Compare commits

..

57 Commits

Author SHA1 Message Date
Nicolas Sarlin
871cc914ee chore(zk): add noise tests for zkv1 2024-12-18 16:39:45 +01:00
Nicolas Sarlin
adc714887c fix: new param naming 2024-12-18 16:39:45 +01:00
Arthur Meyre
4c870fd200 fix(shortint): fix sample count in pbs128 pfail 2024-12-18 16:39:45 +01:00
Arthur Meyre
ce703a6991 test(shortint): add variance check after KS in classic PBS AP 2024-12-18 16:39:45 +01:00
Arthur Meyre
d4d43d434d test(shortint): add normality check after KS in classic AP 2024-12-18 16:39:45 +01:00
Arthur Meyre
1c69df2196 fix(test): fix slighlty wrong log message 2024-12-18 16:39:45 +01:00
Arthur Meyre
67ea7d05ef fix(test): fix test parameters for PBS 128
Disable the bound check as our computation disagrees with the RO
2024-12-18 16:39:45 +01:00
Arthur Meyre
615c26039a test(shortint): add pfail estimation after ms in compression 2024-12-18 16:39:45 +01:00
Arthur Meyre
4c79ca99bc wip: re-exported pbs 128 with symbolic mantissa
- add pbs 128 params with a mantissa setting
- noise with this settings is in line with RO prediction
2024-12-18 16:39:45 +01:00
Arthur Meyre
252182fc38 wip: add pbs 128 tests (noise and pfail) 2024-12-18 16:39:45 +01:00
Arthur Meyre
c079228379 feat(core): add pbs 128 noise formulas 2024-12-18 16:39:45 +01:00
Arthur Meyre
2c246758de test(shortint): add pfail measurement for full compression + AP 2024-12-18 16:39:44 +01:00
Arthur Meyre
93b2d41450 test(shortint): slightly change decryption logic to also get the padding bit out 2024-12-18 16:39:44 +01:00
Arthur Meyre
98acaeb925 test(shortint): add pfail check for first part of compression 2024-12-18 16:39:44 +01:00
Arthur Meyre
4ff2425557 test(shortint): add first part of the compression AP 2024-12-18 16:39:44 +01:00
Arthur Meyre
0fc4f40a0c fix(shortint): accept trvial ciphertexts for compression 2024-12-18 16:39:44 +01:00
Arthur Meyre
db32c8d360 chore: rename helper function to match test usage 2024-12-18 16:39:44 +01:00
Arthur Meyre
37a49b6287 feat(core): add noise formula for packing keyswitch 2024-12-18 16:39:44 +01:00
Arthur Meyre
d8ea767c7a test(shortint): add pfail test for PKE -> KS -> (DP -> KS Compute) -> MS 2024-12-18 16:39:44 +01:00
Arthur Meyre
92645ea83e refactor(test): also accept PKE -> KS to big -> DP -> KS -> MS 2024-12-18 16:39:44 +01:00
Arthur Meyre
6a062a0309 test(shortint): add noise check for PKE -> KS to small key + MS 2024-12-18 16:39:44 +01:00
Arthur Meyre
d7778e8060 wip: hotfix for borrow mut error, this needs a design pass/reflection 2024-12-18 16:39:44 +01:00
Arthur Meyre
e389dbf0a2 feat: add noise formulas for TUniform 132 bits of security 2024-12-18 16:39:44 +01:00
Arthur Meyre
fdd3f342e6 test(hl): add noise check for CompactPublicKey encryption w/ TUniform param 2024-12-18 16:39:44 +01:00
Arthur Meyre
1d6eef6412 test(shortint): add pfail test for the classic AP 2024-12-18 16:39:44 +01:00
Arthur Meyre
9634ed6fe2 refactor: prepare code factorization for noise and pfail 2024-12-18 16:39:44 +01:00
Arthur Meyre
1cbd1a148e test: add shortint atomic pattern noise measurement 2024-12-18 16:39:44 +01:00
Arthur Meyre
bf8012b0dc test(hl): test encryption noise of FheUint ciphertext in HL API 2024-12-18 16:39:44 +01:00
Arthur Meyre
274583145c refactor: remove dev dependency on libm use statrs instead 2024-12-18 16:39:44 +01:00
Nicolas Sarlin
4085a2336e chore(zk): add a test with zkv1 in integer compact list encryption 2024-12-18 16:39:44 +01:00
Nicolas Sarlin
7b5782d491 chore(shortint): new parameters for zkv1 2024-12-18 16:39:44 +01:00
Nicolas Sarlin
32b8d7d420 chore(csprng): prepare release 0.5.0 2024-12-18 16:39:44 +01:00
Nicolas Sarlin
ba20d4b349 chore(doc): fix c_api setup command using removed make target 2024-12-18 16:39:44 +01:00
Nicolas Sarlin
65e54e61b4 chore(csprng)!: moved the RandomGenerator detector into tfhe-csprng 2024-12-18 16:39:44 +01:00
Nicolas Sarlin
f701123c62 chore(tfhe)!: remove arch specific features
BREAKING_CHANGE:
-	The x86_64, x86_64-unix, aarch64, aarch64-unix have been removed, the target
architecture and os family are now automatically detected. A `software_prng`
feature has been added to force the use of a software implementation on older
CPUs
2024-12-18 16:39:44 +01:00
Nicolas Sarlin
3b9a8e9106 chore(csprng)!: remove generator_aarch64_aes feature
BREAKING_CHANGE:
- The `generator_aarch64_aes` feature is no longer supported for tfhe-csprng
2024-12-18 16:39:44 +01:00
Nicolas Sarlin
58192a10ea chore(csprng)!: remove generator_x86_64_aesni feature
BREAKING_CHANGE:
- The `generator_x86_64_aesni` feature is no longer supported for tfhe-csprng
2024-12-18 16:39:44 +01:00
Nicolas Sarlin
da8950ac3e chore(csprng)!: remove seeder_x86_64_rdseed feature
BREAKING_CHANGE:
- The `seeder_x86_64_rdseed` feature is no longer supported for tfhe-csprng
2024-12-18 16:39:44 +01:00
Nicolas Sarlin
9748416db5 chore(csprng)!: remove seeder_unix feature
BREAKING_CHANGE:
- The `seeder_unix` feature is no longer supported for tfhe-csprng
2024-12-18 16:39:44 +01:00
Agnes Leroy
76c6daa814 fix(gpu): fix memory leak 2024-12-18 16:39:43 +01:00
Agnes Leroy
4b19143fac fix(gpu): fix ct degree and noise level after some ops 2024-12-18 16:39:43 +01:00
Agnes Leroy
4f0a3f9feb fix(gpu): fix single gpu on device other than 0 2024-12-18 16:39:43 +01:00
Nicolas Sarlin
dd5f444b22 doc(shortint): add some clarification about shortint size 2024-12-18 16:39:43 +01:00
Nicolas Sarlin
b0e7be546d chore(zk): check that k <= d for zk crs 2024-12-18 16:39:43 +01:00
Nicolas Sarlin
8170037f6e chore(backward): move allow(dead_code) to dispatch variants
This allows to detect unused dispatch enums
2024-12-18 16:39:43 +01:00
Nicolas Sarlin
a9e23e54c0 chore(zk)!: update parameters for zk v2 2024-12-18 16:39:43 +01:00
Nicolas Sarlin
f39d3c2473 feat(zk)!: plug zk v2
BREAKING CHANGE:
- The object ZkVerificationOutCome has been renamed ZkVerificationOutcome.
- Conformance of proofs now checks the scheme version of the CRS. This is
breaking at the shortint and core_crypto levels, and for manually built integer
conformance params.

New CRS will be generated with the V2 Scheme by default, but V1 CRS and proofs
are still accepted, so this is not breaking. New methods have been added to
generate a V1 CRS.
2024-12-18 16:39:43 +01:00
Nicolas Sarlin
fced7f54d9 fix(zk-pok): missing Versionize for ComputeLoadProofFields 2024-12-18 16:39:43 +01:00
Nicolas Sarlin
f9c896917d feat(versionable): "Version" macro now handles transparent attribute 2024-12-18 16:39:43 +01:00
Agnes Leroy
6130623c37 chore(gpu): rework select to avoid using local streams 2024-12-18 16:39:43 +01:00
Mayeul@Zama
f4c76793e1 chore: fix clippy lints after toolchain update 2024-12-18 16:39:43 +01:00
Agnes Leroy
79cd4fe04c chore(gpu): make artifact name unique across different machines 2024-12-18 16:39:43 +01:00
Nicolas Sarlin
cc0727bf39 chore(all): remove the dependency to lazy_static 2024-12-18 16:39:43 +01:00
Arthur Meyre
90d846aca1 chore: remove aliases for gaussian parameters for compact PK
- we are TUniform by default so no more aliases for gaussian parameters
2024-12-18 16:39:43 +01:00
Agnes Leroy
1a3c455832 chore(gpu): run pbs in parallel in difference_check 2024-12-18 16:39:43 +01:00
Mayeul@Zama
75ec0e9581 chore(hlapi): stabilize FheTypes 2024-12-18 16:39:43 +01:00
Agnes Leroy
94f6c6f0ef chore(gpu): add sxm5 vm target 2024-12-18 16:39:43 +01:00
2722 changed files with 78677 additions and 789023 deletions

View File

@@ -1,14 +0,0 @@
[advisories]
ignore = [
# Ignoring unmaintained 'paste' advisory as it is a widely used, low-risk build dependency.
"RUSTSEC-2024-0436",
# Ignoring unmaintained 'bincode' crate. Getting rid of it would be too complex on the short term.
"RUSTSEC-2025-0141",
]
[output]
# Deny advisories that are warnings by default.
# At the moment this works if we allow paste, we might want to disable this in the future if it
# becomes too tedious
deny = ["warnings"]
quiet = false

View File

@@ -8,14 +8,8 @@ root = true
end_of_line = lf
insert_final_newline = true
# 4 space indentation for rust and toml
[*.{rs,toml}]
# 4 space indentation
[*.rs]
charset = utf-8
indent_style = space
indent_size = 4
# 2 for c and js
[*.{js,json,c,h}]
charset = utf-8
indent_style = space
indent_size = 2

3
.gitattributes vendored
View File

@@ -1,3 +0,0 @@
*.hpu filter=lfs diff=lfs merge=lfs -text
*.bcode filter=lfs diff=lfs merge=lfs -text
*.cbor filter=lfs diff=lfs merge=lfs -text

View File

@@ -5,10 +5,6 @@ self-hosted-runner:
- 4090-desktop
- large_windows_16_latest
- large_ubuntu_16
- large_ubuntu_16-22.04
- v80-desktop
- v80-marais
- v80-couperin
# Configuration variables in array of strings defined in your repository or
# organization. `null` means disabling configuration variables check.
# Empty array means no configuration variable is allowed.

View File

@@ -1,121 +0,0 @@
name: Setup Cuda
description: Setup Cuda on Hyperstack or GitHub instance
inputs:
cuda-version:
description: Version of Cuda to use
required: true
gcc-version:
description: Version of GCC to use
required: true
github-instance:
description: Instance is hosted on GitHub
default: 'false'
runs:
using: "composite"
steps:
# Mandatory on hyperstack since a bootable volume is not re-usable yet.
- name: Install dependencies
shell: bash
run: |
wget https://github.com/Kitware/CMake/releases/download/v"${CMAKE_VERSION}"/cmake-"${CMAKE_VERSION}"-linux-x86_64.sh
echo "${CMAKE_SCRIPT_SHA} cmake-${CMAKE_VERSION}-linux-x86_64.sh" > checksum
sha256sum -c checksum
sudo bash cmake-"${CMAKE_VERSION}"-linux-x86_64.sh --skip-license --prefix=/usr/ --exclude-subdir
# Disable unattended-upgrades to avoid lock issues
sudo systemctl mask --now unattended-upgrades
sudo systemctl stop --now unattended-upgrades
sudo apt-get clean
sudo rm -rf /var/lib/apt/lists/*
sudo apt purge -y unattended-upgrades
sudo apt update
sudo apt install -y cmake-format libclang-dev
env:
CMAKE_VERSION: 3.29.6
CMAKE_SCRIPT_SHA: "6e4fada5cba3472ae503a11232b6580786802f0879cead2741672bf65d97488a"
- name: Install GCC
if: inputs.github-instance == 'true'
shell: bash
env:
GCC_VERSION: ${{ inputs.gcc-version }}
run: |
sudo apt-get install gcc-"{GCC_VERSION}" g++-"{GCC_VERSION}"
sudo update-alternatives --install /usr/bin/gcc gcc /usr/bin/gcc-"{GCC_VERSION}" 20
sudo update-alternatives --install /usr/bin/g++ g++ /usr/bin/g++-"{GCC_VERSION}" 20
- name: Check GCC
shell: bash
env:
GCC_VERSION: ${{ inputs.gcc-version }}
run: |
which gcc-"${GCC_VERSION}"
- name: Install CUDA
if: inputs.github-instance == 'true'
shell: bash
env:
CUDA_VERSION: ${{ inputs.cuda-version }}
CUDA_KEYRING_PACKAGE: cuda-keyring_1.1-1_all.deb
CUDA_KEYRING_SHA: "d93190d50b98ad4699ff40f4f7af50f16a76dac3bb8da1eaaf366d47898ff8df"
run: |
# Use Sed to extract a value from a string, this cannot be done with the ${variable//search/replace} pattern.
# shellcheck disable=SC2001
TOOLKIT_VERSION="$(echo "${CUDA_VERSION}" | sed 's/\(.*\)\.\(.*\)/\1-\2/')"
wget https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2204/x86_64/${CUDA_KEYRING_PACKAGE}
echo "${CUDA_KEYRING_SHA} ${CUDA_KEYRING_PACKAGE}" > checksum
sha256sum -c checksum
sudo dpkg -i "${CUDA_KEYRING_PACKAGE}"
# Disable unattended-upgrades to avoid lock issues
sudo systemctl mask --now unattended-upgrades
sudo systemctl stop --now unattended-upgrades
sudo apt-get clean
sudo rm -rf /var/lib/apt/lists/*
sudo apt purge -y unattended-upgrades
sudo apt update
sudo apt -y install cuda-toolkit-"${TOOLKIT_VERSION}"
- name: Export CUDA variables
shell: bash
run: |
find /usr/local -executable -name "nvcc"
CUDA_PATH=/usr/local/cuda-"${CUDA_VERSION}"
{
echo "CUDA_PATH=$CUDA_PATH";
echo "LD_LIBRARY_PATH=$CUDA_PATH/lib64:$LD_LIBRARY_PATH";
echo "CUDA_MODULE_LOADER=EAGER";
echo "PATH=$PATH:$CUDA_PATH/bin";
} >> "${GITHUB_ENV}"
{
echo "PATH=$PATH:$CUDA_PATH/bin";
} >> "${GITHUB_PATH}"
env:
CUDA_VERSION: ${{ inputs.cuda-version }}
# Specify the correct host compilers
- name: Export gcc and g++ variables
shell: bash
run: |
{
echo "CC=/usr/bin/gcc-${GCC_VERSION}";
echo "CXX=/usr/bin/g++-${GCC_VERSION}";
echo "CUDAHOSTCXX=/usr/bin/g++-${GCC_VERSION}";
} >> "${GITHUB_ENV}"
env:
GCC_VERSION: ${{ inputs.gcc-version }}
- name: Check setup
shell: bash
run: |
which nvcc
- name: Check device is detected
shell: bash
run: nvidia-smi

View File

@@ -0,0 +1,53 @@
name: Setup Cuda
description: Setup Cuda on Hyperstack instance
inputs:
cuda-version:
description: Version of Cuda to use
required: true
gcc-version:
description: Version of GCC to use
required: true
cmake-version:
description: Version of cmake to use
default: 3.29.6
runs:
using: "composite"
steps:
# Mandatory on hyperstack since a bootable volume is not re-usable yet.
- name: Install dependencies
shell: bash
run: |
sudo apt update
sudo apt install -y checkinstall zlib1g-dev libssl-dev libclang-dev
wget https://github.com/Kitware/CMake/releases/download/v${{ inputs.cmake-version }}/cmake-${{ inputs.cmake-version }}.tar.gz
tar -zxvf cmake-${{ inputs.cmake-version }}.tar.gz
cd cmake-${{ inputs.cmake-version }}
./bootstrap
make -j"$(nproc)"
sudo make install
- name: Export CUDA variables
shell: bash
run: |
CUDA_PATH=/usr/local/cuda-${{ inputs.cuda-version }}
echo "CUDA_PATH=$CUDA_PATH" >> "${GITHUB_ENV}"
echo "$CUDA_PATH/bin" >> "${GITHUB_PATH}"
echo "LD_LIBRARY_PATH=$CUDA_PATH/lib:$LD_LIBRARY_PATH" >> "${GITHUB_ENV}"
echo "CUDACXX=/usr/local/cuda-${{ inputs.cuda-version }}/bin/nvcc" >> "${GITHUB_ENV}"
# Specify the correct host compilers
- name: Export gcc and g++ variables
shell: bash
run: |
{
echo "CC=/usr/bin/gcc-${{ inputs.gcc-version }}";
echo "CXX=/usr/bin/g++-${{ inputs.gcc-version }}";
echo "CUDAHOSTCXX=/usr/bin/g++-${{ inputs.gcc-version }}";
echo "HOME=/home/ubuntu";
} >> "${GITHUB_ENV}"
- name: Check device is detected
shell: bash
run: nvidia-smi

View File

@@ -7,5 +7,3 @@ updates:
# Check for updates to GitHub Actions every sunday
interval: "weekly"
day: "sunday"
cooldown:
default-days: 7

15
.github/runs-on.yml vendored
View File

@@ -1,15 +0,0 @@
runners:
cpu-big:
family: m6i.32xlarge
image: cpu-tests-eu-west-3
volume: 200gb
spot: false
cpu-small:
family: m6i.4xlarge
image: cpu-tests-eu-west-3
volume: 200gb
spot: false
images:
cpu-tests-eu-west-3:
ami: "ami-0a786ffdb1411fac4" # Ubuntu 24.04

