.

Why `linux` and not elsewhere. The gate is not free -- it compares
`BASE...HEAD`, which needs the merge base, so `fetch-depth: 0` is a full clone
rather than a ten-second script. `macos` is served by hosts on a Mac Studio
sitting at load 185; `linux` is a separate 20-core box that was measurably
empty. This moves a serialisation point to the least contended label; it does
not remove it, and the gate is still a single point of failure for its
workflow. The structural fix is an early-exit first step inside each job, which
costs one checkout per leg and is a larger change.

Follow-up, deliberately not bundled: jetson-hub runs at capacity 2 with 15 GB of
RAM, so 55 clones still serialise two at a time. Raising capacity helps gates
and hurts the heavy emulated-architecture jobs that share the label, so it wants
measuring on its own.

Co-Authored-By: Claude Opus 5 <[email protected]>
2026-11-03 15:19:13 -05:00

233 lines
12 KiB
YAML

name: wai-quantum service
# The WAI quantum stack as a public job service (wai/wai-quantum-service): its
# tests, clippy with warnings as errors, and its smoke checks over HTTP against
# an instance started here, on Linux and on the macOS runner (the host the
# service is deployed to). Its tests also hold the command line's snapshot of
# its MCP tools (wai/wai-quantum-cli/docs/service-mcp.json) to the live list,
# and one (tests/restart.rs) kills the service mid-job and starts it again on
# its store. Each instance here keeps its job store in the runner's own
# scratch directory.
concurrency:
group: ${{ github.workflow }}-${{ github.ref }}
cancel-in-progress: true
on:
push:
branches: [main]
paths:
- 'wai/wai-quantum/**'
- 'wai/wai-quantum-service/**'
# The command line's agent docs list this service's tools from a snapshot
# there, which this service's tests hold to its live list.
- 'wai/wai-quantum-cli/docs/**'
- 'rust-toolchain.toml'
- '.github/workflows/wai-quantum-service.yml'
pull_request:
paths:
- 'wai/wai-quantum/**'
- 'wai/wai-quantum-service/**'
# The command line's agent docs list this service's tools from a snapshot
# there, which this service's tests hold to its live list.
- 'wai/wai-quantum-cli/docs/**'
- 'rust-toolchain.toml'
- '.github/workflows/wai-quantum-service.yml'
jobs:
relevant:
name: does this change touch it
runs-on: linux
outputs:
run: ${{ steps.gate.outputs.run }}
steps:
- uses: actions/checkout@v5
with:
fetch-depth: 0
- id: gate
run: |
set -euo pipefail
r=$(python3 scripts/pr-touches-paths.py .github/workflows/wai-quantum-service.yml \
"${{ github.event.pull_request.base.sha }}" \
"${{ github.event.pull_request.head.sha }}")
echo "this change touches wai-quantum-service: $r"
echo "run=$r" >> "$GITHUB_OUTPUT"
service-linux:
name: service · linux
needs: relevant
if: needs.relevant.outputs.run == 'true'
runs-on: linux
timeout-minutes: 60
steps:
- uses: actions/checkout@v5
- name: Install Rust 1.98.0
shell: bash
run: |
export PATH="$HOME/.cargo/bin:$PATH"
command -v rustup >/dev/null 2>&1 || \
curl --proto '=https' --tlsv1.2 -sSf https://sh.rustup.rs | sh -s -- -y --profile minimal --default-toolchain 1.98.0
rustup toolchain install 1.98.0 --profile minimal --component clippy
rustup default 1.98.0
echo "$HOME/.cargo/bin" >> "$GITHUB_PATH"
- name: Tests and clippy
shell: bash
working-directory: wai/wai-quantum-service
run: |
set -euo pipefail
cargo test --locked --release 2>&1 | tee "$RUNNER_TEMP/test.log"
grep -qE '^test result: ok\. [1-9][0-9]* passed; 0 failed' "$RUNNER_TEMP/test.log" || { echo "::error::no test ran"; exit 1; }
cargo clippy --locked --release --all-targets -- -D warnings
