open-standards/.github/workflows/wai-quantum-cli.yml
Transaction Science 888cc3ba20 CI: a check holds every workflow to the one pinned wasmtime and both its settings
The install steps' comment said every WAI workflow that runs wasmtime pins one
version and passes both determinism settings, but nothing enforced it: a step
that ran a stock or differently pinned wasmtime would have merged green.
scripts/check-wasmtime-pin.py now fails, in the repository checks, any workflow
that

1. downloads a wasmtime release other than 50.0.1, or checks it against any
   SHA-256 but that archive's;
2. installs wasmtime another way (cargo install, the install script, an action);
3. runs wasmtime (run, serve, compile, or a module given to it) or names it as
   a CARGO_TARGET_*_RUNNER without -W nan-canonicalization=y and
   -W relaxed-simd-deterministic=y.

It also fails when it finds no download or no run, so a check that looked in
the wrong place cannot pass. On this tree it finds the downloads and runs it
should. Its self-test (scripts/check-wasmtime-pin-selftest.sh) passes a scratch
workflow shaped like these and fails each way of breaking it that the
self-test lists, matching the reason in the output.

The comment's relaxed_madd claim was measured on arm64 only. It now rests on
both architectures, measured on an arm64 macOS host and an x86_64 Linux host
with a small probe (wasmtime 50.0.1's release for the host, checked against
the digest the release lists):

    host Darwin arm64; wasmtime 50.0.1 (c307ec81d 2026-10-29)
    f32.add(bits 0x7fa00001, 1.0)                     [no flags]: 0x7fe00001
    f32x4.relaxed_madd(1+2^-12, 1+2^-12, -(1+2^-11)) [no flags]: 0x33800000
    f32.add(bits 0x7fa00001, 1.0)                     [both flags]: 0x7fc00000
    f32x4.relaxed_madd(1+2^-12, 1+2^-12, -(1+2^-11)) [both flags]: 0x33800000
    an unknown -W option is refused: error: failed to parse -W / --wasm option:
      unknown option: no-such-option

    host Linux x86_64; wasmtime 50.0.1 (c307ec81d 2026-10-29)
    (the same five lines, bit for bit)

0x33800000 is 2^-24, the fused result; the unfused fadd(fmul) gives 0. So the
NaN setting takes effect on both, and relaxed_madd is fused on both with and
without the relaxed setting. That matches 50.0.1's source (the
wasmtime-internal-cranelift crate, read from the registry): the translator
lowers F32x4/F64x2RelaxedMadd through fma_f32x4/fma_f64x2
(src/translate/code_translator.rs L2530-2539), which emit a native fma on a
host that has one, a fused library call under relaxed-simd-deterministic on a
host that has none, and the unfused fmul and fadd only on a host with no FMA
and the setting off (src/func_environ.rs L5907-5925). The setting is what
makes the result fused on every host. And since an unknown -W option stops
wasmtime, a run that exits 0 with the two flags has parsed them.