View File

@@ -1,22 +1,16 @@
# Add labels in pull request
name: approve_label
name: PR label manager
on:
pull_request:
pull_request_review:
types: [submitted]
permissions: {}
# zizmor: ignore[concurrency-limits] this workflow needs to react to any event in a pull-request
jobs:
trigger-tests:
name: approve_label/trigger-tests
runs-on: ubuntu-latest
permissions:
pull-requests: write # Needed to apply or remove label
pull-requests: write
steps:
- name: Get current labels
uses: snnaplab/get-labels-action@f426df40304808ace3b5282d4f036515f7609576
@@ -40,10 +34,3 @@ jobs:
# We need to use a PAT to be able to trigger `labeled` event for the other workflow.
github_token: ${{ secrets.FHE_ACTIONS_TOKEN }}
labels: approved
- name: Check if maintainer needs to handle label manually
if: ${{ failure() }}
run: |
echo "Pull-request from an external contributor."
echo "A maintainer need to manually add/remove the 'approved' label."
exit 1

View File

@@ -1,114 +0,0 @@
# Run data related tests
name: aws_data_tests
env:
CARGO_TERM_COLOR: always
ACTION_RUN_URL: ${{ github.server_url }}/${{ github.repository }}/actions/runs/${{ github.run_id }}
RUSTFLAGS: "-C target-cpu=native"
RUST_BACKTRACE: "full"
RUST_MIN_STACK: "8388608"
SLACK_CHANNEL: ${{ secrets.SLACK_CHANNEL }}
SLACK_ICON: https://pbs.twimg.com/profile_images/1274014582265298945/OjBKP9kn_400x400.png
SLACK_USERNAME: ${{ secrets.BOT_USERNAME }}
SLACK_WEBHOOK: ${{ secrets.SLACK_WEBHOOK }}
SLACKIFY_MARKDOWN: true
PULL_REQUEST_MD_LINK: ""
CHECKOUT_TOKEN: ${{ secrets.REPO_CHECKOUT_TOKEN || secrets.GITHUB_TOKEN }}
on:
# Allows you to run this workflow manually from the Actions tab as an alternative.
workflow_dispatch:
pull_request:
push:
branches:
- main
permissions:
contents: read
# zizmor: ignore[concurrency-limits] concurrency is managed after instance setup to ensure safe provisioning
jobs:
data-tests:
name: aws_data_tests/data-tests (bpr)
if: (github.event_name == 'push' && github.repository == 'zama-ai/tfhe-rs') ||
github.event_name != 'push'
runs-on: "runs-on=${{ github.run_id }}/runner=cpu-small"
concurrency:
group: ${{ github.workflow_ref }}${{ github.ref == 'refs/heads/main' && github.sha || '' }}
cancel-in-progress: ${{ github.ref != 'refs/heads/main' }}
steps:
- name: Checkout tfhe-rs
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd
with:
persist-credentials: 'true' # Needed to pull lfs data
token: ${{ env.CHECKOUT_TOKEN }}
# Cache key is an aggregated hash of lfs files hashes
- name: Get LFS data sha
id: hash-lfs-data
run: |
SHA=$(git lfs ls-files -l -I utils/tfhe-backward-compat-data,tests/corrupted_inputs_deserialization | sha256sum | cut -d' ' -f1)
echo "sha=${SHA}" >> "${GITHUB_OUTPUT}"
- name: Retrieve data from cache
id: retrieve-data-cache
uses: actions/cache/restore@27d5ce7f107fe9357f9df03efb73ab90386fccae #v5.0.5
with:
path: |
utils/tfhe-backward-compat-data/**/*.cbor
utils/tfhe-backward-compat-data/**/*.bcode
tests/corrupted_inputs_deserialization/**/*.bcode
key: ${{ steps.hash-lfs-data.outputs.sha }}
- name: Pull test data
if: steps.retrieve-data-cache.outputs.cache-hit != 'true'
run: |
make pull_backward_compat_data
make pull_corrupted_inputs_data
# Pull token was stored by action/checkout to be used by lfs, we don't need it anymore
- name: Remove git credentials
run: | # Starting version 6.0, action/checkout uses a dedicated file for git credentials
git config --local --unset-all http.https://github.com/.extraheader || rm "${RUNNER_TEMP}"/git-credentials-*.config
- name: Install latest stable
uses: dtolnay/rust-toolchain@e97e2d8cc328f1b50210efc529dca0028893a2d9 # zizmor: ignore[stale-action-refs] this action doesn't create releases
with:
toolchain: stable
- name: Run backward compatibility tests
run: |
make test_backward_compatibility_ci
- name: Run corrupted inputs deserialization tests
run: |
make test_corrupted_inputs_ci
- name: Store data in cache
if: steps.retrieve-data-cache.outputs.cache-hit != 'true'
continue-on-error: true
uses: actions/cache/save@27d5ce7f107fe9357f9df03efb73ab90386fccae #v5.0.5
with:
path: |
utils/tfhe-backward-compat-data/**/*.cbor
utils/tfhe-backward-compat-data/**/*.bcode
tests/corrupted_inputs_deserialization/**/*.bcode
key: ${{ steps.hash-lfs-data.outputs.sha }}
- name: Set pull-request URL
if: ${{ failure() && github.event_name == 'pull_request' }}
run: |
echo "PULL_REQUEST_MD_LINK=[pull-request](${PR_BASE_URL}${PR_NUMBER}), " >> "${GITHUB_ENV}"
env:
PR_BASE_URL: ${{ vars.PR_BASE_URL }}
PR_NUMBER: ${{ github.event.pull_request.number }}
- name: Slack Notification
if: ${{ failure() || (cancelled() && github.event_name != 'pull_request') }}
continue-on-error: true
uses: rtCamp/action-slack-notify@e31e87e03dd19038e411e38ae27cbad084a90661
env:
SLACK_COLOR: ${{ job.status }}
SLACK_MESSAGE: "Backward compatibility tests finished with status: ${{ job.status }}. (${{ env.PULL_REQUEST_MD_LINK }}[action run](${{ env.ACTION_RUN_URL }}))"

View File

@@ -0,0 +1,117 @@
# Run backward compatibility tests
name: Backward compatibility Tests on CPU
env:
CARGO_TERM_COLOR: always
ACTION_RUN_URL: ${{ github.server_url }}/${{ github.repository }}/actions/runs/${{ github.run_id }}
RUSTFLAGS: "-C target-cpu=native"
RUST_BACKTRACE: "full"
RUST_MIN_STACK: "8388608"
SLACK_CHANNEL: ${{ secrets.SLACK_CHANNEL }}
SLACK_ICON: https://pbs.twimg.com/profile_images/1274014582265298945/OjBKP9kn_400x400.png
SLACK_USERNAME: ${{ secrets.BOT_USERNAME }}
SLACK_WEBHOOK: ${{ secrets.SLACK_WEBHOOK }}
on:
# Allows you to run this workflow manually from the Actions tab as an alternative.
workflow_dispatch:
pull_request:
jobs:
setup-instance:
name: Setup instance (backward-compat-tests)
runs-on: ubuntu-latest
outputs:
runner-name: ${{ steps.start-instance.outputs.label }}
steps:
- name: Start instance
id: start-instance
uses: zama-ai/slab-github-runner@98f0788261a7323d5d695a883e20df36591a92b7
with:
mode: start
github-token: ${{ secrets.SLAB_ACTION_TOKEN }}
slab-url: ${{ secrets.SLAB_BASE_URL }}
job-secret: ${{ secrets.JOB_SECRET }}
backend: aws
profile: cpu-small
backward-compat-tests:
name: Backward compatibility tests
needs: [ setup-instance ]
concurrency:
group: ${{ github.workflow }}_${{ github.ref }}
cancel-in-progress: true
runs-on: ${{ needs.setup-instance.outputs.runner-name }}
steps:
- name: Checkout tfhe-rs
uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683
with:
persist-credentials: 'false'
token: ${{ secrets.FHE_ACTIONS_TOKEN }}
- name: Install latest stable
uses: dtolnay/rust-toolchain@315e265cd78dad1e1dcf3a5074f6d6c47029d5aa
with:
toolchain: stable
- name: Install git-lfs
run: |
sudo apt update && sudo apt -y install git-lfs
- name: Use specific data branch
if: ${{ contains(github.event.pull_request.labels.*.name, 'data_PR') }}
env:
PR_BRANCH: ${{ github.head_ref || github.ref_name }}
run: |
echo "BACKWARD_COMPAT_DATA_BRANCH=${PR_BRANCH}" >> "${GITHUB_ENV}"
- name: Get backward compat branch
id: backward_compat_branch
run: |
BRANCH="$(make backward_compat_branch)"
echo "branch=${BRANCH}" >> "${GITHUB_OUTPUT}"
- name: Clone test data
uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683
with:
persist-credentials: 'false'
repository: zama-ai/tfhe-backward-compat-data
path: tfhe/tfhe-backward-compat-data
lfs: 'true'
ref: ${{ steps.backward_compat_branch.outputs.branch }}
- name: Run backward compatibility tests
run: |
make test_backward_compatibility_ci
- name: Slack Notification
if: ${{ failure() }}
continue-on-error: true
uses: rtCamp/action-slack-notify@c33737706dea87cd7784c687dadc9adf1be59990
env:
SLACK_COLOR: ${{ job.status }}
SLACK_MESSAGE: "Backward compatibility tests finished with status: ${{ job.status }}. (${{ env.ACTION_RUN_URL }})"
teardown-instance:
name: Teardown instance (backward-compat-tests)
if: ${{ always() && needs.setup-instance.result != 'skipped' }}
needs: [ setup-instance, backward-compat-tests ]
runs-on: ubuntu-latest
steps:
- name: Stop instance
id: stop-instance
uses: zama-ai/slab-github-runner@98f0788261a7323d5d695a883e20df36591a92b7
with:
mode: stop
github-token: ${{ secrets.SLAB_ACTION_TOKEN }}
slab-url: ${{ secrets.SLAB_BASE_URL }}
job-secret: ${{ secrets.JOB_SECRET }}
label: ${{ needs.setup-instance.outputs.runner-name }}
- name: Slack Notification
if: ${{ failure() }}
continue-on-error: true
uses: rtCamp/action-slack-notify@c33737706dea87cd7784c687dadc9adf1be59990
env:
SLACK_COLOR: ${{ job.status }}
SLACK_MESSAGE: "Instance teardown (backward-compat-tests) finished with status: ${{ job.status }}. (${{ env.ACTION_RUN_URL }})"

View File

@@ -1,5 +1,5 @@
# Run a small subset of tests to ensure quick feedback.
name: aws_tfhe_fast_tests
name: Fast AWS Tests on CPU
env:
CARGO_TERM_COLOR: always
@@ -11,32 +11,22 @@ env:
SLACK_ICON: https://pbs.twimg.com/profile_images/1274014582265298945/OjBKP9kn_400x400.png
SLACK_USERNAME: ${{ secrets.BOT_USERNAME }}
SLACK_WEBHOOK: ${{ secrets.SLACK_WEBHOOK }}
SLACKIFY_MARKDOWN: true
IS_PULL_REQUEST: ${{ github.event_name == 'pull_request' }}
PULL_REQUEST_MD_LINK: ""
CHECKOUT_TOKEN: ${{ secrets.REPO_CHECKOUT_TOKEN || secrets.GITHUB_TOKEN }}
on:
# Allows you to run this workflow manually from the Actions tab as an alternative.
workflow_dispatch:
pull_request:
permissions:
contents: read
# zizmor: ignore[concurrency-limits] concurrency is managed after instance setup to ensure safe provisioning
jobs:
should-run:
name: aws_tfhe_fast_tests/should-run
runs-on: ubuntu-latest
permissions:
pull-requests: read # Needed to check for file change
pull-requests: write
outputs:
csprng_test: ${{ env.IS_PULL_REQUEST == 'false' || steps.changed-files.outputs.csprng_any_changed }}
zk_pok_test: ${{ env.IS_PULL_REQUEST == 'false' || steps.changed-files.outputs.zk_pok_any_changed }}
versionable_test: ${{ env.IS_PULL_REQUEST == 'false' || steps.changed-files.outputs.versionable_any_changed }}
safe_serialize_test: ${{ env.IS_PULL_REQUEST == 'false' || steps.changed-files.outputs.safe_serialize_any_changed }}
core_crypto_test: ${{ env.IS_PULL_REQUEST == 'false' ||
steps.changed-files.outputs.core_crypto_any_changed ||
steps.changed-files.outputs.dependencies_any_changed }}
@@ -61,16 +51,16 @@ jobs:
any_file_changed: ${{ env.IS_PULL_REQUEST == 'false' || steps.aggregated-changes.outputs.any_changed }}
steps:
- name: Checkout tfhe-rs
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd
uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683
with:
fetch-depth: 0
persist-credentials: "false"
token: ${{ env.CHECKOUT_TOKEN }}
token: ${{ secrets.FHE_ACTIONS_TOKEN }}
- name: Check for file changes
id: changed-files
uses: tj-actions/changed-files@9426d40962ed5378910ee2e21d5f8c6fcbf2dd96 # v47.0.6
uses: tj-actions/changed-files@bab30c2299617f6615ec02a68b9a40d10bd21366
with:
since_last_remote_commit: true
files_yaml: |
dependencies:
- tfhe/Cargo.toml
@@ -79,7 +69,6 @@ jobs:
- tfhe-zk-pok/**
- utils/tfhe-versionable/**
- utils/tfhe-versionable-derive/**
- utils/tfhe-safe-serialize/**
csprng:
- tfhe-csprng/**
zk_pok:
@@ -87,8 +76,6 @@ jobs:
versionable:
- utils/tfhe-versionable/**
- utils/tfhe-versionable-derive/**
safe_serialize:
- utils/tfhe-safe-serialize/**
core_crypto:
- tfhe/src/core_crypto/**
boolean:
@@ -116,7 +103,7 @@ jobs:
user_docs:
- tfhe/src/**
- '!tfhe/src/c_api/**'
- 'tfhe/docs/**/**.md'
- 'tfhe/docs/**.md'
- README.md
- name: Aggregate file changes
@@ -125,7 +112,6 @@ jobs:
steps.changed-files.outputs.csprng_any_changed == 'true' ||
steps.changed-files.outputs.zk_pok_any_changed == 'true' ||
steps.changed-files.outputs.versionable_any_changed == 'true' ||
steps.changed-files.outputs.safe_serialize_any_changed == 'true' ||
steps.changed-files.outputs.core_crypto_any_changed == 'true' ||
steps.changed-files.outputs.boolean_any_changed == 'true' ||
steps.changed-files.outputs.shortint_any_changed == 'true' ||
@@ -136,24 +122,44 @@ jobs:
run: |
echo "any_changed=true" >> "$GITHUB_OUTPUT"
setup-instance:
name: Setup instance (fast-tests)
if: github.event_name != 'pull_request' ||
needs.should-run.outputs.any_file_changed == 'true'
needs: should-run
runs-on: ubuntu-latest
outputs:
runner-name: ${{ steps.start-instance.outputs.label }}
steps:
- name: Start instance
id: start-instance
uses: zama-ai/slab-github-runner@98f0788261a7323d5d695a883e20df36591a92b7
with:
mode: start
github-token: ${{ secrets.SLAB_ACTION_TOKEN }}
slab-url: ${{ secrets.SLAB_BASE_URL }}
job-secret: ${{ secrets.JOB_SECRET }}
backend: aws
profile: cpu-big
fast-tests:
name: Fast CPU tests
needs: should-run
if: github.event_name == 'workflow_dispatch' ||
(github.event_name != 'workflow_dispatch' && needs.should-run.outputs.any_file_changed == 'true')
if: github.event_name != 'pull_request' ||
(github.event_name == 'pull_request' && needs.setup-instance.result != 'skipped')
needs: [ should-run, setup-instance ]
concurrency:
group: ${{ github.workflow_ref }}
group: ${{ github.workflow }}_${{ github.ref }}
cancel-in-progress: true
runs-on: "runs-on=${{ github.run_id }}/runner=cpu-big"
runs-on: ${{ needs.setup-instance.outputs.runner-name }}
steps:
- name: Checkout tfhe-rs
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd
uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683
with:
persist-credentials: "false"
token: ${{ env.CHECKOUT_TOKEN }}
persist-credentials: 'false'
token: ${{ secrets.FHE_ACTIONS_TOKEN }}
- name: Install latest stable
uses: dtolnay/rust-toolchain@e97e2d8cc328f1b50210efc529dca0028893a2d9 # zizmor: ignore[stale-action-refs] this action doesn't create releases
uses: dtolnay/rust-toolchain@315e265cd78dad1e1dcf3a5074f6d6c47029d5aa
with:
toolchain: stable
@@ -161,11 +167,9 @@ jobs:
if: needs.should-run.outputs.csprng_test == 'true'
run: |
make test_tfhe_csprng
make test_tfhe_csprng_big_endian
- name: Run tfhe-zk-pok tests
# Always run it to catch non deterministic bugs earlier
# if: needs.should-run.outputs.zk_pok_test == 'true'
if: needs.should-run.outputs.zk_pok_test == 'true'
run: |
make test_zk_pok
@@ -174,11 +178,6 @@ jobs:
run: |
make test_versionable
- name: Run tfhe-safe-serialize tests
if: needs.should-run.outputs.safe_serialize_test == 'true'
run: |
make test_safe_serialize
- name: Run core tests
if: needs.should-run.outputs.core_crypto_test == 'true'
run: |
@@ -200,7 +199,7 @@ jobs:
- name: Node cache restoration
id: node-cache
uses: actions/cache/restore@27d5ce7f107fe9357f9df03efb73ab90386fccae #v5.0.5
uses: actions/cache/restore@1bd1e32a3bdc45362d1e726936510720a7c30a57 #v4.2.0
with:
path: |
~/.nvm
@@ -213,7 +212,7 @@ jobs:
make install_node
- name: Node cache save
uses: actions/cache/save@27d5ce7f107fe9357f9df03efb73ab90386fccae #v5.0.5
uses: actions/cache/save@1bd1e32a3bdc45362d1e726936510720a7c30a57 #v4.2.0
if: steps.node-cache.outputs.cache-hit != 'true'
with:
path: |
@@ -255,18 +254,34 @@ jobs:
run: |
make test_zk
- name: Set pull-request URL
if: ${{ failure() && github.event_name == 'pull_request' }}
run: |
echo "PULL_REQUEST_MD_LINK=[pull-request](${PR_BASE_URL}${PR_NUMBER}), " >> "${GITHUB_ENV}"
env:
PR_BASE_URL: ${{ vars.PR_BASE_URL }}
PR_NUMBER: ${{ github.event.pull_request.number }}
- name: Slack Notification
if: ${{ failure() && env.SECRETS_AVAILABLE == 'true' }}
if: ${{ failure() }}
continue-on-error: true
uses: rtCamp/action-slack-notify@e31e87e03dd19038e411e38ae27cbad084a90661
uses: rtCamp/action-slack-notify@c33737706dea87cd7784c687dadc9adf1be59990
env:
SLACK_COLOR: ${{ job.status }}
SLACK_MESSAGE: "Fast AWS tests finished with status: ${{ job.status }}. (${{ env.PULL_REQUEST_MD_LINK }}[action run](${{ env.ACTION_RUN_URL }}))"
SLACK_MESSAGE: "Fast AWS tests finished with status: ${{ job.status }}. (${{ env.ACTION_RUN_URL }})"
teardown-instance:
name: Teardown instance (fast-tests)
if: ${{ always() && needs.setup-instance.result != 'skipped' }}
needs: [ setup-instance, fast-tests ]
runs-on: ubuntu-latest
steps:
- name: Stop instance
id: stop-instance
uses: zama-ai/slab-github-runner@98f0788261a7323d5d695a883e20df36591a92b7
with:
mode: stop
github-token: ${{ secrets.SLAB_ACTION_TOKEN }}
slab-url: ${{ secrets.SLAB_BASE_URL }}
job-secret: ${{ secrets.JOB_SECRET }}
label: ${{ needs.setup-instance.outputs.runner-name }}
- name: Slack Notification
if: ${{ failure() }}
continue-on-error: true
uses: rtCamp/action-slack-notify@c33737706dea87cd7784c687dadc9adf1be59990
env:
SLACK_COLOR: ${{ job.status }}
SLACK_MESSAGE: "Instance teardown (fast-tests) finished with status: ${{ job.status }}. (${{ env.ACTION_RUN_URL }})"