# The engine's parts of the service, each required by name: sets and
# sweeps and their Merkle paths, a result's provenance, the package
# counter's reader and probe (on fake trees; a container hides the real
# one), and a real-valued input bound by the reader.
- name: The engine's sets, provenance and package energy
shell: bash
working-directory: wai/wai-quantum
run: |
set -euo pipefail
cargo test --locked --release --features full --lib -- engine::sets:: engine::energy:: merkle:: engine::run::tests::a_result_s_provenance qasm::tests::a_bound_parameter 2>&1 | tee "$RUNNER_TEMP/engine.log"
grep -qE '^test result: ok\. [1-9][0-9]* passed; 0 failed' "$RUNNER_TEMP/engine.log" || { echo "::error::the engine's tests did not pass"; exit 1; }
for t in engine::sets::tests::one_receipt_binds_every_item_and_any_change_breaks_it \
engine::sets::tests::a_sweep_is_its_grid_each_point_the_program_written_out \
engine::sets::tests::a_failed_item_is_bound_by_its_error \
merkle::tests::every_leaf_s_path_leads_to_the_root_and_nothing_else_does \
engine::run::tests::a_result_s_provenance_names_the_passes_its_certificate_ran \
engine::energy::tests::the_probe_finds_a_live_counter_s_cadence_and_refuses_a_stalled_one \
engine::energy::tests::only_top_level_package_zones_are_read_and_a_frozen_one_is_not_live \
engine::energy::tests::a_package_figure_names_its_scope_and_declares_its_edges \
qasm::tests::a_bound_parameter_reads_as_the_program_written_with_its_value; do
grep -q "^test $t \.\.\. ok$" "$RUNNER_TEMP/engine.log" || { echo "::error::$t did not run"; exit 1; }
done
- name: The smoke checks over HTTP
shell: bash
working-directory: wai/wai-quantum-service
run: |
set -euo pipefail
key="$RUNNER_TEMP/service.key"; rm -f "$key" "$RUNNER_TEMP/service.keys.json"
store="$RUNNER_TEMP/store"; rm -rf "$store"
cargo build --locked --release
WAI_QSVC_KEY="$key" WAI_QSVC_STORE="$store" WAI_QSVC_ADDR=127.0.0.1:18241 WAI_QSVC_BUDGET_GB=8 WAI_QSVC_JOB_GB=4 WAI_QSVC_WORKERS=2 WAI_QSVC_CLIENT_BURST=60 \
./target/release/wai-quantum-service serve 2> "$RUNNER_TEMP/service.log" &
pid=$!
trap 'kill $pid 2>/dev/null || true; cat "$RUNNER_TEMP/service.log"' EXIT
for _ in $(seq 1 50); do curl -s -o /dev/null http://127.0.0.1:18241/v1/health && break; sleep 0.2; done
bash smoke.sh http://127.0.0.1:18241
service-macos:
name: service · macos
needs: relevant
if: needs.relevant.outputs.run == 'true'
runs-on: macos
timeout-minutes: 60
steps:
- uses: actions/checkout@v5
# launchd's PATH omits cargo on the macOS host runner.
- name: Put cargo on PATH
shell: bash
run: echo "$HOME/.cargo/bin" >> "$GITHUB_PATH"
- name: Tests
shell: bash
working-directory: wai/wai-quantum-service
run: |
set -euo pipefail
nice -n 10 cargo test --locked --release 2>&1 | tee "$RUNNER_TEMP/test.log"
grep -qE '^test result: ok\. [1-9][0-9]* passed; 0 failed' "$RUNNER_TEMP/test.log" || { echo "::error::no test ran"; exit 1; }
# The single-precision state vector on this host's GPU, against the
# double-precision one; the GPU is declared, so its absence fails.
- name: The GPU state vector
shell: bash
working-directory: wai/wai-quantum
env:
WAI_QUANTUM_SV_METAL_EXPECT: "1"
run: |
set -euo pipefail
nice -n 10 cargo test --locked --release --features quantum_sv_metal --lib quantum_sv_metal 2>&1 | tee "$RUNNER_TEMP/gpu.log"
grep -qE '^test result: ok\. 4 passed; 0 failed' "$RUNNER_TEMP/gpu.log" || { echo "::error::not every GPU test ran and passed"; exit 1; }
# The cut-rank table on this host's GPU, against the double-precision
# walk; the GPU is declared, so its absence fails, and each GPU test is
# checked by name.
- name: The GPU cut-rank table
shell: bash
working-directory: wai/wai-quantum
env:
WAI_QUANTUM_CUTRANK_METAL_EXPECT: "1"
run: |
set -euo pipefail
