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]>
152 lines
7.6 KiB
YAML
152 lines
7.6 KiB
YAML
name: wai-quantum tensor-engine member
|
|
|
|
# The fabric's tensor-engine member: the gate's products as exact bf16 limb
|
|
# convolutions on a tensor processor's matrix engine, accumulated in fp32. The
|
|
# kernel runs under the engine's kernel toolchain (Apache-2.0, pinned below),
|
|
# whose simulator is the target here: every certificate says it was earned in
|
|
# a simulator.
|
|
#
|
|
# - tensor-engine-sim (the container runner): the device tests must find the
|
|
# declared target, agree with the CPU on the arithmetic probe and on random
|
|
# states (out-of-range runs included), reproduce all four pinned
|
|
# certificates, and catch every mutant, the limb size the engine's inputs
|
|
# cannot hold among them.
|
|
# - tensor-engine-absent: with no kernel toolchain named, the same build finds
|
|
# no member and says why; the device tests run over no member with the
|
|
# declaration `none`.
|
|
|
|
concurrency:
|
|
group: ${{ github.workflow }}-${{ github.ref }}
|
|
cancel-in-progress: true
|
|
|
|
on:
|
|
pull_request:
|
|
paths:
|
|
- 'wai/wai-quantum/**'
|
|
- 'rust-toolchain.toml'
|
|
- '.github/workflows/wai-quantum-tensor-engine.yml'
|
|
schedule:
|
|
- cron: '27 4 * * *'
|
|
|
|
env:
|
|
KERNEL_TOOLCHAIN: nki==0.6.0+31049202112.g85070674
|
|
|
|
jobs:
|
|
# Does this pull request's own diff touch the paths below? This forge matches a
|
|
# workflow's `paths:` against the push, so a merge of main into a branch fires the
|
|
# workflow whatever the pull request changes; every job here waits on this answer.
|
|
# A push to main, a schedule, or anything the script cannot determine answers
|
|
# `true` (scripts/pr-touches-paths.py).
|
|
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-tensor-engine.yml \
|
|
"${{ github.event.pull_request.base.sha }}" \
|
|
"${{ github.event.pull_request.head.sha }}")
|
|
echo "this change touches wai-quantum-tensor-engine: $r"
|
|
echo "run=$r" >> "$GITHUB_OUTPUT"
|
|
|
|
tensor-engine-sim:
|
|
name: member · tensor engine, the kernel toolchain's simulator (linux)
|
|
needs: relevant
|
|
if: needs.relevant.outputs.run == 'true'
|
|
runs-on: linux
|
|
timeout-minutes: 60
|
|
container:
|
|
image: debian@sha256:913f6706df59a68922d1dd08f78c2476560a8d367897200a6005b00e5f67c2d5
|
|
options: --memory=6g --memory-swap=6g
|
|
env:
|
|
WAI_QUANTUM_TENSOR_ENGINE_EXPECT: tensor-engine-simulator
|
|
steps:
|
|
- name: Tools
|
|
shell: bash
|
|
run: |
|
|
set -euo pipefail
|
|
apt-get update -qq
|
|
apt-get install -y -qq --no-install-recommends ca-certificates curl git build-essential python3 python3-venv >/dev/null
|
|
# No Node for a checkout action: the forge's archive of the exact commit,
|
|
# its embedded commit id checked.
|
|
- name: Fetch the source
|
|
shell: bash
|
|
run: |
|
|
set -euo pipefail
|
|
want="${{ github.event.pull_request.head.sha || github.sha }}"
|
|
rm -rf src src.tar.gz src.tar && mkdir src
|
|
curl -sSf -o src.tar.gz "http://100.64.224.0:3400/Transaction-Science/open-standards/archive/$want.tar.gz"
|
|
gzip -dc src.tar.gz > src.tar
|
|
got="$(git get-tar-commit-id < src.tar)"
|
|
[ "$got" = "$want" ] || { echo "::error::the archive is of $got, not $want"; exit 1; }
|
|
tar -xf src.tar -C src --strip-components=1
|
|
rm src.tar.gz src.tar
|
|
- name: The member's tests on the simulator
|
|
shell: bash
|
|
working-directory: src/wai/wai-quantum
|
|
run: |
|
|
set -euo pipefail
|
|
cache=/ci-cache/wai-quantum-tensor-engine
|
|
mkdir -p "$cache"
|
|
exec 9>"$cache/.lock"
|
|
flock 9
|
|
venv="$cache/kernel-toolchain"