View File

@@ -1,4 +1,4 @@
name: aws_tfhe_integer_tests
name: AWS Unsigned Integer Tests on CPU
env:
CARGO_TERM_COLOR: always
@@ -10,54 +10,46 @@ env:
SLACK_ICON: https://pbs.twimg.com/profile_images/1274014582265298945/OjBKP9kn_400x400.png
SLACK_USERNAME: ${{ secrets.BOT_USERNAME }}
SLACK_WEBHOOK: ${{ secrets.SLACK_WEBHOOK }}
SLACKIFY_MARKDOWN: true
PULL_REQUEST_MD_LINK: ""
# We clear the cache to reduce memory pressure because of the numerous processes of cargo
# nextest
TFHE_RS_CLEAR_IN_MEMORY_KEY_CACHE: "1"
NO_BIG_PARAMS: FALSE
CHECKOUT_TOKEN: ${{ secrets.REPO_CHECKOUT_TOKEN || secrets.GITHUB_TOKEN }}
on:
# Allows you to run this workflow manually from the Actions tab as an alternative.
workflow_dispatch:
pull_request:
types: [ labeled ]
types: [labeled]
push:
branches:
- main
permissions:
contents: read
# zizmor: ignore[concurrency-limits] concurrency is managed after instance setup to ensure safe provisioning
jobs:
should-run:
name: aws_tfhe_integer_tests/should-run
if:
(github.event_name == 'push' && github.repository == 'zama-ai/tfhe-rs') ||
(github.event_name == 'schedule' && github.repository == 'zama-ai/tfhe-rs') ||
(github.event_name == 'pull_request' && contains(github.event.label.name, 'approved')) ||
github.event_name == 'workflow_dispatch'
runs-on: ubuntu-latest
permissions:
pull-requests: read # Needed to check for file change
pull-requests: write
outputs:
integer_test: ${{ github.event_name == 'workflow_dispatch' ||
steps.changed-files.outputs.integer_any_changed }}
steps:
- name: Checkout tfhe-rs
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd
uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683
with:
fetch-depth: 0
persist-credentials: 'false'
token: ${{ env.CHECKOUT_TOKEN }}
token: ${{ secrets.FHE_ACTIONS_TOKEN }}
persist-credentials: "false"
- name: Check for file changes
id: changed-files
uses: tj-actions/changed-files@9426d40962ed5378910ee2e21d5f8c6fcbf2dd96 # v47.0.6
uses: tj-actions/changed-files@bab30c2299617f6615ec02a68b9a40d10bd21366
with:
since_last_remote_commit: true
files_yaml: |
integer:
- tfhe/Cargo.toml
@@ -69,28 +61,45 @@ jobs:
- tfhe/src/integer/**
- .github/workflows/aws_tfhe_integer_tests.yml
unsigned-integer-tests:
name: aws_tfhe_integer_tests/unsigned-integer-tests
setup-instance:
name: Setup instance (unsigned-integer-tests)
needs: should-run
if:
(github.event_name == 'push' && github.repository == 'zama-ai/tfhe-rs' && needs.should-run.outputs.integer_test == 'true') ||
(github.event_name == 'schedule' && github.repository == 'zama-ai/tfhe-rs') ||
(github.event.action == 'labeled' && github.event.label.name == 'approved' && needs.should-run.outputs.integer_test == 'true') ||
(github.event_name == 'pull_request' && contains(github.event.label.name, 'approved')) ||
github.event_name == 'workflow_dispatch'
runs-on: ubuntu-latest
outputs:
runner-name: ${{ steps.start-instance.outputs.label }}
steps:
- name: Start instance
id: start-instance
uses: zama-ai/slab-github-runner@98f0788261a7323d5d695a883e20df36591a92b7
with:
mode: start
github-token: ${{ secrets.SLAB_ACTION_TOKEN }}
slab-url: ${{ secrets.SLAB_BASE_URL }}
job-secret: ${{ secrets.JOB_SECRET }}
backend: aws
profile: cpu-big
unsigned-integer-tests:
name: Unsigned integer tests
needs: setup-instance
concurrency:
group: ${{ github.workflow_ref }}${{ github.ref == 'refs/heads/main' && github.sha || '' }}
group: ${{ github.workflow }}_${{ github.ref }}${{ github.ref == 'refs/heads/main' && github.sha || '' }}
cancel-in-progress: ${{ github.ref != 'refs/heads/main' }}
runs-on: "runs-on=${{ github.run_id }}/runner=cpu-big"
timeout-minutes: 480 # 8 hours
runs-on: ${{ needs.setup-instance.outputs.runner-name }}
steps:
- name: Checkout tfhe-rs
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd
uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683
with:
persist-credentials: "false"
token: ${{ env.CHECKOUT_TOKEN }}
token: ${{ secrets.FHE_ACTIONS_TOKEN }}
- name: Install latest stable
uses: dtolnay/rust-toolchain@e97e2d8cc328f1b50210efc529dca0028893a2d9 # zizmor: ignore[stale-action-refs] this action doesn't create releases
uses: dtolnay/rust-toolchain@315e265cd78dad1e1dcf3a5074f6d6c47029d5aa
with:
toolchain: stable
@@ -113,20 +122,36 @@ jobs:
- name: Run unsigned integer tests
run: |
AVX512_SUPPORT=ON NO_BIG_PARAMS="${NO_BIG_PARAMS}" BIG_TESTS_INSTANCE=TRUE make test_unsigned_integer_ci
- name: Set pull-request URL
if: ${{ failure() && github.event_name == 'pull_request' }}
run: |
echo "PULL_REQUEST_MD_LINK=[pull-request](${PR_BASE_URL}${PR_NUMBER}), " >> "${GITHUB_ENV}"
env:
PR_BASE_URL: ${{ vars.PR_BASE_URL }}
PR_NUMBER: ${{ github.event.pull_request.number }}
AVX512_SUPPORT=ON NO_BIG_PARAMS=${{ env.NO_BIG_PARAMS }} BIG_TESTS_INSTANCE=TRUE make test_unsigned_integer_ci
- name: Slack Notification
if: ${{ failure() || (cancelled() && github.event_name != 'pull_request') }}
if: ${{ failure() }}
continue-on-error: true
uses: rtCamp/action-slack-notify@e31e87e03dd19038e411e38ae27cbad084a90661
uses: rtCamp/action-slack-notify@c33737706dea87cd7784c687dadc9adf1be59990
env:
SLACK_COLOR: ${{ job.status }}
SLACK_MESSAGE: "Unsigned Integer tests finished with status: ${{ job.status }}. (${{ env.PULL_REQUEST_MD_LINK }}[action run](${{ env.ACTION_RUN_URL }}))"
SLACK_MESSAGE: "Unsigned Integer tests finished with status: ${{ job.status }}. (${{ env.ACTION_RUN_URL }})"
teardown-instance:
name: Teardown instance (unsigned-integer-tests)
if: ${{ always() && needs.setup-instance.result != 'skipped' }}
needs: [setup-instance, unsigned-integer-tests]
runs-on: ubuntu-latest
steps:
- name: Stop instance
id: stop-instance
uses: zama-ai/slab-github-runner@98f0788261a7323d5d695a883e20df36591a92b7
with:
mode: stop
github-token: ${{ secrets.SLAB_ACTION_TOKEN }}
slab-url: ${{ secrets.SLAB_BASE_URL }}
job-secret: ${{ secrets.JOB_SECRET }}
label: ${{ needs.setup-instance.outputs.runner-name }}
- name: Slack Notification
if: ${{ failure() }}
continue-on-error: true
uses: rtCamp/action-slack-notify@c33737706dea87cd7784c687dadc9adf1be59990
env:
SLACK_COLOR: ${{ job.status }}
SLACK_MESSAGE: "Instance teardown (unsigned-integer-tests) finished with status: ${{ job.status }}. (${{ env.ACTION_RUN_URL }})"

View File

@@ -1,115 +0,0 @@
name: aws_tfhe_noise_checks
env:
CARGO_TERM_COLOR: always
ACTION_RUN_URL: ${{ github.server_url }}/${{ github.repository }}/actions/runs/${{ github.run_id }}
RUSTFLAGS: "-C target-cpu=native"
RUST_BACKTRACE: "full"
RUST_MIN_STACK: "8388608"
SLACK_CHANNEL: ${{ secrets.SLACK_CHANNEL }}
SLACK_ICON: https://pbs.twimg.com/profile_images/1274014582265298945/OjBKP9kn_400x400.png
SLACK_USERNAME: ${{ secrets.BOT_USERNAME }}
SLACK_WEBHOOK: ${{ secrets.SLACK_WEBHOOK }}
SLACKIFY_MARKDOWN: true
PULL_REQUEST_MD_LINK: ""
CHECKOUT_TOKEN: ${{ secrets.REPO_CHECKOUT_TOKEN || secrets.GITHUB_TOKEN }}
on:
# Allows you to run this workflow manually from the Actions tab as an alternative.
workflow_dispatch:
permissions:
contents: read
# zizmor: ignore[concurrency-limits] only Zama organization members can trigger this workflow
jobs:
setup-instance:
name: aws_tfhe_noise_checks/setup-instance
runs-on: ubuntu-latest
outputs:
runner-name: ${{ steps.start-remote-instance.outputs.label || steps.start-github-instance.outputs.runner_group }}
steps:
- name: Start remote instance
id: start-remote-instance
if: env.SECRETS_AVAILABLE == 'true'
uses: zama-ai/slab-github-runner@5aee5d157f4a0201e5eaefc9cc648e5f9f5472a5 # v1.6.0
with:
mode: start
github-token: ${{ secrets.SLAB_ACTION_TOKEN }}
slab-url: ${{ secrets.SLAB_BASE_URL }}
job-secret: ${{ secrets.JOB_SECRET }}
backend: aws
# We want an hpc7a more compute, will be faster
profile: bench
# This instance will be spawned especially for pull-request from forked repository
- name: Start GitHub instance
id: start-github-instance
if: env.SECRETS_AVAILABLE == 'false'
run: |
echo "Cannot run this without secrets"
exit 1
noise-checks:
name: aws_tfhe_noise_checks/noise-checks
needs: setup-instance
runs-on: ${{ needs.setup-instance.outputs.runner-name }}
timeout-minutes: 1440
steps:
- name: Checkout tfhe-rs
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd
with:
persist-credentials: 'false'
token: ${{ env.CHECKOUT_TOKEN }}
- name: Install latest stable
uses: dtolnay/rust-toolchain@e97e2d8cc328f1b50210efc529dca0028893a2d9 # zizmor: ignore[stale-action-refs] this action doesn't create releases
with:
toolchain: stable
- name: Run noise checks
timeout-minutes: 1440
run: |
make test_noise_check
- name: Set pull-request URL
if: ${{ !success() }}
run: |
echo "PULL_REQUEST_MD_LINK=[pull-request](${PR_BASE_URL}${PR_NUMBER}), " >> "${GITHUB_ENV}"
env:
PR_BASE_URL: ${{ vars.PR_BASE_URL }}
PR_NUMBER: ${{ github.event.pull_request.number }}
- name: Slack Notification
if: ${{ !success() }}
continue-on-error: true
uses: rtCamp/action-slack-notify@e31e87e03dd19038e411e38ae27cbad084a90661
env:
SLACK_COLOR: ${{ job.status }}
SLACK_MESSAGE: "Noise checks tests finished with status: ${{ job.status }}. (${{ env.PULL_REQUEST_MD_LINK }}[action run](${{ env.ACTION_RUN_URL }}))"
teardown-instance:
name: aws_tfhe_noise_checks/teardown-instance
if: ${{ always() && needs.setup-instance.result == 'success' }}
needs: [ setup-instance, noise-checks ]
runs-on: ubuntu-latest
steps:
- name: Stop remote instance
id: stop-instance
if: env.SECRETS_AVAILABLE == 'true'
uses: zama-ai/slab-github-runner@5aee5d157f4a0201e5eaefc9cc648e5f9f5472a5 # v1.6.0
with:
mode: stop
github-token: ${{ secrets.SLAB_ACTION_TOKEN }}
slab-url: ${{ secrets.SLAB_BASE_URL }}
job-secret: ${{ secrets.JOB_SECRET }}
label: ${{ needs.setup-instance.outputs.runner-name }}
- name: Slack Notification
if: ${{ !success() }}
uses: rtCamp/action-slack-notify@e31e87e03dd19038e411e38ae27cbad084a90661
env:
SLACK_COLOR: ${{ job.status }}
SLACK_MESSAGE: "Instance teardown (noise-checks) finished with status: ${{ job.status }}. (${{ env.ACTION_RUN_URL }})"