Co-Authored-By: Claude Opus 5.5 <[email protected]>
2026-11-04 12:28:47 -05:00

211 lines
13 KiB
YAML

name: wai-quantum command line
# The wai-quantum command line (wai/wai-quantum-cli): its tests and clippy with
# warnings as errors, with and without `submit`; its smoke checks (smoke.sh: lint,
# plan, run, verify and a forged result refused, key, draw, MCP over stdio, the
# research tools); and the WASI component, whose output under wasmtime equals the
# native build's byte for byte. On macOS, also with the GPU state vector. Its tests
# include the model harness (src/eval.rs): every task's reference solved and an
# empty program not, a scripted run, a model server on this machine asked and one
# elsewhere refused, and the corpus (wai/quantum-conformance/eval/) held to the
# task set.
concurrency:
group: ${{ github.workflow }}-${{ github.ref }}
cancel-in-progress: true
on:
push:
branches: [main]
paths:
- 'wai/wai-quantum/**'
- 'wai/wai-quantum-cli/**'
- 'wai/wai-quantum-qpu/**'
- 'wai/quantum-conformance/eval/**'
- 'rust-toolchain.toml'
- '.github/workflows/wai-quantum-cli.yml'
pull_request:
paths:
- 'wai/wai-quantum/**'
- 'wai/wai-quantum-cli/**'
- 'wai/wai-quantum-qpu/**'
- 'wai/quantum-conformance/eval/**'
- 'rust-toolchain.toml'
- '.github/workflows/wai-quantum-cli.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-cli.yml \
"${{ github.event.pull_request.base.sha }}" \
"${{ github.event.pull_request.head.sha }}")
echo "this change touches the wai-quantum command line: $r"
echo "run=$r" >> "$GITHUB_OUTPUT"
cli-linux:
name: command line · linux
needs: relevant
if: needs.relevant.outputs.run == 'true'
runs-on: linux
timeout-minutes: 60
steps:
- uses: actions/checkout@v5
- name: Install Rust 1.98.0 with clippy and the wasm32-wasip2 target
shell: bash
run: |
export PATH="$HOME/.cargo/bin:$PATH"
command -v rustup >/dev/null 2>&1 || \
curl --proto '=https' --tlsv1.2 -sSf https://sh.rustup.rs | sh -s -- -y --profile minimal --default-toolchain 1.98.0
rustup toolchain install 1.98.0 --profile minimal --component clippy
rustup default 1.98.0
rustup target add --toolchain 1.98.0 wasm32-wasip2
echo "$HOME/.cargo/bin" >> "$GITHUB_PATH"
- name: Tests and clippy, with and without submit
shell: bash
working-directory: wai/wai-quantum-cli
run: |
set -euo pipefail
cargo test --locked --release 2>&1 | tee "$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; }
# Your own providers end to end (tests/providers.rs), against the
# reference provider this binary serves, and their jobs as MCP
# tasks, each test by name.
for t in a_person_adds_tests_plans_and_runs_on_a_provider_and_the_receipt_verifies \
a_run_that_is_not_free_is_refused_until_the_person_confirms_it \
a_credential_is_kept_in_a_file_only_its_owner_reads_and_appears_in_nothing_printed \
a_provider_over_http_runs_with_its_credential_in_the_header \
an_agent_lists_estimates_is_refused_and_runs_only_with_confirm_cost \
an_agent_runs_on_a_provider_as_a_task_and_the_server_answers_while_it_waits \
providers::tasks::tests::a_job_reads_as_its_task \
providers::tasks::tests::a_cancelled_request_ends_its_wait_and_a_change_wakes_it; do
grep -q "^test $t \.\.\. ok$" "$RUNNER_TEMP/test.log" || { echo "::error::$t did not run and pass"; exit 1; }
done
# The model harness and the views in an agent's host, each test by name.
for t in eval::tests::every_task_s_reference_is_solved_and_an_empty_program_is_not \
eval::tests::a_scripted_run_reports_what_the_engine_found \
eval::tests::a_model_on_this_machine_is_asked_and_one_elsewhere_is_refused \
eval::tests::the_corpus_is_the_task_set_written_three_ways \
eval::tests::an_answer_is_read_for_its_program_and_scored \
mcp::tests::a_host_that_draws_views_gets_them_and_one_that_does_not_gets_words; do