|
|
if ! "$venv/bin/python" -I -c 'import nki, ml_dtypes' 2>/dev/null || [ "$(cat "$venv/pinned" 2>/dev/null)" != "$KERNEL_TOOLCHAIN" ]; then
|
|
rm -rf "$venv"
|
|
python3 -m venv "$venv"
|
|
"$venv/bin/pip" install -q --extra-index-url https://pip.repos.neuron.amazonaws.com "$KERNEL_TOOLCHAIN" numpy ml_dtypes # allow-name: the kernel toolchain's package index
|
|
echo "$KERNEL_TOOLCHAIN" > "$venv/pinned"
|
|
fi
|
|
export RUSTUP_HOME="$cache/rustup" CARGO_HOME="$cache/cargo" CARGO_TARGET_DIR="$cache/target"
|
|
export PATH="$CARGO_HOME/bin:$PATH" WAI_QUANTUM_TENSOR_ENGINE_PYTHON="$venv/bin/python" WAI_QUANTUM_TENSOR_ENGINE_CACHE="$RUNNER_TEMP/tensor-engine"
|
|
if ! command -v rustup >/dev/null; then
|
|
curl --proto '=https' --tlsv1.2 -sSf https://sh.rustup.rs | sh -s -- -y --no-modify-path --profile minimal --default-toolchain none
|
|
fi
|
|
rustup toolchain install
|
|
cargo test --locked --features quantum_fabric_tensor_engine --lib -- quantum_fabric_tensor_engine:: quantum_fabric::tensor_engine:: 2>&1 | tee "$RUNNER_TEMP/unit.log"
|
|
grep -qE '^test result: ok\. [1-9][0-9]* passed; 0 failed' "$RUNNER_TEMP/unit.log" || { echo "::error::no unit test ran"; exit 1; }
|
|
key="$WAI_QUANTUM_TENSOR_ENGINE_EXPECT"
|
|
report="$RUNNER_TEMP/tensor-engine-report.txt"; : > "$report"
|
|
WAI_QUANTUM_TENSOR_ENGINE_REPORT="$report" cargo test --locked --release --features quantum_fabric_tensor_engine --test tensor_engine_member -- --test-threads=1 2>&1 | tee "$RUNNER_TEMP/device.log"
|
|
grep -qE '^test result: ok\. 9 passed; 0 failed' "$RUNNER_TEMP/device.log" || { echo "::error::not every device test ran and passed"; exit 1; }
|
|
cat "$report"
|
|
grep -qx "tensor-engine devices: as declared ($key)" "$report" || { echo "::error::the target found is not the declared one"; exit 1; }
|
|
grep -q "^tensor-engine member: key=$key .*; simulator:" "$report" || { echo "::error::the member does not say it ran in a simulator"; exit 1; }
|
|
n="$(grep -c "^tensor-engine certified: key=$key .* parity=bit-exact " "$report")"
|
|
[ "$n" = 4 ] || { echo "::error::$n of 4 certificates were earned"; exit 1; }
|
|
grep -qx "tensor-engine mutants: key=$key killed 4 of 4" "$report" || { echo "::error::a mutant survived"; exit 1; }
|
|
|
|
tensor-engine-absent:
|
|
name: member · tensor engine absent (linux-arm64)
|
|
needs: relevant
|
|
if: needs.relevant.outputs.run == 'true'
|
|
runs-on: linux-arm64
|
|
timeout-minutes: 60
|
|
env:
|
|
WAI_QUANTUM_TENSOR_ENGINE_EXPECT: none
|
|
steps:
|
|
- uses: actions/checkout@v5
|
|
- name: Install Rust 1.98.0
|
|
shell: bash
|
|
run: |
|
|
export PATH="$HOME/.cargo/bin:$PATH"
|
|
command -v rustup >/dev/null 2>&1 || \
|
|
curl --proto '=https' --tlsv1.2 -sSf https://sh.rustup.rs | sh -s -- -y --profile minimal --default-toolchain 1.98.0
|
|
rustup toolchain install 1.98.0 --profile minimal
|
|
rustup default 1.98.0
|
|
echo "$HOME/.cargo/bin" >> "$GITHUB_PATH"
|
|
- name: No kernel toolchain here, and the member says so
|
|
shell: bash
|
|
working-directory: wai/wai-quantum
|
|
run: |
|
|
set -euo pipefail
|
|
report="$RUNNER_TEMP/tensor-engine-report.txt"; : > "$report"
|
|
WAI_QUANTUM_TENSOR_ENGINE_REPORT="$report" cargo test --locked --features quantum_fabric_tensor_engine --test tensor_engine_member -- --test-threads=1 2>&1 | tee "$RUNNER_TEMP/device.log"
|
|
grep -qE '^test result: ok\. 9 passed; 0 failed' "$RUNNER_TEMP/device.log" || { echo "::error::not every device test ran and passed"; exit 1; }
|
|
grep -qx 'tensor-engine devices: as declared (none)' "$report" || { echo "::error::the absent case was not established"; exit 1; }
|