View File

@@ -1,4 +1,4 @@
name: aws_tfhe_signed_integer_tests
name: AWS Signed Integer Tests on CPU
env:
CARGO_TERM_COLOR: always
@@ -10,32 +10,22 @@ env:
SLACK_ICON: https://pbs.twimg.com/profile_images/1274014582265298945/OjBKP9kn_400x400.png
SLACK_USERNAME: ${{ secrets.BOT_USERNAME }}
SLACK_WEBHOOK: ${{ secrets.SLACK_WEBHOOK }}
SLACKIFY_MARKDOWN: true
PULL_REQUEST_MD_LINK: ""
# We clear the cache to reduce memory pressure because of the numerous processes of cargo
# nextest
TFHE_RS_CLEAR_IN_MEMORY_KEY_CACHE: "1"
NO_BIG_PARAMS: FALSE
CHECKOUT_TOKEN: ${{ secrets.REPO_CHECKOUT_TOKEN || secrets.GITHUB_TOKEN }}
on:
# Allows you to run this workflow manually from the Actions tab as an alternative.
workflow_dispatch:
pull_request:
types: [ labeled ]
types: [labeled]
push:
branches:
- main
permissions:
contents: read
# zizmor: ignore[concurrency-limits] concurrency is managed after instance setup to ensure safe provisioning
jobs:
should-run:
name: aws_tfhe_signed_integer_tests/should-run
if:
(github.event_name == 'push' && github.repository == 'zama-ai/tfhe-rs') ||
(github.event_name == 'schedule' && github.repository == 'zama-ai/tfhe-rs') ||
@@ -43,22 +33,23 @@ jobs:
github.event_name == 'workflow_dispatch'
runs-on: ubuntu-latest
permissions:
pull-requests: read # Needed to check for file change
pull-requests: write
outputs:
integer_test: ${{ github.event_name == 'workflow_dispatch' ||
steps.changed-files.outputs.integer_any_changed }}
steps:
- name: Checkout tfhe-rs
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd
uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683
with:
fetch-depth: 0
persist-credentials: 'false'
token: ${{ env.CHECKOUT_TOKEN }}
token: ${{ secrets.FHE_ACTIONS_TOKEN }}
persist-credentials: "false"
- name: Check for file changes
id: changed-files
uses: tj-actions/changed-files@9426d40962ed5378910ee2e21d5f8c6fcbf2dd96 # v47.0.6
uses: tj-actions/changed-files@bab30c2299617f6615ec02a68b9a40d10bd21366
with:
since_last_remote_commit: true
files_yaml: |
integer:
- tfhe/Cargo.toml
@@ -70,27 +61,45 @@ jobs:
- tfhe/src/integer/**
- .github/workflows/aws_tfhe_signed_integer_tests.yml
signed-integer-tests:
name: aws_tfhe_signed_integer_tests/signed-integer-tests
setup-instance:
name: Setup instance (unsigned-integer-tests)
needs: should-run
if:
(github.event_name == 'push' && github.repository == 'zama-ai/tfhe-rs' && needs.should-run.outputs.integer_test == 'true') ||
(github.event_name == 'schedule' && github.repository == 'zama-ai/tfhe-rs') ||
(github.event.action == 'labeled' && github.event.label.name == 'approved' && needs.should-run.outputs.integer_test == 'true') ||
(github.event_name == 'pull_request' && contains(github.event.label.name, 'approved')) ||
github.event_name == 'workflow_dispatch'
runs-on: ubuntu-latest
outputs:
runner-name: ${{ steps.start-instance.outputs.label }}
steps:
- name: Start instance
id: start-instance
uses: zama-ai/slab-github-runner@98f0788261a7323d5d695a883e20df36591a92b7
with:
mode: start
github-token: ${{ secrets.SLAB_ACTION_TOKEN }}
slab-url: ${{ secrets.SLAB_BASE_URL }}
job-secret: ${{ secrets.JOB_SECRET }}
backend: aws
profile: cpu-big
signed-integer-tests:
name: Signed integer tests
needs: setup-instance
concurrency:
group: ${{ github.workflow_ref }}${{ github.ref == 'refs/heads/main' && github.sha || '' }}
group: ${{ github.workflow }}_${{ github.ref }}${{ github.ref == 'refs/heads/main' && github.sha || '' }}
cancel-in-progress: ${{ github.ref != 'refs/heads/main' }}
runs-on: "runs-on=${{ github.run_id }}/runner=cpu-big"
runs-on: ${{ needs.setup-instance.outputs.runner-name }}
steps:
- name: Checkout tfhe-rs
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd
uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683
with:
persist-credentials: "false"
token: ${{ env.CHECKOUT_TOKEN }}
token: ${{ secrets.FHE_ACTIONS_TOKEN }}
- name: Install latest stable
uses: dtolnay/rust-toolchain@e97e2d8cc328f1b50210efc529dca0028893a2d9 # zizmor: ignore[stale-action-refs] this action doesn't create releases
uses: dtolnay/rust-toolchain@315e265cd78dad1e1dcf3a5074f6d6c47029d5aa
with:
toolchain: stable
@@ -117,20 +126,36 @@ jobs:
- name: Run signed integer tests
run: |
AVX512_SUPPORT=ON NO_BIG_PARAMS="${NO_BIG_PARAMS}" BIG_TESTS_INSTANCE=TRUE make test_signed_integer_ci
- name: Set pull-request URL
if: ${{ failure() && github.event_name == 'pull_request' }}
run: |
echo "PULL_REQUEST_MD_LINK=[pull-request](${PR_BASE_URL}${PR_NUMBER}), " >> "${GITHUB_ENV}"
env:
PR_BASE_URL: ${{ vars.PR_BASE_URL }}
PR_NUMBER: ${{ github.event.pull_request.number }}
AVX512_SUPPORT=ON NO_BIG_PARAMS=${{ env.NO_BIG_PARAMS }} BIG_TESTS_INSTANCE=TRUE make test_signed_integer_ci
- name: Slack Notification
if: ${{ failure() || (cancelled() && github.event_name != 'pull_request') }}
if: ${{ failure() }}
continue-on-error: true
uses: rtCamp/action-slack-notify@e31e87e03dd19038e411e38ae27cbad084a90661
uses: rtCamp/action-slack-notify@c33737706dea87cd7784c687dadc9adf1be59990
env:
SLACK_COLOR: ${{ job.status }}
SLACK_MESSAGE: "Signed Integer tests finished with status: ${{ job.status }}. (${{ env.PULL_REQUEST_MD_LINK }}[action run](${{ env.ACTION_RUN_URL }}))"
SLACK_MESSAGE: "Signed Integer tests finished with status: ${{ job.status }}. (${{ env.ACTION_RUN_URL }})"
teardown-instance:
name: Teardown instance (signed-integer-tests)
if: ${{ always() && needs.setup-instance.result != 'skipped' }}
needs: [setup-instance, signed-integer-tests]
runs-on: ubuntu-latest
steps:
- name: Stop instance
id: stop-instance
uses: zama-ai/slab-github-runner@98f0788261a7323d5d695a883e20df36591a92b7
with:
mode: stop
github-token: ${{ secrets.SLAB_ACTION_TOKEN }}
slab-url: ${{ secrets.SLAB_BASE_URL }}
job-secret: ${{ secrets.JOB_SECRET }}
label: ${{ needs.setup-instance.outputs.runner-name }}
- name: Slack Notification
if: ${{ failure() }}
continue-on-error: true
uses: rtCamp/action-slack-notify@c33737706dea87cd7784c687dadc9adf1be59990
env:
SLACK_COLOR: ${{ job.status }}
SLACK_MESSAGE: "Instance teardown (signed-integer-tests) finished with status: ${{ job.status }}. (${{ env.ACTION_RUN_URL }})"

View File

@@ -1,4 +1,4 @@
name: aws_tfhe_tests
name: AWS Tests on CPU
env:
CARGO_TERM_COLOR: always
@@ -10,11 +10,7 @@ env:
SLACK_ICON: https://pbs.twimg.com/profile_images/1274014582265298945/OjBKP9kn_400x400.png
SLACK_USERNAME: ${{ secrets.BOT_USERNAME }}
SLACK_WEBHOOK: ${{ secrets.SLACK_WEBHOOK }}
SLACKIFY_MARKDOWN: true
IS_PULL_REQUEST: ${{ github.event_name == 'pull_request' }}
PULL_REQUEST_MD_LINK: ""
CHECKOUT_TOKEN: ${{ secrets.REPO_CHECKOUT_TOKEN || secrets.GITHUB_TOKEN }}
on:
# Allows you to run this workflow manually from the Actions tab as an alternative.
@@ -25,19 +21,13 @@ on:
# Nightly tests @ 1AM after each work day
- cron: "0 1 * * MON-FRI"
permissions:
contents: read
# zizmor: ignore[concurrency-limits] concurrency is managed after instance setup to ensure safe provisioning
jobs:
should-run:
name: aws_tfhe_tests/should-run
runs-on: ubuntu-latest
if: github.event_name != 'schedule' ||
(github.event_name == 'schedule' && github.repository == 'zama-ai/tfhe-rs')
permissions:
pull-requests: read # Needed to check for file change
pull-requests: write
outputs:
csprng_test: ${{ env.IS_PULL_REQUEST == 'false' || steps.changed-files.outputs.csprng_any_changed }}
zk_pok_test: ${{ env.IS_PULL_REQUEST == 'false' || steps.changed-files.outputs.zk_pok_any_changed }}
@@ -70,16 +60,16 @@ jobs:
any_file_changed: ${{ env.IS_PULL_REQUEST == 'false' || steps.aggregated-changes.outputs.any_changed }}
steps:
- name: Checkout tfhe-rs
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd
uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683
with:
fetch-depth: 0
persist-credentials: 'false'
token: ${{ env.CHECKOUT_TOKEN }}
token: ${{ secrets.FHE_ACTIONS_TOKEN }}
- name: Check for file changes
id: changed-files
uses: tj-actions/changed-files@9426d40962ed5378910ee2e21d5f8c6fcbf2dd96 # v47.0.6
uses: tj-actions/changed-files@bab30c2299617f6615ec02a68b9a40d10bd21366
with:
since_last_remote_commit: true
files_yaml: |
dependencies:
- tfhe/Cargo.toml
@@ -121,7 +111,7 @@ jobs:
user_docs:
- tfhe/src/**
- '!tfhe/src/c_api/**'
- 'tfhe/docs/**/**.md'
- 'tfhe/docs/**.md'
- README.md
- name: Aggregate file changes
@@ -141,24 +131,44 @@ jobs:
run: |
echo "any_changed=true" >> "$GITHUB_OUTPUT"
cpu-tests:
name: aws_tfhe_tests/cpu-tests
needs: should-run
setup-instance:
name: Setup instance (cpu-tests)
if: github.event_name != 'pull_request' ||
(github.event.action == 'labeled' && github.event.label.name == 'approved' && needs.should-run.outputs.any_file_changed == 'true')
needs: should-run
runs-on: ubuntu-latest
outputs:
runner-name: ${{ steps.start-instance.outputs.label }}
steps:
- name: Start instance
id: start-instance
uses: zama-ai/slab-github-runner@98f0788261a7323d5d695a883e20df36591a92b7
with:
mode: start
github-token: ${{ secrets.SLAB_ACTION_TOKEN }}
slab-url: ${{ secrets.SLAB_BASE_URL }}
job-secret: ${{ secrets.JOB_SECRET }}
backend: aws
profile: cpu-big
cpu-tests:
name: CPU tests
if: github.event_name != 'pull_request' ||
(github.event_name == 'pull_request' && needs.setup-instance.result != 'skipped')
needs: [ should-run, setup-instance ]
concurrency:
group: ${{ github.workflow_ref }}_${{github.event_name}}
group: ${{ github.workflow }}_${{github.event_name}}_${{ github.ref }}
cancel-in-progress: true
runs-on: "runs-on=${{ github.run_id }}/runner=cpu-big"
runs-on: ${{ needs.setup-instance.outputs.runner-name }}
steps:
- name: Checkout tfhe-rs
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd
uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683
with:
persist-credentials: 'false'
token: ${{ env.CHECKOUT_TOKEN }}
token: ${{ secrets.FHE_ACTIONS_TOKEN }}
- name: Install latest stable
uses: dtolnay/rust-toolchain@e97e2d8cc328f1b50210efc529dca0028893a2d9 # zizmor: ignore[stale-action-refs] this action doesn't create releases
uses: dtolnay/rust-toolchain@315e265cd78dad1e1dcf3a5074f6d6c47029d5aa
with:
toolchain: stable
@@ -224,18 +234,34 @@ jobs:
make test_trivium
make test_kreyvium
- name: Set pull-request URL
if: ${{ failure() && github.event_name == 'pull_request' }}
run: |
echo "PULL_REQUEST_MD_LINK=[pull-request](${PR_BASE_URL}${PR_NUMBER}), " >> "${GITHUB_ENV}"
env:
PR_BASE_URL: ${{ vars.PR_BASE_URL }}
PR_NUMBER: ${{ github.event.pull_request.number }}
- name: Slack Notification
if: ${{ failure() || (cancelled() && github.event_name != 'pull_request') }}
if: ${{ failure() }}
continue-on-error: true
uses: rtCamp/action-slack-notify@e31e87e03dd19038e411e38ae27cbad084a90661
uses: rtCamp/action-slack-notify@c33737706dea87cd7784c687dadc9adf1be59990
env:
SLACK_COLOR: ${{ job.status }}
SLACK_MESSAGE: "CPU tests finished with status: ${{ job.status }}. (${{ env.PULL_REQUEST_MD_LINK }}[action run](${{ env.ACTION_RUN_URL }}))"
SLACK_MESSAGE: "CPU tests finished with status: ${{ job.status }}. (${{ env.ACTION_RUN_URL }})"
teardown-instance:
name: Teardown instance (cpu-tests)
if: ${{ always() && needs.setup-instance.result != 'skipped' }}
needs: [ setup-instance, cpu-tests ]
runs-on: ubuntu-latest
steps:
- name: Stop instance
id: stop-instance
uses: zama-ai/slab-github-runner@98f0788261a7323d5d695a883e20df36591a92b7
with:
mode: stop
github-token: ${{ secrets.SLAB_ACTION_TOKEN }}
slab-url: ${{ secrets.SLAB_BASE_URL }}
job-secret: ${{ secrets.JOB_SECRET }}
label: ${{ needs.setup-instance.outputs.runner-name }}
- name: Slack Notification
if: ${{ failure() }}
continue-on-error: true
uses: rtCamp/action-slack-notify@c33737706dea87cd7784c687dadc9adf1be59990
env:
SLACK_COLOR: ${{ job.status }}
SLACK_MESSAGE: "Instance teardown (cpu-tests) finished with status: ${{ job.status }}. (${{ env.ACTION_RUN_URL }})"

View File

@@ -1,4 +1,4 @@
name: aws_tfhe_wasm_tests
name: AWS WASM Tests on CPU
env:
CARGO_TERM_COLOR: always
@@ -10,79 +10,48 @@ env:
SLACK_ICON: https://pbs.twimg.com/profile_images/1274014582265298945/OjBKP9kn_400x400.png
SLACK_USERNAME: ${{ secrets.BOT_USERNAME }}
SLACK_WEBHOOK: ${{ secrets.SLACK_WEBHOOK }}
SLACKIFY_MARKDOWN: true
PULL_REQUEST_MD_LINK: ""
CHECKOUT_TOKEN: ${{ secrets.REPO_CHECKOUT_TOKEN || secrets.GITHUB_TOKEN }}
on:
# Allows you to run this workflow manually from the Actions tab as an alternative.
workflow_dispatch:
pull_request:
types: [labeled]
permissions:
contents: read
# zizmor: ignore[concurrency-limits] concurrency is managed after instance setup to ensure safe provisioning
types: [ labeled ]
jobs:
should-run:
name: aws_tfhe_wasm_tests/should-run
if: github.event_name == 'workflow_dispatch' || contains(github.event.label.name, 'approved')
setup-instance:
name: Setup instance (wasm-tests)
if: ${{ github.event_name == 'workflow_dispatch' || contains(github.event.label.name, 'approved') }}
runs-on: ubuntu-latest
permissions:
pull-requests: read # Needed to check for file change
outputs:
wasm_test: ${{ github.event_name == 'workflow_dispatch' ||
steps.changed-files.outputs.wasm_any_changed }}
runner-name: ${{ steps.start-instance.outputs.label }}
steps:
- name: Checkout tfhe-rs
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd
- name: Start instance
id: start-instance
uses: zama-ai/slab-github-runner@98f0788261a7323d5d695a883e20df36591a92b7
with:
fetch-depth: 0
persist-credentials: "false"
token: ${{ env.CHECKOUT_TOKEN }}
- name: Check for file changes
id: changed-files
uses: tj-actions/changed-files@9426d40962ed5378910ee2e21d5f8c6fcbf2dd96 # v47.0.6
with:
files_yaml: |
wasm:
- Cargo.toml
- tfhe/Cargo.toml
- tfhe-csprng/**
- tfhe-fft/**
- tfhe-zk-pok/**
- tfhe/src/core_crypto/**
- tfhe/src/shortint/**
- tfhe/src/integer/**
- tfhe/src/high_level_api/**
- tfhe/src/js_on_wasm_api/**
- tfhe/js_on_wasm_tests/**
- tfhe/web_wasm_parallel_tests/**
- utils/tfhe-versionable/**
- utils/tfhe-safe-serialize/**
- .github/workflows/aws_tfhe_wasm_tests.yml
mode: start
github-token: ${{ secrets.SLAB_ACTION_TOKEN }}
slab-url: ${{ secrets.SLAB_BASE_URL }}
job-secret: ${{ secrets.JOB_SECRET }}
backend: aws
profile: cpu-small
wasm-tests:
name: aws_tfhe_wasm_tests/wasm-tests
needs: should-run
if: github.event_name == 'workflow_dispatch' ||
(contains(github.event.label.name, 'approved') && needs.should-run.outputs.wasm_test == 'true')
name: WASM tests
needs: setup-instance
concurrency:
group: ${{ github.workflow_ref }}_${{github.event_name}}
group: ${{ github.workflow }}_${{ github.ref }}
cancel-in-progress: true
runs-on: "runs-on=${{ github.run_id }}/runner=cpu-small"
runs-on: ${{ needs.setup-instance.outputs.runner-name }}
steps:
- name: Checkout tfhe-rs
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd
uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683
with:
persist-credentials: "false"
token: ${{ env.CHECKOUT_TOKEN }}
persist-credentials: 'false'
token: ${{ secrets.FHE_ACTIONS_TOKEN }}
- name: Install latest stable
uses: dtolnay/rust-toolchain@e97e2d8cc328f1b50210efc529dca0028893a2d9 # zizmor: ignore[stale-action-refs] this action doesn't create releases
uses: dtolnay/rust-toolchain@315e265cd78dad1e1dcf3a5074f6d6c47029d5aa
with:
toolchain: stable
@@ -92,7 +61,7 @@ jobs:
- name: Node cache restoration
id: node-cache
uses: actions/cache/restore@27d5ce7f107fe9357f9df03efb73ab90386fccae #v5.0.5
uses: actions/cache/restore@1bd1e32a3bdc45362d1e726936510720a7c30a57 #v4.2.0
with:
path: |
~/.nvm
@@ -105,7 +74,7 @@ jobs:
make install_node
- name: Node cache save
uses: actions/cache/save@27d5ce7f107fe9357f9df03efb73ab90386fccae #v5.0.5
uses: actions/cache/save@1bd1e32a3bdc45362d1e726936510720a7c30a57 #v4.2.0
if: steps.node-cache.outputs.cache-hit != 'true'
with:
path: |
@@ -117,8 +86,6 @@ jobs:
run: |
make install_chrome_browser
make install_chrome_web_driver
make install_firefox_browser
make install_firefox_web_driver
- name: Run fmt checks
run: |
@@ -128,37 +95,42 @@ jobs:
run: |
make test_nodejs_wasm_api_ci
- name: Run wasm_par_mq tests
run: |
make test_wasm_par_mq_chrome_ci
make test_wasm_par_mq_firefox_ci
- name: Run parallel wasm tests
run: |
make test_web_js_api_parallel_chrome_ci
make test_web_js_api_parallel_firefox_ci
- name: Run cross origin wasm tests
run: |
make test_web_js_api_cross_origin_chrome_ci
make test_web_js_api_cross_origin_firefox_ci
- name: Run x86_64/wasm zk compatibility tests
run: |
make test_zk_wasm_x86_compat_ci
- name: Set pull-request URL
if: ${{ failure() && github.event_name == 'pull_request' }}
run: |
echo "PULL_REQUEST_MD_LINK=[pull-request](${PR_BASE_URL}${PR_NUMBER}), " >> "${GITHUB_ENV}"
env:
PR_BASE_URL: ${{ vars.PR_BASE_URL }}
PR_NUMBER: ${{ github.event.pull_request.number }}
- name: Slack Notification
if: ${{ failure() || (cancelled() && github.event_name != 'pull_request') }}
if: ${{ failure() }}
continue-on-error: true
uses: rtCamp/action-slack-notify@e31e87e03dd19038e411e38ae27cbad084a90661
uses: rtCamp/action-slack-notify@c33737706dea87cd7784c687dadc9adf1be59990
env:
SLACK_COLOR: ${{ job.status }}
SLACK_MESSAGE: "WASM tests finished with status: ${{ job.status }}. (${{ env.PULL_REQUEST_MD_LINK }}[action run](${{ env.ACTION_RUN_URL }}))"
SLACK_MESSAGE: "WASM tests finished with status: ${{ job.status }}. (${{ env.ACTION_RUN_URL }})"
teardown-instance:
name: Teardown instance (wasm-tests)
if: ${{ always() && needs.setup-instance.result != 'skipped' }}
needs: [ setup-instance, wasm-tests ]
runs-on: ubuntu-latest
steps:
- name: Stop instance
id: stop-instance
uses: zama-ai/slab-github-runner@98f0788261a7323d5d695a883e20df36591a92b7
with:
mode: stop
github-token: ${{ secrets.SLAB_ACTION_TOKEN }}
slab-url: ${{ secrets.SLAB_BASE_URL }}
job-secret: ${{ secrets.JOB_SECRET }}
label: ${{ needs.setup-instance.outputs.runner-name }}
- name: Slack Notification
if: ${{ failure() }}
continue-on-error: true
uses: rtCamp/action-slack-notify@c33737706dea87cd7784c687dadc9adf1be59990
env:
SLACK_COLOR: ${{ job.status }}
SLACK_MESSAGE: "Instance teardown (wasm-tests) finished with status: ${{ job.status }}. (${{ env.ACTION_RUN_URL }})"