grep -q "^test $t \.\.\. ok$" "$RUNNER_TEMP/test.log" || { echo "::error::$t did not run and pass"; exit 1; }
done
# The token counter needs a model's vocabulary to run; here it is only read.
python3 -I -c 'import ast, sys; ast.parse(open(sys.argv[1]).read())' ../quantum-conformance/eval/token_cost.py
cargo clippy --locked --release --all-targets -- -D warnings
cargo clippy --locked --release --all-targets --no-default-features -- -D warnings
- name: The smoke checks
shell: bash
working-directory: wai/wai-quantum-cli
run: |
set -euo pipefail
cargo build --locked --release
bash smoke.sh ./target/release/wai-quantum
# The release binary, pinned by version and SHA-256, rather than a build from source.
# Every WAI workflow that runs wasmtime pins this one version, and every run passes
# -W nan-canonicalization=y -W relaxed-simd-deterministic=y: every NaN a float
# instruction makes is the canonical one, and each relaxed SIMD instruction has one
# behaviour on every host. A stock wasmtime leaves both off. The fixed relaxed
# behaviour is the engine's own: on 50.0.1 relaxed_madd is fused (measured on arm64 and
# x86_64), where WebAssembly's deterministic profile prescribes the unfused result.
# The repository checks (scripts/check-wasmtime-pin.py) fail a workflow that pins another
# release or runs wasmtime without both settings.
- name: Install wasmtime 50.0.1
shell: bash
run: |
set -euo pipefail
[ "$(uname -m)" = x86_64 ] || { echo "::error::the pinned wasmtime archive is x86_64; this host is $(uname -m)"; exit 1; }
if ! command -v xz >/dev/null 2>&1; then
SUDO=""; [ "$(id -u)" -ne 0 ] && SUDO="sudo"
$SUDO apt-get update -qq
$SUDO apt-get install -y -qq xz-utils
fi
v=50.0.1
f="wasmtime-v$v-x86_64-linux.tar.xz"
curl --proto '=https' --tlsv1.2 -sSfL -o "$RUNNER_TEMP/$f" \
"https://github.com/bytecodealliance/wasmtime/releases/download/v$v/$f"
echo "685466455dd42a14774a86c36c8f2e45cbb9bd6d92a850139315df60a75498a5 $RUNNER_TEMP/$f" | sha256sum -c -
tar -xJf "$RUNNER_TEMP/$f" -C "$RUNNER_TEMP"
echo "$RUNNER_TEMP/wasmtime-v$v-x86_64-linux" >> "$GITHUB_PATH"
# The same source as a WASI component: a circuit's state hash, a run's
# result and a receipt signed with the published key are the native
# build's, byte for byte.
- name: The WASI component gives the native build's bytes
shell: bash
working-directory: wai/wai-quantum-cli
run: |
set -euo pipefail
cargo build --locked --release --target wasm32-wasip2 --no-default-features
w=target/wasm32-wasip2/release/wai-quantum.wasm
n=target/release/wai-quantum
printf 'OPENQASM 3.0; qubit[3] q; h q[0]; cx q[0], q[1]; t q[2]; cx q[1], q[2];' > "$RUNNER_TEMP/p.qasm"
for args in "lab run ghz -n 5" "lab receipt ghz -n 3 --demo-key"; do
$n $args > "$RUNNER_TEMP/native.txt" 2> /dev/null
wasmtime run -W nan-canonicalization=y -W relaxed-simd-deterministic=y "$w" $args > "$RUNNER_TEMP/wasi.txt" 2> /dev/null
cmp "$RUNNER_TEMP/native.txt" "$RUNNER_TEMP/wasi.txt" || { echo "::error::$args differs under wasmtime"; diff "$RUNNER_TEMP/native.txt" "$RUNNER_TEMP/wasi.txt" || true; exit 1; }
done
$n run --demo-key --json --shots 100 < "$RUNNER_TEMP/p.qasm" 2> /dev/null | python3 -c 'import json,sys; j=json.load(sys.stdin); j["result"].pop("elapsed_ms"); j["result"]["receipt"].pop("digest"); j["result"]["receipt"].pop("signature"); print(json.dumps(j, sort_keys=True))' > "$RUNNER_TEMP/native.json"
wasmtime run -W nan-canonicalization=y -W relaxed-simd-deterministic=y "$w" run --demo-key --json --shots 100 < "$RUNNER_TEMP/p.qasm" 2> /dev/null | python3 -c 'import json,sys; j=json.load(sys.stdin); j["result"].pop("elapsed_ms"); j["result"]["receipt"].pop("digest"); j["result"]["receipt"].pop("signature"); print(json.dumps(j, sort_keys=True))' > "$RUNNER_TEMP/wasi.json"
cmp "$RUNNER_TEMP/native.json" "$RUNNER_TEMP/wasi.json" || { echo "::error::a run's result differs under wasmtime"; exit 1; }