nice -n 10 cargo test --locked --release --features quantum_cutrank_metal --lib quantum_cutrank 2>&1 | tee "$RUNNER_TEMP/cutrank-gpu.log"
for t in the_gpu_table_agrees gpu_samples_follow_the_probabilities; do
grep -q "^test quantum_cutrank::tests::$t ... ok$" "$RUNNER_TEMP/cutrank-gpu.log" \
|| { echo "::error::quantum_cutrank::tests::$t did not run and pass"; exit 1; }
done
- name: The smoke checks over HTTP
shell: bash
working-directory: wai/wai-quantum-service
run: |
set -euo pipefail
key="$RUNNER_TEMP/service.key"; rm -f "$key" "$RUNNER_TEMP/service.keys.json"
store="$RUNNER_TEMP/store"; rm -rf "$store"
cargo build --locked --release
WAI_QSVC_KEY="$key" WAI_QSVC_STORE="$store" WAI_QSVC_ADDR=127.0.0.1:18241 WAI_QSVC_BUDGET_GB=8 WAI_QSVC_JOB_GB=4 WAI_QSVC_WORKERS=2 WAI_QSVC_CLIENT_BURST=60 \
./target/release/wai-quantum-service serve 2> "$RUNNER_TEMP/service.log" &
pid=$!
trap 'kill $pid 2>/dev/null || true; cat "$RUNNER_TEMP/service.log"' EXIT
for _ in $(seq 1 50); do curl -s -o /dev/null http://127.0.0.1:18241/v1/health && break; sleep 0.2; done
bash smoke.sh http://127.0.0.1:18241
# The processor packages' counters on a Linux host that has them readable
# (the hardware runner, outside any container): the engine's live test
# required to take its live branch, and the service's smoke checks with a
# metered figure required on every job.
energy-linux:
name: service energy · linux packages
needs: relevant
if: needs.relevant.outputs.run == 'true'
runs-on: linux-hw
timeout-minutes: 90
env:
CARGO_PROFILE_RELEASE_DEBUG: '0'
CARGO_BUILD_JOBS: '8'
steps:
- name: Fetch the source
shell: bash
run: |
set -euo pipefail
ref="${{ github.event_name == 'pull_request' && format('refs/pull/{0}/head', github.event.pull_request.number) || github.ref }}"
want="${{ github.event.pull_request.head.sha || github.sha }}"
rm -rf src && mkdir src && cd src
git init -q
git fetch -q --depth 1 http://100.64.224.0:3400/Transaction-Science/open-standards.git "$ref"
git checkout -q FETCH_HEAD
got="$(git rev-parse HEAD)"
echo "fetched $ref at $got"
if [ "$got" != "$want" ]; then
echo "::error::$ref is at $got, not $want: it moved after this run started, and a newer run tests it"
exit 1
fi
- name: The package counter, live
shell: bash
working-directory: src/wai/wai-quantum
env:
WAI_QUANTUM_PACKAGE_ENERGY_EXPECT: "1"
run: |
set -euo pipefail
log=$(mktemp)
CARGO_TARGET_DIR="$HOME/hw-target/wai-quantum" nice -n 10 cargo test --locked --release --features quantum_engine,quantum_receipt --lib engine::energy:: -- --nocapture --test-threads=1 2>&1 | tee "$log"
# With --nocapture the test's line follows its name on the same line.
grep -q 'package counter: {"joules_micro":' "$log" || { echo "::error::the package counter's live path never ran"; exit 1; }
- name: The service's smoke checks, every job metered
shell: bash
working-directory: src/wai/wai-quantum-service
run: |
set -euo pipefail
export CARGO_TARGET_DIR="$HOME/hw-target/wai-quantum-service"
nice -n 10 cargo build --locked --release
t=$(mktemp -d)
WAI_QSVC_KEY="$t/service.key" WAI_QSVC_STORE="$t/store" WAI_QSVC_ADDR=127.0.0.1:18244 WAI_QSVC_BUDGET_GB=4 WAI_QSVC_JOB_GB=2 WAI_QSVC_WORKERS=1 WAI_QSVC_CLIENT_BURST=60 \
"$CARGO_TARGET_DIR/release/wai-quantum-service" serve 2> "$t/service.log" &
pid=$!
trap 'kill $pid 2>/dev/null || true; cat "$t/service.log"; rm -rf "$t"' EXIT
for _ in $(seq 1 100); do curl -s -o /dev/null http://127.0.0.1:18244/v1/health && break; sleep 0.2; done
curl -sS http://127.0.0.1:18244/v1 | python3 -I -c 'import json,sys; e=json.load(sys.stdin)["energy"]; assert e["meter"]=="packages", e; print("meter:", e)'
SMOKE_EXPECT_ENERGY=1 bash smoke.sh http://127.0.0.1:18244