View File

@@ -1,89 +0,0 @@
# On PR: when snapshot files are modified, generates a diff report between
# the base branch and the PR snapshots, then posts it as a PR comment.
# Helps reviewers understand what versioned types changed.
name: backward_compat_pr_change_report
on:
pull_request:
env:
CHECKOUT_TOKEN: ${{ secrets.REPO_CHECKOUT_TOKEN || secrets.GITHUB_TOKEN }}
permissions:
contents: read
concurrency:
group: ${{ github.workflow }}-${{ github.head_ref }}
cancel-in-progress: true
jobs:
should-run:
name: backward_compat_pr_change_report/should-run
runs-on: ubuntu-latest
permissions:
pull-requests: read # Needed to check for file change
outputs:
backward_report: ${{ steps.changed-files.outputs.backward_any_changed }}
steps:
- name: Checkout tfhe-rs
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd
with:
fetch-depth: 0
persist-credentials: 'false'
token: ${{ env.CHECKOUT_TOKEN }}
- name: Check for file changes
id: changed-files
uses: tj-actions/changed-files@9426d40962ed5378910ee2e21d5f8c6fcbf2dd96 # v47.0.6
with:
files_yaml: |
backward:
- utils/tfhe-lints/snapshots/*.json
change-report:
name: backward_compat_pr_change_report/change-report (bpr)
runs-on: ubuntu-latest
needs: should-run
if:
needs.should-run.outputs.backward_report == 'true'
permissions:
pull-requests: write # To send and modify message in the PR
steps:
- name: Checkout PR head
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd
with:
persist-credentials: 'false'
path: head
- name: Checkout base branch
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd
with:
persist-credentials: 'false'
ref: ${{ github.event.pull_request.base.sha }}
path: base
- name: Generate diff report
id: report
run: |
cd head && make backward_snapshot_report \
BASE_SNAPSHOT_DIR=../base/utils/tfhe-lints/snapshots \
HEAD_SNAPSHOT_DIR=utils/tfhe-lints/snapshots \
OUTPUT_FILE=../report.md
if [ -s ../report.md ]; then
echo "has_report=true" >> "$GITHUB_OUTPUT"
elif [ -f ../report.md ]; then
echo "has_report=false" >> "$GITHUB_OUTPUT"
else
echo "::error::report.md was not created — something went wrong"
exit 1
fi
- name: Post/refresh backward-compat report
if: steps.report.outputs.has_report == 'true'
uses: marocchino/sticky-pull-request-comment@0ea0beb66eb9baf113663a64ec522f60e49231c0
with:
header: backward-compat-snapshot
hide_and_recreate: true
hide_classify: OUTDATED
path: report.md

View File

@@ -1,54 +0,0 @@
# Generates snapshots from code and diffs against committed base files.
# Ensures snapshots are up to date on PRs and catches stale ones on main.
name: backward_compat_snapshot_consistency
env:
CARGO_TERM_COLOR: always
ACTION_RUN_URL: ${{ github.server_url }}/${{ github.repository }}/actions/runs/${{ github.run_id }}
RUST_BACKTRACE: "full"
RUST_MIN_STACK: "8388608"
SLACK_CHANNEL: ${{ secrets.SLACK_CHANNEL }}
SLACK_ICON: https://pbs.twimg.com/profile_images/1274014582265298945/OjBKP9kn_400x400.png
SLACK_USERNAME: ${{ secrets.BOT_USERNAME }}
SLACK_WEBHOOK: ${{ secrets.SLACK_WEBHOOK }}
SLACKIFY_MARKDOWN: true
on:
workflow_dispatch:
pull_request:
push:
branches:
- main
permissions:
contents: read
jobs:
snapshot-consistency:
name: backward_compat_snapshot_consistency/snapshot-consistency (bpr)
runs-on: ubuntu-latest
concurrency:
group: ${{ github.workflow }}-${{ github.head_ref || github.sha }}
cancel-in-progress: ${{ github.event_name != 'push' }}
steps:
- name: Checkout tfhe-rs
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd
with:
persist-credentials: 'false'
- name: Generate snapshots from current code
run: make backward_snapshot_head
- name: Check snapshot consistency
run: ./scripts/check_snapshot_consistency.sh
- name: Slack Notification
if: ${{ failure() || (cancelled() && github.event_name == 'push') }}
continue-on-error: true
uses: rtCamp/action-slack-notify@e31e87e03dd19038e411e38ae27cbad084a90661
env:
SLACK_COLOR: ${{ job.status }}
SLACK_MESSAGE: >-
Backward compatibility snapshot consistency check: ${{ job.status }}.
Snapshots may be outdated — run `make backward_snapshot_base` and commit.
[See details](${{ env.ACTION_RUN_URL }})

144
.github/workflows/benchmark_boolean.yml vendored Normal file
View File

@@ -0,0 +1,144 @@
# Run boolean benchmarks on an AWS instance and return parsed results to Slab CI bot.
name: Boolean benchmarks
on:
workflow_dispatch:
schedule:
# Weekly benchmarks will be triggered each Saturday at 1a.m.
- cron: '0 1 * * 6'
env:
CARGO_TERM_COLOR: always
RESULTS_FILENAME: parsed_benchmark_results_${{ github.sha }}.json
ACTION_RUN_URL: ${{ github.server_url }}/${{ github.repository }}/actions/runs/${{ github.run_id }}
RUST_BACKTRACE: "full"
RUST_MIN_STACK: "8388608"
SLACK_CHANNEL: ${{ secrets.SLACK_CHANNEL }}
SLACK_ICON: https://pbs.twimg.com/profile_images/1274014582265298945/OjBKP9kn_400x400.png
SLACK_USERNAME: ${{ secrets.BOT_USERNAME }}
SLACK_WEBHOOK: ${{ secrets.SLACK_WEBHOOK }}
jobs:
setup-instance:
name: Setup instance (boolean-benchmarks)
runs-on: ubuntu-latest
if: github.event_name != 'schedule' ||
(github.event_name == 'schedule' && github.repository == 'zama-ai/tfhe-rs')
outputs:
runner-name: ${{ steps.start-instance.outputs.label }}
steps:
- name: Start instance
id: start-instance
uses: zama-ai/slab-github-runner@98f0788261a7323d5d695a883e20df36591a92b7
with:
mode: start
github-token: ${{ secrets.SLAB_ACTION_TOKEN }}
slab-url: ${{ secrets.SLAB_BASE_URL }}
job-secret: ${{ secrets.JOB_SECRET }}
backend: aws
profile: bench
boolean-benchmarks:
name: Execute boolean benchmarks in EC2
needs: setup-instance
runs-on: ${{ needs.setup-instance.outputs.runner-name }}
concurrency:
group: ${{ github.workflow }}_${{ github.ref }}
cancel-in-progress: ${{ github.ref != 'refs/heads/main' }}
continue-on-error: true
steps:
- name: Checkout tfhe-rs repo with tags
uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683
with:
fetch-depth: 0
token: ${{ secrets.FHE_ACTIONS_TOKEN }}
- name: Get benchmark details
run: |
{
echo "BENCH_DATE=$(date --iso-8601=seconds)";
echo "COMMIT_DATE=$(git --no-pager show -s --format=%cd --date=iso8601-strict ${{ github.sha }})";
echo "COMMIT_HASH=$(git describe --tags --dirty)";
} >> "${GITHUB_ENV}"
- name: Install rust
uses: dtolnay/rust-toolchain@315e265cd78dad1e1dcf3a5074f6d6c47029d5aa
with:
toolchain: nightly
- name: Run benchmarks with AVX512
run: |
make bench_boolean
- name: Parse results
run: |
python3 ./ci/benchmark_parser.py target/criterion ${{ env.RESULTS_FILENAME }} \
--database tfhe_rs \
--hardware "hpc7a.96xlarge" \
--project-version "${{ env.COMMIT_HASH }}" \
--branch ${{ github.ref_name }} \
--commit-date "${{ env.COMMIT_DATE }}" \
--bench-date "${{ env.BENCH_DATE }}" \
--walk-subdirs \
--name-suffix avx512
- name: Measure key sizes
run: |
make measure_boolean_key_sizes
- name: Parse key sizes results
run: |
python3 ./ci/benchmark_parser.py tfhe/boolean_key_sizes.csv ${{ env.RESULTS_FILENAME }} \
--object-sizes \
--append-results
- name: Upload parsed results artifact
uses: actions/upload-artifact@b4b15b8c7c6ac21ea08fcf65892d2ee8f75cf882
with:
name: ${{ github.sha }}_boolean
path: ${{ env.RESULTS_FILENAME }}
- name: Checkout Slab repo
uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683
with:
repository: zama-ai/slab
path: slab
token: ${{ secrets.FHE_ACTIONS_TOKEN }}
- name: Send data to Slab
shell: bash
run: |
python3 slab/scripts/data_sender.py ${{ env.RESULTS_FILENAME }} "${{ secrets.JOB_SECRET }}" \
--slab-url "${{ secrets.SLAB_URL }}"
- name: Slack Notification
if: ${{ failure() }}
continue-on-error: true
uses: rtCamp/action-slack-notify@c33737706dea87cd7784c687dadc9adf1be59990
env:
SLACK_COLOR: ${{ job.status }}
SLACK_MESSAGE: "Boolean benchmarks finished with status: ${{ job.status }}. (${{ env.ACTION_RUN_URL }})"
teardown-instance:
name: Teardown instance (boolean-benchmarks)
if: ${{ always() && needs.setup-instance.result != 'skipped' }}
needs: [ setup-instance, boolean-benchmarks ]
runs-on: ubuntu-latest
steps:
- name: Stop instance
id: stop-instance
uses: zama-ai/slab-github-runner@98f0788261a7323d5d695a883e20df36591a92b7
with:
mode: stop
github-token: ${{ secrets.SLAB_ACTION_TOKEN }}
slab-url: ${{ secrets.SLAB_BASE_URL }}
job-secret: ${{ secrets.JOB_SECRET }}
label: ${{ needs.setup-instance.outputs.runner-name }}
- name: Slack Notification
if: ${{ failure() }}
continue-on-error: true
uses: rtCamp/action-slack-notify@c33737706dea87cd7784c687dadc9adf1be59990
env:
SLACK_COLOR: ${{ job.status }}
SLACK_MESSAGE: "Instance teardown (boolean-benchmarks) finished with status: ${{ job.status }}. (${{ env.ACTION_RUN_URL }})"

View File

@@ -0,0 +1,132 @@
# Run core crypto benchmarks on an AWS instance and return parsed results to Slab CI bot.
name: Core crypto benchmarks
on:
workflow_dispatch:
env:
CARGO_TERM_COLOR: always
RESULTS_FILENAME: parsed_benchmark_results_${{ github.sha }}.json
ACTION_RUN_URL: ${{ github.server_url }}/${{ github.repository }}/actions/runs/${{ github.run_id }}
RUST_BACKTRACE: "full"
RUST_MIN_STACK: "8388608"
SLACK_CHANNEL: ${{ secrets.SLACK_CHANNEL }}
SLACK_ICON: https://pbs.twimg.com/profile_images/1274014582265298945/OjBKP9kn_400x400.png
SLACK_USERNAME: ${{ secrets.BOT_USERNAME }}
SLACK_WEBHOOK: ${{ secrets.SLACK_WEBHOOK }}
jobs:
setup-instance:
name: Setup instance (core-crypto-benchmarks)
runs-on: ubuntu-latest
if: github.event_name != 'schedule' ||
(github.event_name == 'schedule' && github.repository == 'zama-ai/tfhe-rs')
outputs:
runner-name: ${{ steps.start-instance.outputs.label }}
steps:
- name: Start instance
id: start-instance
uses: zama-ai/slab-github-runner@98f0788261a7323d5d695a883e20df36591a92b7
with:
mode: start
github-token: ${{ secrets.SLAB_ACTION_TOKEN }}
slab-url: ${{ secrets.SLAB_BASE_URL }}
job-secret: ${{ secrets.JOB_SECRET }}
backend: aws
profile: bench
core-crypto-benchmarks:
name: Execute core crypto benchmarks in EC2
needs: setup-instance
runs-on: ${{ needs.setup-instance.outputs.runner-name }}
concurrency:
group: ${{ github.workflow }}_${{ github.ref }}
cancel-in-progress: ${{ github.ref != 'refs/heads/main' }}
steps:
- name: Checkout tfhe-rs repo with tags
uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683
with:
fetch-depth: 0
token: ${{ secrets.FHE_ACTIONS_TOKEN }}
- name: Get benchmark details
run: |
{
echo "BENCH_DATE=$(date --iso-8601=seconds)";
echo "COMMIT_DATE=$(git --no-pager show -s --format=%cd --date=iso8601-strict ${{ github.sha }})";
echo "COMMIT_HASH=$(git describe --tags --dirty)";
} >> "${GITHUB_ENV}"
- name: Install rust
uses: dtolnay/rust-toolchain@315e265cd78dad1e1dcf3a5074f6d6c47029d5aa
with:
toolchain: nightly
- name: Run benchmarks with AVX512
run: |
make bench_pbs
make bench_pbs128
make bench_ks
- name: Parse results
run: |
python3 ./ci/benchmark_parser.py target/criterion ${{ env.RESULTS_FILENAME }} \
--database tfhe_rs \
--hardware "hpc7a.96xlarge" \
--project-version "${{ env.COMMIT_HASH }}" \
--branch ${{ github.ref_name }} \
--commit-date "${{ env.COMMIT_DATE }}" \
--bench-date "${{ env.BENCH_DATE }}" \
--name-suffix avx512 \
--walk-subdirs
- name: Upload parsed results artifact
uses: actions/upload-artifact@b4b15b8c7c6ac21ea08fcf65892d2ee8f75cf882
with:
name: ${{ github.sha }}_core_crypto
path: ${{ env.RESULTS_FILENAME }}
- name: Checkout Slab repo
uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683
with:
repository: zama-ai/slab
path: slab
token: ${{ secrets.FHE_ACTIONS_TOKEN }}
- name: Send data to Slab
shell: bash
run: |
python3 slab/scripts/data_sender.py ${{ env.RESULTS_FILENAME }} "${{ secrets.JOB_SECRET }}" \
--slab-url "${{ secrets.SLAB_URL }}"
- name: Slack Notification
if: ${{ failure() }}
continue-on-error: true
uses: rtCamp/action-slack-notify@c33737706dea87cd7784c687dadc9adf1be59990
env:
SLACK_COLOR: ${{ job.status }}
SLACK_MESSAGE: "PBS benchmarks finished with status: ${{ job.status }}. (${{ env.ACTION_RUN_URL }})"
teardown-instance:
name: Teardown instance (core-crypto-benchmarks)
if: ${{ always() && needs.setup-instance.result != 'skipped' }}
needs: [ setup-instance, core-crypto-benchmarks ]
runs-on: ubuntu-latest
steps:
- name: Stop instance
id: stop-instance
uses: zama-ai/slab-github-runner@98f0788261a7323d5d695a883e20df36591a92b7
with:
mode: stop
github-token: ${{ secrets.SLAB_ACTION_TOKEN }}
slab-url: ${{ secrets.SLAB_BASE_URL }}
job-secret: ${{ secrets.JOB_SECRET }}
label: ${{ needs.setup-instance.outputs.runner-name }}
- name: Slack Notification
if: ${{ failure() }}
continue-on-error: true
uses: rtCamp/action-slack-notify@c33737706dea87cd7784c687dadc9adf1be59990
env:
SLACK_COLOR: ${{ job.status }}
SLACK_MESSAGE: "Instance teardown (core-crypto-benchmarks) finished with status: ${{ job.status }}. (${{ env.ACTION_RUN_URL }})"