# A compile to Clifford+T (rotation synthesis in exact and fixed-point
# arithmetic, the passes, the certificate and its seal): byte for byte.
printf 'OPENQASM 3.0; qubit[2] q; h q[0]; rz(0.3) q[0]; cx q[0], q[1]; ry(1.1) q[1]; cp(0.7) q[0], q[1];' > "$RUNNER_TEMP/angles.qasm"
$n compile --demo-key --json --level 2 --epsilon 1e-8 < "$RUNNER_TEMP/angles.qasm" > "$RUNNER_TEMP/native-compile.json" 2> /dev/null
wasmtime run -W nan-canonicalization=y -W relaxed-simd-deterministic=y "$w" compile --demo-key --json --level 2 --epsilon 1e-8 < "$RUNNER_TEMP/angles.qasm" > "$RUNNER_TEMP/wasi-compile.json" 2> /dev/null
cmp "$RUNNER_TEMP/native-compile.json" "$RUNNER_TEMP/wasi-compile.json" || { echo "::error::a compile differs under wasmtime"; exit 1; }
# A run on a device target: the layout search, the noise derived and
# the density matrix, the same bits.
for t in synthetic-line-5 synthetic-all-5; do
$n run --demo-key --json --shots 100 --target $t < "$RUNNER_TEMP/p.qasm" 2> /dev/null | python3 -c 'import json,sys; j=json.load(sys.stdin); j["result"]["receipt"].pop("digest"); j["result"]["receipt"].pop("signature"); print(json.dumps(j, sort_keys=True))' > "$RUNNER_TEMP/native-$t.json"
wasmtime run -W nan-canonicalization=y -W relaxed-simd-deterministic=y "$w" run --demo-key --json --shots 100 --target $t < "$RUNNER_TEMP/p.qasm" 2> /dev/null | python3 -c 'import json,sys; j=json.load(sys.stdin); j["result"]["receipt"].pop("digest"); j["result"]["receipt"].pop("signature"); print(json.dumps(j, sort_keys=True))' > "$RUNNER_TEMP/wasi-$t.json"
cmp "$RUNNER_TEMP/native-$t.json" "$RUNNER_TEMP/wasi-$t.json" || { echo "::error::a run on $t differs under wasmtime"; exit 1; }
done
# A resource estimate (the counts, each rotation's synthesis, the
# planar model's factory search and the frontier, every receipt and
# the seal): byte for byte.
printf 'OPENQASM 3.0; qubit[4] q; h q; ctrl(3) @ x q[0], q[1], q[2], q[3]; rz(0.3) q[0]; ry(1.1) q[2]; t q[1];' > "$RUNNER_TEMP/est.qasm"
$n estimate --demo-key --json --qubit us-e3 < "$RUNNER_TEMP/est.qasm" > "$RUNNER_TEMP/native-estimate.json" 2> /dev/null
wasmtime run -W nan-canonicalization=y -W relaxed-simd-deterministic=y "$w" estimate --demo-key --json --qubit us-e3 < "$RUNNER_TEMP/est.qasm" > "$RUNNER_TEMP/wasi-estimate.json" 2> /dev/null
cmp "$RUNNER_TEMP/native-estimate.json" "$RUNNER_TEMP/wasi-estimate.json" || { echo "::error::an estimate differs under wasmtime"; exit 1; }
echo "the WASI component's output equals the native build's"
cli-macos:
name: command line · 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 and clippy, with the GPU state vector
shell: bash
working-directory: wai/wai-quantum-cli
run: |
set -euo pipefail
nice -n 10 cargo test --locked --release --features metal 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; }
nice -n 10 cargo clippy --locked --release --all-targets --features metal -- -D warnings
- name: The smoke checks, and a run on the GPU
shell: bash
working-directory: wai/wai-quantum-cli
run: |
set -euo pipefail
nice -n 10 cargo build --locked --release --features metal
bash smoke.sh ./target/release/wai-quantum
printf 'OPENQASM 3.0; qubit[12] q; bit[12] c; h q; rx(0.3) q[0]; cx q[0], q[1]; c = measure q;' > "$RUNNER_TEMP/gpu.qasm"
./target/release/wai-quantum run "$RUNNER_TEMP/gpu.qasm" --method statevector_single --shots 200 --demo-key --json > "$RUNNER_TEMP/gpu.json" 2> /dev/null
python3 -c 'import json,sys; j=json.load(open(sys.argv[1])); r=j["result"]; assert r["accuracy"]["class"]=="single" and sum(r["counts"].values())==200, r' "$RUNNER_TEMP/gpu.json"
echo "statevector_single ran on $(python3 -c 'import json,sys; print(json.load(open(sys.argv[1]))["result"]["accuracy"]["device"])' "$RUNNER_TEMP/gpu.json")"