View File

@@ -1,130 +0,0 @@
# Run benchmarks on an AWS instance and return parsed results to Slab CI bot.
name: benchmark_cpu
run-name: ${{ inputs.command }}::${{ inputs.bench_type}} (${{ inputs.op_flavor }}, ${{ inputs.precisions_set }}, ${{ inputs.params_type }})
on:
workflow_dispatch:
inputs:
command:
description: "Benchmark command to run"
type: choice
options:
- integer
- signed_integer
- integer_compression
- integer_zk
- msm_zk
- shortint
- shortint_oprf
- hlapi_unsigned
- hlapi_signed
- hlapi_erc7984
- hlapi_dex
- hlapi_noise_squash
- hlapi_kvstore
- tfhe_zk_pok
- boolean
- pbs
- pbs128
- ks
- ks_pbs
op_flavor:
description: "Operations set to run"
type: choice
default: default
options:
- default
- fast_default
- smart
- unchecked
- misc
precisions_set:
description: "Bit precisions set"
type: choice
default: fast
options:
- fast
- all
- documentation
bench_type:
description: "Benchmarks type"
type: choice
default: latency
options:
- latency
- throughput
- both
params_type:
description: "Parameters type"
type: choice
default: classical
options:
- classical
- multi_bit
- classical + multi_bit
- classical_documentation
- multi_bit_documentation
- classical_documentation + multi_bit_documentation
permissions: {}
# zizmor: ignore[concurrency-limits] only Zama organization members can trigger this workflow
jobs:
parse-inputs:
name: benchmark_cpu/parse-inputs
runs-on: ubuntu-latest
outputs:
additional_file_to_parse: ${{ steps.set_file_to_parse.outputs.additional_file_to_parse }}
steps:
- name: Get additional file to parse
id: set_file_to_parse
shell: python
env:
INPUTS_COMMAND: ${{ inputs.command }}
run: |
import os
inputs_command = os.environ["INPUTS_COMMAND"]
output_file = os.environ["GITHUB_OUTPUT"]
files_to_parse = []
if inputs_command == "integer_zk":
files_to_parse.append("pke_zk_crs_sizes.csv")
elif inputs_command == "hlapi_erc7984":
files_to_parse.append("erc7984_pbs_count.csv")
elif inputs_command == "hlapi_dex":
files_to_parse.extend(
[
"dex_swap_request_update_dex_balance_pbs_count.csv",
"dex_swap_request_finalize_pbs_count.csv",
"dex_swap_claim_prepare_pbs_count.csv",
"dex_swap_claim_update_dex_balance_pbs_count.csv"
]
)
with open(output_file, "a") as f:
f.write(f"""additional_file_to_parse={",".join(files_to_parse)}\n""")
run-benchmarks:
name: benchmark_cpu/run-benchmarks
needs: parse-inputs
uses: ./.github/workflows/benchmark_cpu_common.yml
with:
command: ${{ inputs.command }}
op_flavor: ${{ inputs.op_flavor }}
bench_type: ${{ inputs.bench_type }}
params_type: ${{ inputs.params_type }}
precisions_set: ${{ inputs.precisions_set }}
additional_file_to_parse: ${{ needs.parse-inputs.outputs.additional_file_to_parse }}
secrets:
BOT_USERNAME: ${{ secrets.BOT_USERNAME }}
SLACK_CHANNEL: ${{ secrets.SLACK_CHANNEL }}
SLACK_WEBHOOK: ${{ secrets.SLACK_WEBHOOK }}
REPO_CHECKOUT_TOKEN: ${{ secrets.REPO_CHECKOUT_TOKEN }}
JOB_SECRET: ${{ secrets.JOB_SECRET }}
SLAB_ACTION_TOKEN: ${{ secrets.SLAB_ACTION_TOKEN }}
SLAB_URL: ${{ secrets.SLAB_URL }}
SLAB_BASE_URL: ${{ secrets.SLAB_BASE_URL }}

View File

@@ -1,277 +0,0 @@
# Run benchmarks on an instance and return parsed results to Slab CI bot.
name: benchmark_cpu_common
on:
workflow_call:
inputs:
command: # Any make recipes stripped of the "bench_" prefix in the Makefile
type: string # Use comma separated values to generate an array
required: true
op_flavor:
type: string # Use comma separated values to generate an array
default: default
bench_type:
type: string
default: latency
params_type:
type: string
default: classical
precisions_set:
type: string
default: fast
additional_recipe: # Make recipes to run aside the benchmarks.
type: string # Use comma separated values to generate an array
additional_file_to_parse: # Other files to parse, located under tfhe-benchmark/ directory
type: string # Use comma separated values to generate an array
additional_results_type:
type: string
default: object-size
secrets:
REPO_CHECKOUT_TOKEN:
required: true
SLAB_ACTION_TOKEN:
required: true
SLAB_BASE_URL:
required: true
SLAB_URL:
required: true
JOB_SECRET:
required: true
SLACK_CHANNEL:
required: true
BOT_USERNAME:
required: true
SLACK_WEBHOOK:
required: true
env:
CARGO_TERM_COLOR: always
RESULTS_FILENAME: parsed_benchmark_results_${{ github.sha }}.json
ACTION_RUN_URL: ${{ github.server_url }}/${{ github.repository }}/actions/runs/${{ github.run_id }}
RUST_BACKTRACE: "full"
RUST_MIN_STACK: "8388608"
SLACK_CHANNEL: ${{ secrets.SLACK_CHANNEL }}
SLACK_ICON: https://pbs.twimg.com/profile_images/1274014582265298945/OjBKP9kn_400x400.png
SLACK_USERNAME: ${{ secrets.BOT_USERNAME }}
SLACK_WEBHOOK: ${{ secrets.SLACK_WEBHOOK }}
permissions: {}
# zizmor: ignore[concurrency-limits] caller workflow is responsible for the concurrency
jobs:
prepare-matrix:
name: benchmark_cpu_common/prepare-matrix
runs-on: ubuntu-latest
outputs:
command: ${{ steps.set_matrix_args.outputs.command }}
op_flavor: ${{ steps.set_matrix_args.outputs.op_flavor }}
bench_type: ${{ steps.set_matrix_args.outputs.bench_type }}
params_type: ${{ steps.set_matrix_args.outputs.params_type }}
steps:
- name: Parse user inputs
shell: python
env:
INPUTS_COMMAND: ${{ inputs.command }}
INPUTS_OP_FLAVOR: ${{ inputs.op_flavor }}
INPUTS_BENCH_TYPE: ${{ inputs.bench_type }}
INPUTS_PARAMS_TYPE: ${{ inputs.params_type }}
run: |
import os
inputs_command = os.environ["INPUTS_COMMAND"]
inputs_op_flavor = os.environ["INPUTS_OP_FLAVOR"]
inputs_bench_type = os.environ["INPUTS_BENCH_TYPE"]
inputs_params_type = os.environ["INPUTS_PARAMS_TYPE"]
env_file = os.environ["GITHUB_ENV"]
split_command = inputs_command.replace(" ", "").split(",")
split_op_flavor = inputs_op_flavor.replace(" ", "").split(",")
if inputs_bench_type == "both":
bench_type = ["latency", "throughput"]
else:
bench_type = [inputs_bench_type, ]
if "+" in inputs_params_type:
split_params_type= inputs_params_type.replace(" ", "").split("+")
else:
split_params_type = [inputs_params_type, ]
with open(env_file, "a") as f:
for env_name, values_to_join in [
("COMMAND", split_command),
("OP_FLAVOR", split_op_flavor),
("BENCH_TYPE", bench_type),
("PARAMS_TYPE", split_params_type),
]:
f.write(f"""{env_name}=["{'", "'.join(values_to_join)}"]\n""")
- name: Set matrix arguments outputs
id: set_matrix_args
run: | # zizmor: ignore[template-injection] these env variable are safe
{
echo "command=${{ toJSON(env.COMMAND) }}";
echo "op_flavor=${{ toJSON(env.OP_FLAVOR) }}";
echo "bench_type=${{ toJSON(env.BENCH_TYPE) }}";
echo "params_type=${{ toJSON(env.PARAMS_TYPE) }}";
} >> "${GITHUB_OUTPUT}"
setup-instance:
name: benchmark_cpu_common/setup-instance
needs: prepare-matrix
runs-on: ubuntu-latest
outputs:
runner-name: ${{ steps.start-instance.outputs.label }}
steps:
- name: Start instance
id: start-instance
uses: zama-ai/slab-github-runner@5aee5d157f4a0201e5eaefc9cc648e5f9f5472a5 # v1.6.0
with:
mode: start
github-token: ${{ secrets.SLAB_ACTION_TOKEN }}
slab-url: ${{ secrets.SLAB_BASE_URL }}
job-secret: ${{ secrets.JOB_SECRET }}
backend: aws
profile: bench
integer-benchmarks:
name: benchmark_cpu_common/integer-benchmarks
needs: [ prepare-matrix, setup-instance ]
runs-on: ${{ needs.setup-instance.outputs.runner-name }}
timeout-minutes: 1440 # 24 hours
strategy:
max-parallel: 1
matrix:
command: ${{ fromJSON(needs.prepare-matrix.outputs.command) }}
op_flavor: ${{ fromJSON(needs.prepare-matrix.outputs.op_flavor) }}
bench_type: ${{ fromJSON(needs.prepare-matrix.outputs.bench_type) }}
params_type: ${{ fromJSON(needs.prepare-matrix.outputs.params_type) }}
steps:
- name: Checkout tfhe-rs repo with tags
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd
with:
fetch-depth: 0
persist-credentials: 'false'
token: ${{ secrets.REPO_CHECKOUT_TOKEN }}
- name: Get benchmark details
run: |
COMMIT_DATE=$(git --no-pager show -s --format=%cd --date=iso8601-strict "${SHA}");
{
echo "BENCH_DATE=$(date --iso-8601=seconds)";
echo "COMMIT_DATE=${COMMIT_DATE}";
echo "COMMIT_HASH=$(git describe --tags --dirty)";
} >> "${GITHUB_ENV}"
env:
SHA: ${{ github.sha }}
- name: Install rust
uses: dtolnay/rust-toolchain@e97e2d8cc328f1b50210efc529dca0028893a2d9 # zizmor: ignore[stale-action-refs] this action doesn't create releases
with:
toolchain: nightly
- name: Run benchmarks with AVX512
run: |
make BIT_SIZES_SET="${PRECISIONS_SET}" BENCH_OP_FLAVOR="${OP_FLAVOR}" BENCH_TYPE="${BENCH_TYPE}" BENCH_PARAM_TYPE="${BENCH_PARAMS_TYPE}" bench_"${BENCH_COMMAND}"
env:
OP_FLAVOR: ${{ matrix.op_flavor }}
BENCH_TYPE: ${{ matrix.bench_type }}
BENCH_PARAMS_TYPE: ${{ matrix.params_type }}
BENCH_COMMAND: ${{ matrix.command }}
PRECISIONS_SET: ${{ inputs.precisions_set }}
- name: Parse results
run: |
python3 ./ci/benchmark_parser.py target/criterion "${RESULTS_FILENAME}" \
--database tfhe_rs \
--hardware "hpc7a.96xlarge" \
--project-version "${COMMIT_HASH}" \
--branch "${REF_NAME}" \
--commit-date "${COMMIT_DATE}" \
--bench-date "${BENCH_DATE}" \
--walk-subdirs \
--name-suffix avx512 \
--bench-type "${BENCH_TYPE}"
env:
REF_NAME: ${{ github.ref_name }}
BENCH_TYPE: ${{ matrix.bench_type }}
- name: Run additional benchmarks
if: ${{ inputs.additional_recipe }}
run: |
targets_list="${targets}"
IFS=','
for target in $targets_list; do
make "$target"
done
env:
targets: ${{ inputs.additional_recipe }}
- name: Parse additional benchmarks results files
if: ${{ inputs.additional_file_to_parse }}
run: |
filenames_list="${filenames}"
IFS=','
for filename in $filenames_list; do
python3 ./ci/benchmark_parser.py "tfhe-benchmark/${filename}" "${RESULTS_FILENAME}" \
--"${results_type}" \
--append-results
done
env:
filenames: ${{ inputs.additional_file_to_parse }}
results_type: ${{ inputs.additional_results_type }}
- name: Upload parsed results artifact
uses: actions/upload-artifact@043fb46d1a93c77aae656e7c1c64a875d1fc6a0a
with:
name: ${{ github.sha }}_${{ matrix.command }}_${{ matrix.op_flavor }}_${{ matrix.bench_type }}_${{ matrix.params_type }}
path: ${{ env.RESULTS_FILENAME }}
- name: Checkout Slab repo
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd
with:
repository: zama-ai/slab
path: slab
persist-credentials: 'false'
token: ${{ secrets.REPO_CHECKOUT_TOKEN }}
- name: Send data to Slab
shell: bash
run: |
python3 slab/scripts/data_sender.py "${RESULTS_FILENAME}" "${JOB_SECRET}" \
--slab-url "${SLAB_URL}"
env:
JOB_SECRET: ${{ secrets.JOB_SECRET }}
SLAB_URL: ${{ secrets.SLAB_URL }}
- name: Slack Notification
if: ${{ failure() }}
continue-on-error: true
uses: rtCamp/action-slack-notify@e31e87e03dd19038e411e38ae27cbad084a90661
env:
SLACK_COLOR: ${{ job.status }}
SLACK_MESSAGE: "CPU benchmarks finished with status: ${{ job.status }}. (${{ env.ACTION_RUN_URL }})"
teardown-instance:
name: benchmark_cpu_common/teardown-instance
if: ${{ always() && needs.setup-instance.result == 'success' }}
needs: [ setup-instance, integer-benchmarks ]
runs-on: ubuntu-latest
steps:
- name: Stop instance
id: stop-instance
uses: zama-ai/slab-github-runner@5aee5d157f4a0201e5eaefc9cc648e5f9f5472a5 # v1.6.0
with:
mode: stop
github-token: ${{ secrets.SLAB_ACTION_TOKEN }}
slab-url: ${{ secrets.SLAB_BASE_URL }}
job-secret: ${{ secrets.JOB_SECRET }}
label: ${{ needs.setup-instance.outputs.runner-name }}
- name: Slack Notification
if: ${{ failure() }}
uses: rtCamp/action-slack-notify@e31e87e03dd19038e411e38ae27cbad084a90661
env:
SLACK_COLOR: ${{ job.status }}
SLACK_MESSAGE: "Instance teardown (cpu-benchmarks) finished with status: ${{ job.status }}. (${{ env.ACTION_RUN_URL }})"

View File

@@ -1,216 +0,0 @@
# Run CPU latencies benchmarks AWS VMs and return parsed results to Slab CI bot.
name: benchmark_cpu_weekly
on:
schedule:
# Weekly schedules are separated in two groups to avoid spawning too many the machines at once thus risking resource shortages.
# Group 1
# -------
# Weekly benchmarks will be triggered each Saturday at 1a.m.
- cron: '0 1 * * 6'
# Group 2
# -------
# Weekly benchmarks will be triggered each Sunday at 3a.m.
- cron: '0 3 * * 0'
# Quarterly benchmarks will be triggered right before the end of the quarter, the 25th of the current month at 4a.m.
# These benchmarks are far longer to execute, hence the reason to run them only four times a year.
- cron: '0 4 25 MAR,JUN,SEP,DEC *'
permissions: {}
# zizmor: ignore[concurrency-limits] only GitHub can trigger this workflow
jobs:
prepare-inputs:
name: benchmark_cpu_weekly/prepare-inputs
if: github.repository == 'zama-ai/tfhe-rs'
runs-on: ubuntu-latest
outputs:
is_weekly_bench_group_1: ${{ steps.check_bench_group_1.outputs.is_weekly_bench_group_1 }}
is_weekly_bench_group_2: ${{ steps.check_bench_group_2.outputs.is_weekly_bench_group_2 }}
is_quarterly_bench: ${{ steps.check_quarterly_bench.outputs.is_quarterly_bench }}
op_flavor: ${{ steps.set_op_flavor.outputs.op_flavor }}
precisions_set: ${{ steps.set_precisions_set.outputs.precisions_set }}
steps:
- name: Check is weekly bench group 1
id: check_bench_group_1
run: | # zizmor: ignore[template-injection] this env variable is safe
echo "is_weekly_bench_group_1=${{ github.event.schedule == '0 1 * * 6' }}" >> "${GITHUB_OUTPUT}"
- name: Check is weekly bench group 2
id: check_bench_group_2
run: | # zizmor: ignore[template-injection] this env variable is safe
echo "is_weekly_bench_group_2=${{ github.event.schedule == '0 3 * * 0' }}" >> "${GITHUB_OUTPUT}"
- name: Check is quarterly bench
id: check_quarterly_bench
run: | # zizmor: ignore[template-injection] this env variable is safe
echo "is_quarterly_bench=${{ github.event.schedule == '0 4 25 MAR,JUN,SEP,DEC *' }}" >> "${GITHUB_OUTPUT}"
- name: Weekly benchmarks
if: steps.check_bench_group_1.outputs.is_weekly_bench_group_1 == 'true' ||
steps.check_bench_group_2.outputs.is_weekly_bench_group_2 == 'true'
run: |
echo "OP_FLAVOR=default" >> "${GITHUB_ENV}"
echo "PRECISIONS_SET=fast" >> "${GITHUB_ENV}"
- name: Quarterly benchmarks
if: steps.check_quarterly_bench.outputs.is_quarterly_bench == 'true'
run: |
echo "OP_FLAVOR=\"default,unchecked\"" >> "${GITHUB_ENV}"
echo "PRECISIONS_SET=all" >> "${GITHUB_ENV}"
- name: Set operation flavor output
id: set_op_flavor
run: | # zizmor: ignore[template-injection] this env variable is safe
echo "op_flavor=${{ env.OP_FLAVOR }}" >> "${GITHUB_OUTPUT}"
- name: Set bit precisions output
id: set_precisions_set
run: | # zizmor: ignore[template-injection] this env variable is safe
echo "precisions_set=${{ env.PRECISIONS_SET }}" >> "${GITHUB_OUTPUT}"
run-benchmarks-integer:
name: benchmark_cpu_weekly/run-benchmarks-integer
if: needs.prepare-inputs.outputs.is_weekly_bench_group_1 == 'true' || needs.prepare-inputs.outputs.is_quarterly_bench == 'true'
needs: prepare-inputs
uses: ./.github/workflows/benchmark_cpu_common.yml
with:
command: integer,signed_integer, integer_compression
op_flavor: ${{ needs.prepare-inputs.outputs.op_flavor }}
precisions_set: ${{ needs.prepare-inputs.outputs.precisions_set }}
secrets:
BOT_USERNAME: ${{ secrets.BOT_USERNAME }}
SLACK_CHANNEL: ${{ secrets.SLACK_CHANNEL }}
SLACK_WEBHOOK: ${{ secrets.SLACK_WEBHOOK }}
REPO_CHECKOUT_TOKEN: ${{ secrets.REPO_CHECKOUT_TOKEN }}
JOB_SECRET: ${{ secrets.JOB_SECRET }}
SLAB_ACTION_TOKEN: ${{ secrets.SLAB_ACTION_TOKEN }}
SLAB_URL: ${{ secrets.SLAB_URL }}
SLAB_BASE_URL: ${{ secrets.SLAB_BASE_URL }}
run-benchmarks-integer-zk-pke:
name: benchmark_cpu_weekly/run-benchmarks-integer-zk-pke
if: needs.prepare-inputs.outputs.is_weekly_bench_group_1 == 'true'
needs: prepare-inputs
uses: ./.github/workflows/benchmark_cpu_common.yml
with:
command: integer_zk
additional_file_to_parse: pke_zk_crs_sizes.csv
secrets:
BOT_USERNAME: ${{ secrets.BOT_USERNAME }}
SLACK_CHANNEL: ${{ secrets.SLACK_CHANNEL }}
SLACK_WEBHOOK: ${{ secrets.SLACK_WEBHOOK }}
REPO_CHECKOUT_TOKEN: ${{ secrets.REPO_CHECKOUT_TOKEN }}
JOB_SECRET: ${{ secrets.JOB_SECRET }}
SLAB_ACTION_TOKEN: ${{ secrets.SLAB_ACTION_TOKEN }}
SLAB_URL: ${{ secrets.SLAB_URL }}
SLAB_BASE_URL: ${{ secrets.SLAB_BASE_URL }}
run-benchmarks-hlapi-erc7984:
name: benchmark_cpu_weekly/run-benchmarks-hlapi-erc7984
if: needs.prepare-inputs.outputs.is_weekly_bench_group_2 == 'true'
needs: prepare-inputs
uses: ./.github/workflows/benchmark_cpu_common.yml
with:
command: hlapi_erc7984
additional_file_to_parse: erc7984_pbs_count.csv
secrets:
BOT_USERNAME: ${{ secrets.BOT_USERNAME }}
SLACK_CHANNEL: ${{ secrets.SLACK_CHANNEL }}
SLACK_WEBHOOK: ${{ secrets.SLACK_WEBHOOK }}
REPO_CHECKOUT_TOKEN: ${{ secrets.REPO_CHECKOUT_TOKEN }}
JOB_SECRET: ${{ secrets.JOB_SECRET }}
SLAB_ACTION_TOKEN: ${{ secrets.SLAB_ACTION_TOKEN }}
SLAB_URL: ${{ secrets.SLAB_URL }}
SLAB_BASE_URL: ${{ secrets.SLAB_BASE_URL }}
run-benchmarks-hlapi-dex:
name: benchmark_cpu_weekly/run-benchmarks-hlapi-dex
if: needs.prepare-inputs.outputs.is_weekly_bench_group_1 == 'true'
needs: prepare-inputs
uses: ./.github/workflows/benchmark_cpu_common.yml
with:
command: hlapi_dex
additional_file_to_parse: dex_swap_request_update_dex_balance_pbs_count.csv,dex_swap_request_finalize_pbs_count.csv,dex_swap_claim_prepare_pbs_count.csv,dex_swap_claim_update_dex_balance_pbs_count.csv
secrets:
BOT_USERNAME: ${{ secrets.BOT_USERNAME }}
SLACK_CHANNEL: ${{ secrets.SLACK_CHANNEL }}
SLACK_WEBHOOK: ${{ secrets.SLACK_WEBHOOK }}
REPO_CHECKOUT_TOKEN: ${{ secrets.REPO_CHECKOUT_TOKEN }}
JOB_SECRET: ${{ secrets.JOB_SECRET }}
SLAB_ACTION_TOKEN: ${{ secrets.SLAB_ACTION_TOKEN }}
SLAB_URL: ${{ secrets.SLAB_URL }}
SLAB_BASE_URL: ${{ secrets.SLAB_BASE_URL }}
run-benchmarks-core-crypto:
name: benchmark_cpu_weekly/run-benchmarks-core-crypto
if: needs.prepare-inputs.outputs.is_weekly_bench_group_1 == 'true'
needs: prepare-inputs
uses: ./.github/workflows/benchmark_cpu_common.yml
with:
command: ks,pbs,pbs128,ks_pbs
secrets:
BOT_USERNAME: ${{ secrets.BOT_USERNAME }}
SLACK_CHANNEL: ${{ secrets.SLACK_CHANNEL }}
SLACK_WEBHOOK: ${{ secrets.SLACK_WEBHOOK }}
REPO_CHECKOUT_TOKEN: ${{ secrets.REPO_CHECKOUT_TOKEN }}
JOB_SECRET: ${{ secrets.JOB_SECRET }}
SLAB_ACTION_TOKEN: ${{ secrets.SLAB_ACTION_TOKEN }}
SLAB_URL: ${{ secrets.SLAB_URL }}
SLAB_BASE_URL: ${{ secrets.SLAB_BASE_URL }}
run-benchmarks-shortint:
name: benchmark_cpu_weekly/run-benchmarks-shortint
if: needs.prepare-inputs.outputs.is_weekly_bench_group_2 == 'true' || needs.prepare-inputs.outputs.is_quarterly_bench == 'true'
needs: prepare-inputs
uses: ./.github/workflows/benchmark_cpu_common.yml
with:
op_flavor: ${{ needs.prepare-inputs.outputs.op_flavor }}
command: shortint
secrets:
BOT_USERNAME: ${{ secrets.BOT_USERNAME }}
SLACK_CHANNEL: ${{ secrets.SLACK_CHANNEL }}
SLACK_WEBHOOK: ${{ secrets.SLACK_WEBHOOK }}
REPO_CHECKOUT_TOKEN: ${{ secrets.REPO_CHECKOUT_TOKEN }}
JOB_SECRET: ${{ secrets.JOB_SECRET }}
SLAB_ACTION_TOKEN: ${{ secrets.SLAB_ACTION_TOKEN }}
SLAB_URL: ${{ secrets.SLAB_URL }}
SLAB_BASE_URL: ${{ secrets.SLAB_BASE_URL }}
run-benchmarks-boolean:
name: benchmark_cpu_weekly/run-benchmarks-boolean
if: needs.prepare-inputs.outputs.is_weekly_bench_group_2 == 'true'
needs: prepare-inputs
uses: ./.github/workflows/benchmark_cpu_common.yml
with:
command: boolean
additional_recipe: measure_boolean_key_sizes
additional_file_to_parse: boolean_key_sizes.csv
secrets:
BOT_USERNAME: ${{ secrets.BOT_USERNAME }}
SLACK_CHANNEL: ${{ secrets.SLACK_CHANNEL }}
SLACK_WEBHOOK: ${{ secrets.SLACK_WEBHOOK }}
REPO_CHECKOUT_TOKEN: ${{ secrets.REPO_CHECKOUT_TOKEN }}
JOB_SECRET: ${{ secrets.JOB_SECRET }}
SLAB_ACTION_TOKEN: ${{ secrets.SLAB_ACTION_TOKEN }}
SLAB_URL: ${{ secrets.SLAB_URL }}
SLAB_BASE_URL: ${{ secrets.SLAB_BASE_URL }}
run-benchmarks-tfhe-zk-pok:
name: benchmark_cpu_weekly/run-benchmarks-tfhe-zk-pok
if: needs.prepare-inputs.outputs.is_weekly_bench_group_1 == 'true'
needs: prepare-inputs
uses: ./.github/workflows/benchmark_cpu_common.yml
with:
command: tfhe_zk_pok
secrets:
BOT_USERNAME: ${{ secrets.BOT_USERNAME }}
SLACK_CHANNEL: ${{ secrets.SLACK_CHANNEL }}
SLACK_WEBHOOK: ${{ secrets.SLACK_WEBHOOK }}
REPO_CHECKOUT_TOKEN: ${{ secrets.REPO_CHECKOUT_TOKEN }}
JOB_SECRET: ${{ secrets.JOB_SECRET }}
SLAB_ACTION_TOKEN: ${{ secrets.SLAB_ACTION_TOKEN }}
SLAB_URL: ${{ secrets.SLAB_URL }}
SLAB_BASE_URL: ${{ secrets.SLAB_BASE_URL }}

View File

@@ -1,153 +0,0 @@
# Run sizes benchmarks on an instance and return parsed results to Slab CI bot.
name: Ciphertext and Keys sizes benchmarks
on:
workflow_dispatch:
schedule:
# Monthly benchmarks will be triggered each 24th of the month at 1a.m.
- cron: '0 1 24 * 6'
env:
CARGO_TERM_COLOR: always
RESULTS_FILENAME: parsed_benchmark_results_${{ github.sha }}.json
ACTION_RUN_URL: ${{ github.server_url }}/${{ github.repository }}/actions/runs/${{ github.run_id }}
RUST_BACKTRACE: "full"
RUST_MIN_STACK: "8388608"
SLACK_CHANNEL: ${{ secrets.SLACK_CHANNEL }}
SLACK_ICON: https://pbs.twimg.com/profile_images/1274014582265298945/OjBKP9kn_400x400.png
SLACK_USERNAME: ${{ secrets.BOT_USERNAME }}
SLACK_WEBHOOK: ${{ secrets.SLACK_WEBHOOK }}
permissions: {}
# zizmor: ignore[concurrency-limits] only Zama organization members and GitHub can trigger this workflow
jobs:
setup-instance:
name: Setup instance (sizes-benchmarks)
if: github.event_name == 'workflow_dispatch' ||
(github.event_name == 'schedule' && github.repository == 'zama-ai/tfhe-rs')
runs-on: ubuntu-latest
outputs:
runner-name: ${{ steps.start-instance.outputs.label }}
steps:
- name: Start instance
id: start-instance
uses: zama-ai/slab-github-runner@5aee5d157f4a0201e5eaefc9cc648e5f9f5472a5 # v1.6.0
with:
mode: start
github-token: ${{ secrets.SLAB_ACTION_TOKEN }}
slab-url: ${{ secrets.SLAB_BASE_URL }}
job-secret: ${{ secrets.JOB_SECRET }}
backend: aws
profile: cpu-big
sizes-benchmarks:
name: Execute sizes client benchmarks
needs: setup-instance
if: needs.setup-instance.result != 'skipped'
runs-on: ${{ needs.setup-instance.outputs.runner-name }}
steps:
- name: Checkout tfhe-rs repo with tags
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd
with:
fetch-depth: 0
persist-credentials: 'false'
token: ${{ secrets.REPO_CHECKOUT_TOKEN }}
- name: Get benchmark details
run: |
COMMIT_DATE=$(git --no-pager show -s --format=%cd --date=iso8601-strict "${SHA}");
{
echo "BENCH_DATE=$(date --iso-8601=seconds)";
echo "COMMIT_DATE=${COMMIT_DATE}";
echo "COMMIT_HASH=$(git describe --tags --dirty)";
} >> "${GITHUB_ENV}"
env:
SHA: ${{ github.sha }}
- name: Install rust
uses: dtolnay/rust-toolchain@b3b07ba8b418998c39fb20f53e8b695cdcc8de1b # zizmor: ignore[stale-action-refs] this action doesn't create releases
with:
toolchain: nightly
- name: Measure public key and ciphertext sizes in HL Api
run: |
make measure_hlapi_compact_pk_ct_sizes
- name: Parse key and ciphertext sizes results
run: |
python3 ./ci/benchmark_parser.py tfhe-benchmark/hlapi_ct_key_sizes.csv "${RESULTS_FILENAME}" \
--database tfhe_rs \
--hardware "m6i.32xlarge" \
--project-version "${COMMIT_HASH}" \
--branch "${REF_NAME}" \
--commit-date "${COMMIT_DATE}" \
--bench-date "${BENCH_DATE}" \
--object-sizes
env:
REF_NAME: ${{ github.ref_name }}
- name: Measure key sizes in shortint
run: |
make measure_shortint_key_sizes
- name: Parse key sizes results
run: |
python3 ./ci/benchmark_parser.py tfhe-benchmark/shortint_key_sizes.csv "${RESULTS_FILENAME}" \
--object-sizes \
--append-results
- name: Upload parsed results artifact
uses: actions/upload-artifact@043fb46d1a93c77aae656e7c1c64a875d1fc6a0a
with:
name: ${{ github.sha }}_ct_key_sizes
path: ${{ env.RESULTS_FILENAME }}
- name: Checkout Slab repo
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd
with:
repository: zama-ai/slab
path: slab
persist-credentials: 'false'
token: ${{ secrets.REPO_CHECKOUT_TOKEN }}
- name: Send data to Slab
shell: bash
run: |
python3 slab/scripts/data_sender.py "${RESULTS_FILENAME}" "${JOB_SECRET}" \
--slab-url "${SLAB_URL}"
env:
JOB_SECRET: ${{ secrets.JOB_SECRET }}
SLAB_URL: ${{ secrets.SLAB_URL }}
- name: Slack Notification
if: ${{ failure() }}
continue-on-error: true
uses: rtCamp/action-slack-notify@e31e87e03dd19038e411e38ae27cbad084a90661
env:
SLACK_COLOR: ${{ job.status }}
SLACK_MESSAGE: "Sizes benchmarks finished with status: ${{ job.status }}. (${{ env.ACTION_RUN_URL }})"
teardown-instance:
name: Teardown instance (sizes-benchmarks)
if: ${{ always() && needs.setup-instance.result == 'success' }}
needs: [ setup-instance, sizes-benchmarks ]
runs-on: ubuntu-latest
steps:
- name: Stop instance
id: stop-instance
uses: zama-ai/slab-github-runner@5aee5d157f4a0201e5eaefc9cc648e5f9f5472a5 # v1.6.0
with:
mode: stop
github-token: ${{ secrets.SLAB_ACTION_TOKEN }}
slab-url: ${{ secrets.SLAB_BASE_URL }}
job-secret: ${{ secrets.JOB_SECRET }}
label: ${{ needs.setup-instance.outputs.runner-name }}
- name: Slack Notification
if: ${{ failure() }}
uses: rtCamp/action-slack-notify@e31e87e03dd19038e411e38ae27cbad084a90661
env:
SLACK_COLOR: ${{ job.status }}
SLACK_MESSAGE: "Instance teardown (sizes-benchmarks) finished with status: ${{ job.status }}. (${{ env.ACTION_RUN_URL }})"

View File

@@ -1,224 +0,0 @@
# Run all benchmarks displayed in the public documentation.
name: benchmark_documentation
on:
workflow_dispatch:
inputs:
run-cpu-benchmarks:
description: "Run CPU benchmarks"
type: boolean
default: true
# GPU benchmarks are split because of resource scarcity.
run-gpu-integer-benchmarks:
description: "Run GPU integer benchmarks"
type: boolean
default: true
run-gpu-core-crypto-benchmarks:
description: "Run GPU core-crypto benchmarks"
type: boolean
default: true
run-gpu-zk-benchmarks:
description: "Run GPU ZK benchmarks"
type: boolean
default: true
run-hpu-benchmarks:
description: "Run HPU benchmarks"
type: boolean
default: true
generate-svgs:
description: "Generate SVG tables"
type: boolean
default: true
permissions: {}
# zizmor: ignore[concurrency-limits] only Zama organization members can trigger this workflow
jobs:
run-benchmarks-cpu-integer:
name: benchmark_documentation/run-benchmarks-cpu-integer
uses: ./.github/workflows/benchmark_cpu_common.yml
if: inputs.run-cpu-benchmarks
with:
command: integer,hlapi_erc7984
op_flavor: fast_default
bench_type: both
precisions_set: documentation
secrets:
BOT_USERNAME: ${{ secrets.BOT_USERNAME }}
SLACK_CHANNEL: ${{ secrets.SLACK_CHANNEL }}
SLACK_WEBHOOK: ${{ secrets.SLACK_WEBHOOK }}
REPO_CHECKOUT_TOKEN: ${{ secrets.REPO_CHECKOUT_TOKEN }}
JOB_SECRET: ${{ secrets.JOB_SECRET }}
SLAB_ACTION_TOKEN: ${{ secrets.SLAB_ACTION_TOKEN }}
SLAB_URL: ${{ secrets.SLAB_URL }}
SLAB_BASE_URL: ${{ secrets.SLAB_BASE_URL }}
run-benchmarks-cpu-zk-server:
name: benchmark_documentation/run-benchmarks-cpu-zk-server
uses: ./.github/workflows/benchmark_cpu_common.yml
if: inputs.run-cpu-benchmarks
with:
command: integer_zk
op_flavor: default
bench_type: both
secrets:
BOT_USERNAME: ${{ secrets.BOT_USERNAME }}
SLACK_CHANNEL: ${{ secrets.SLACK_CHANNEL }}
SLACK_WEBHOOK: ${{ secrets.SLACK_WEBHOOK }}
REPO_CHECKOUT_TOKEN: ${{ secrets.REPO_CHECKOUT_TOKEN }}
JOB_SECRET: ${{ secrets.JOB_SECRET }}
SLAB_ACTION_TOKEN: ${{ secrets.SLAB_ACTION_TOKEN }}
SLAB_URL: ${{ secrets.SLAB_URL }}
SLAB_BASE_URL: ${{ secrets.SLAB_BASE_URL }}
run-benchmarks-cpu-zk-client:
name: benchmark_documentation/run-benchmarks-cpu-zk-client
uses: ./.github/workflows/benchmark_wasm_client_common.yml
if: inputs.run-cpu-benchmarks
with:
browser: chrome
secrets:
BOT_USERNAME: ${{ secrets.BOT_USERNAME }}
SLACK_CHANNEL: ${{ secrets.SLACK_CHANNEL }}
SLACK_WEBHOOK: ${{ secrets.SLACK_WEBHOOK }}
REPO_CHECKOUT_TOKEN: ${{ secrets.REPO_CHECKOUT_TOKEN }}
JOB_SECRET: ${{ secrets.JOB_SECRET }}
SLAB_ACTION_TOKEN: ${{ secrets.SLAB_ACTION_TOKEN }}
SLAB_URL: ${{ secrets.SLAB_URL }}
SLAB_BASE_URL: ${{ secrets.SLAB_BASE_URL }}
run-benchmarks-gpu-integer:
name: benchmark_documentation/run-benchmarks-gpu-integer
uses: ./.github/workflows/benchmark_gpu_common.yml
if: inputs.run-gpu-integer-benchmarks
with:
profile: multi-h100-sxm5
hardware_name: n3-H100-SXM5x8
command: integer_multi_bit,hlapi_erc7984
op_flavor: fast_default
bench_type: both
precisions_set: documentation
secrets:
BOT_USERNAME: ${{ secrets.BOT_USERNAME }}
SLACK_CHANNEL: ${{ secrets.SLACK_CHANNEL }}
SLACK_WEBHOOK: ${{ secrets.SLACK_WEBHOOK }}
REPO_CHECKOUT_TOKEN: ${{ secrets.REPO_CHECKOUT_TOKEN }}
JOB_SECRET: ${{ secrets.JOB_SECRET }}
SLAB_ACTION_TOKEN: ${{ secrets.SLAB_ACTION_TOKEN }}
SLAB_URL: ${{ secrets.SLAB_URL }}
SLAB_BASE_URL: ${{ secrets.SLAB_BASE_URL }}
run-benchmarks-hpu-integer:
name: benchmark_documentation/run-benchmarks-hpu-integer
uses: ./.github/workflows/benchmark_hpu_common.yml
if: inputs.run-hpu-benchmarks
with:
command: integer,hlapi_erc7984
op_flavor: default
bench_type: both
precisions_set: documentation
v80_pcie_dev: 24
v80_serial_number: XFL12NWY3ZKG
secrets:
BOT_USERNAME: ${{ secrets.BOT_USERNAME }}
SLACK_CHANNEL: ${{ secrets.SLACK_CHANNEL }}
SLACK_WEBHOOK: ${{ secrets.SLACK_WEBHOOK }}
REPO_CHECKOUT_TOKEN: ${{ secrets.REPO_CHECKOUT_TOKEN }}
JOB_SECRET: ${{ secrets.JOB_SECRET }}
SLAB_ACTION_TOKEN: ${{ secrets.SLAB_ACTION_TOKEN }}
SLAB_URL: ${{ secrets.SLAB_URL }}
SLAB_BASE_URL: ${{ secrets.SLAB_BASE_URL }}
SSH_PRIVATE_KEY: ${{ secrets.SSH_PRIVATE_KEY }}
run-benchmarks-cpu-core-crypto:
name: benchmark_documentation/run-benchmarks-cpu-core-crypto
uses: ./.github/workflows/benchmark_cpu_common.yml
if: inputs.run-cpu-benchmarks
with:
command: pbs, ks_pbs
bench_type: latency
params_type: classical_documentation + multi_bit_documentation
secrets:
BOT_USERNAME: ${{ secrets.BOT_USERNAME }}
SLACK_CHANNEL: ${{ secrets.SLACK_CHANNEL }}
SLACK_WEBHOOK: ${{ secrets.SLACK_WEBHOOK }}
REPO_CHECKOUT_TOKEN: ${{ secrets.REPO_CHECKOUT_TOKEN }}
JOB_SECRET: ${{ secrets.JOB_SECRET }}
SLAB_ACTION_TOKEN: ${{ secrets.SLAB_ACTION_TOKEN }}
SLAB_URL: ${{ secrets.SLAB_URL }}
SLAB_BASE_URL: ${{ secrets.SLAB_BASE_URL }}
run-benchmarks-gpu-core-crypto:
name: benchmark_documentation/run-benchmarks-gpu-core-crypto
uses: ./.github/workflows/benchmark_gpu_common.yml
if: inputs.run-gpu-core-crypto-benchmarks
with:
profile: multi-h100-sxm5
hardware_name: n3-H100-SXM5x8
command: pbs, ks_pbs
bench_type: latency
params_type: classical_documentation + multi_bit_documentation
secrets:
BOT_USERNAME: ${{ secrets.BOT_USERNAME }}
SLACK_CHANNEL: ${{ secrets.SLACK_CHANNEL }}
SLACK_WEBHOOK: ${{ secrets.SLACK_WEBHOOK }}
REPO_CHECKOUT_TOKEN: ${{ secrets.REPO_CHECKOUT_TOKEN }}
JOB_SECRET: ${{ secrets.JOB_SECRET }}
SLAB_ACTION_TOKEN: ${{ secrets.SLAB_ACTION_TOKEN }}
SLAB_URL: ${{ secrets.SLAB_URL }}
SLAB_BASE_URL: ${{ secrets.SLAB_BASE_URL }}
run-benchmarks-gpu-zk-server:
name: benchmark_documentation/run-benchmarks-gpu-zk-server
uses: ./.github/workflows/benchmark_gpu_common.yml
if: inputs.run-gpu-zk-benchmarks
with:
profile: multi-h100-sxm5
hardware_name: n3-H100-SXM5x8
command: integer_zk
op_flavor: default
bench_type: both
secrets:
BOT_USERNAME: ${{ secrets.BOT_USERNAME }}
SLACK_CHANNEL: ${{ secrets.SLACK_CHANNEL }}
SLACK_WEBHOOK: ${{ secrets.SLACK_WEBHOOK }}
REPO_CHECKOUT_TOKEN: ${{ secrets.REPO_CHECKOUT_TOKEN }}
JOB_SECRET: ${{ secrets.JOB_SECRET }}
SLAB_ACTION_TOKEN: ${{ secrets.SLAB_ACTION_TOKEN }}
SLAB_URL: ${{ secrets.SLAB_URL }}
SLAB_BASE_URL: ${{ secrets.SLAB_BASE_URL }}
generate-svgs-with-benchmarks-run:
name: benchmark-documentation/generate-svgs-with-benchmarks-run
if: ${{ always() &&
(inputs.run-cpu-benchmarks || inputs.run-gpu-integer-benchmarks || inputs.run-gpu-core-crypto-benchmarks || inputs.run-gpu-zk-benchmarks || inputs.run-hpu-benchmarks) &&
inputs.generate-svgs }}
needs: [
run-benchmarks-cpu-integer, run-benchmarks-gpu-integer, run-benchmarks-hpu-integer,
run-benchmarks-cpu-zk-server, run-benchmarks-cpu-zk-client,
run-benchmarks-cpu-core-crypto, run-benchmarks-gpu-core-crypto,
run-benchmarks-gpu-zk-server
]
uses: ./.github/workflows/generate_svgs.yml
with:
time_span_days: 5
generate-cpu-svgs: ${{ inputs.run-cpu-benchmarks }}
generate-gpu-svgs: ${{ inputs.run-gpu-integer-benchmarks || inputs.run-gpu-core-crypto-benchmarks || inputs.run-gpu-zk-benchmarks }}
generate-hpu-svgs: ${{ inputs.run-hpu-benchmarks }}
secrets:
DATA_EXTRACTOR_DATABASE_USER: ${{ secrets.DATA_EXTRACTOR_DATABASE_USER }}
DATA_EXTRACTOR_DATABASE_HOST: ${{ secrets.DATA_EXTRACTOR_DATABASE_HOST }}
DATA_EXTRACTOR_DATABASE_PASSWORD: ${{ secrets.DATA_EXTRACTOR_DATABASE_PASSWORD }}
generate-svgs-without-benchmarks-run:
name: benchmark-documentation/generate-svgs-without-benchmarks-run
if: ${{ !(inputs.run-cpu-benchmarks || inputs.run-gpu-integer-benchmarks || inputs.run-gpu-core-crypto-benchmarks || inputs.run-gpu-zk-benchmarks || inputs.run-hpu-benchmarks) &&
inputs.generate-svgs }}
uses: ./.github/workflows/generate_svgs.yml
with:
time_span_days: 60
secrets:
DATA_EXTRACTOR_DATABASE_USER: ${{ secrets.DATA_EXTRACTOR_DATABASE_USER }}
DATA_EXTRACTOR_DATABASE_HOST: ${{ secrets.DATA_EXTRACTOR_DATABASE_HOST }}
DATA_EXTRACTOR_DATABASE_PASSWORD: ${{ secrets.DATA_EXTRACTOR_DATABASE_PASSWORD }}

141
.github/workflows/benchmark_erc20.yml vendored Normal file
View File

@@ -0,0 +1,141 @@
# Run all ERC20 benchmarks on an AWS instance and return parsed results to Slab CI bot.
name: ERC20 benchmarks
on:
workflow_dispatch:
schedule:
# Weekly benchmarks will be triggered each Saturday at 5a.m.
- cron: '0 5 * * 6'
env:
CARGO_TERM_COLOR: always
RESULTS_FILENAME: parsed_benchmark_results_${{ github.sha }}.json
ACTION_RUN_URL: ${{ github.server_url }}/${{ github.repository }}/actions/runs/${{ github.run_id }}
RUST_BACKTRACE: "full"
RUST_MIN_STACK: "8388608"
SLACK_CHANNEL: ${{ secrets.SLACK_CHANNEL }}
SLACK_ICON: https://pbs.twimg.com/profile_images/1274014582265298945/OjBKP9kn_400x400.png
SLACK_USERNAME: ${{ secrets.BOT_USERNAME }}
SLACK_WEBHOOK: ${{ secrets.SLACK_WEBHOOK }}
jobs:
setup-instance:
name: Setup instance (erc20-benchmarks)
runs-on: ubuntu-latest
if: github.event_name == 'workflow_dispatch' ||
(github.event_name == 'schedule' && github.repository == 'zama-ai/tfhe-rs')
outputs:
runner-name: ${{ steps.start-instance.outputs.label }}
steps:
- name: Start instance
id: start-instance
uses: zama-ai/slab-github-runner@98f0788261a7323d5d695a883e20df36591a92b7
with:
mode: start
github-token: ${{ secrets.SLAB_ACTION_TOKEN }}
slab-url: ${{ secrets.SLAB_BASE_URL }}
job-secret: ${{ secrets.JOB_SECRET }}
backend: aws
profile: bench
erc20-benchmarks:
name: Execute ERC20 benchmarks
needs: setup-instance
runs-on: ${{ needs.setup-instance.outputs.runner-name }}
concurrency:
group: ${{ github.workflow }}_${{ github.ref }}
cancel-in-progress: ${{ github.ref != 'refs/heads/main' }}
continue-on-error: true
timeout-minutes: 720 # 12 hours
steps:
- name: Checkout tfhe-rs repo with tags
uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683
with:
fetch-depth: 0
token: ${{ secrets.FHE_ACTIONS_TOKEN }}
- name: Get benchmark details
run: |
{
echo "BENCH_DATE=$(date --iso-8601=seconds)";
echo "COMMIT_DATE=$(git --no-pager show -s --format=%cd --date=iso8601-strict ${{ github.sha }})";
echo "COMMIT_HASH=$(git describe --tags --dirty)";
} >> "${GITHUB_ENV}"
- name: Install rust
uses: dtolnay/rust-toolchain@315e265cd78dad1e1dcf3a5074f6d6c47029d5aa
with:
toolchain: nightly
- name: Checkout Slab repo
uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683
with:
repository: zama-ai/slab
path: slab
token: ${{ secrets.FHE_ACTIONS_TOKEN }}
- name: Run benchmarks
run: |
make bench_hlapi_erc20
- name: Parse results
run: |
python3 ./ci/benchmark_parser.py target/criterion ${{ env.RESULTS_FILENAME }} \
--database tfhe_rs \
--hardware "hpc7a.96xlarge" \
--project-version "${{ env.COMMIT_HASH }}" \
--branch ${{ github.ref_name }} \
--commit-date "${{ env.COMMIT_DATE }}" \
--bench-date "${{ env.BENCH_DATE }}" \
--walk-subdirs \
--name-suffix avx512
- name: Parse PBS counts
run: |
python3 ./ci/benchmark_parser.py tfhe/erc20_pbs_count.csv ${{ env.RESULTS_FILENAME }} \
--object-sizes \
--append-results
- name: Upload parsed results artifact
uses: actions/upload-artifact@b4b15b8c7c6ac21ea08fcf65892d2ee8f75cf882
with:
name: ${{ github.sha }}_erc20
path: ${{ env.RESULTS_FILENAME }}
- name: Send data to Slab
shell: bash
run: |
python3 slab/scripts/data_sender.py ${{ env.RESULTS_FILENAME }} "${{ secrets.JOB_SECRET }}" \
--slab-url "${{ secrets.SLAB_URL }}"
- name: Slack Notification
if: ${{ failure() }}
continue-on-error: true
uses: rtCamp/action-slack-notify@c33737706dea87cd7784c687dadc9adf1be59990
env:
SLACK_COLOR: ${{ job.status }}
SLACK_MESSAGE: "ERC20 benchmarks finished with status: ${{ job.status }}. (${{ env.ACTION_RUN_URL }})"
teardown-instance:
name: Teardown instance (erc20-benchmarks)
if: ${{ always() && needs.setup-instance.result != 'skipped' }}
needs: [ setup-instance, erc20-benchmarks ]
runs-on: ubuntu-latest
steps:
- name: Stop instance
id: stop-instance
uses: zama-ai/slab-github-runner@98f0788261a7323d5d695a883e20df36591a92b7
with:
mode: stop
github-token: ${{ secrets.SLAB_ACTION_TOKEN }}
slab-url: ${{ secrets.SLAB_BASE_URL }}
job-secret: ${{ secrets.JOB_SECRET }}
label: ${{ needs.setup-instance.outputs.runner-name }}
- name: Slack Notification
if: ${{ failure() }}
continue-on-error: true
uses: rtCamp/action-slack-notify@c33737706dea87cd7784c687dadc9adf1be59990
env:
SLACK_COLOR: ${{ job.status }}
SLACK_MESSAGE: "Instance teardown (erc20-benchmarks) finished with status: ${{ job.status }}. (${{ env.ACTION_RUN_URL }})"

View File

@@ -1,162 +0,0 @@
# Run CUDA benchmarks on a Hyperstack VM and return parsed results to Slab CI bot.
name: benchmark_gpu
run-name: ${{ inputs.command }}::${{ inputs.bench_type}} (${{ inputs.profile }}, ${{ inputs.op_flavor }}, ${{ inputs.precisions_set }}, ${{ inputs.params_type }})
on:
workflow_dispatch:
inputs:
profile:
description: "Instance type"
required: true
type: choice
options:
- "l40 (n3-L40x1)"
- "4-l40 (n3-L40x4)"
- "multi-a100-nvlink (n3-A100x8-NVLink)"
- "single-h100 (n3-H100x1)"
- "2-h100 (n3-H100x2)"
- "4-h100 (n3-H100x4)"
- "multi-h100 (n3-H100x8)"
- "multi-h100-nvlink (n3-H100x8-NVLink)"
- "multi-h100-sxm5 (n3-H100-SXM5x8)"
command:
description: "Benchmark command to run"
type: choice
default: integer
options:
- integer
- integer_compression
- pbs
- pbs128
- ks
- ks_pbs
- tfhe_zk_pok
- msm_zk
- integer_zk
- integer_zk_experimental
- integer_aes
- integer_aes256
- hlapi_erc7984
- hlapi_dex
- hlapi_noise_squash
op_flavor:
description: "Operations set to run"
type: choice
default: default
options:
- default
- fast_default
- unchecked
precisions_set:
description: "Bit precisions set"
type: choice
default: fast
options:
- fast
- all
- documentation
bench_type:
description: "Benchmarks type"
type: choice
default: latency
options:
- latency
- throughput
- both
params_type:
description: "Parameters type"
type: choice
default: multi_bit
options:
- classical
- multi_bit
- classical + multi_bit
- classical_documentation
- multi_bit_documentation
- classical_documentation + multi_bit_documentation
permissions: {}
# zizmor: ignore[concurrency-limits] only Zama organization members can trigger this workflow
jobs:
parse-inputs:
name: benchmark_gpu/parse-inputs
runs-on: ubuntu-latest
outputs:
profile: ${{ steps.parse_profile.outputs.profile }}
hardware_name: ${{ steps.parse_hardware_name.outputs.name }}
additional_file_to_parse: ${{ steps.set_file_to_parse.outputs.additional_file_to_parse }}
env:
INPUTS_PROFILE: ${{ inputs.profile }}
steps:
- name: Parse profile
id: parse_profile
run: |
# Use Sed to extract a value from a string, this cannot be done with the ${variable//search/replace} pattern.
# shellcheck disable=SC2001
PROFILE=$(echo "${INPUTS_PROFILE}" | sed 's|\(.*\)[[:space:]](.*)|\1|')
echo "profile=${PROFILE}" >> "${GITHUB_OUTPUT}"
- name: Parse hardware name
id: parse_hardware_name
run: |
# Use Sed to extract a value from a string, this cannot be done with the ${variable//search/replace} pattern.
# shellcheck disable=SC2001
NAME=$(echo "${INPUTS_PROFILE}" | sed 's|.*[[:space:]](\(.*\))|\1|')
echo "name=${NAME}" >> "${GITHUB_OUTPUT}"
- name: Get additional file to parse
id: set_file_to_parse
shell: python
env:
INPUTS_COMMAND: ${{ inputs.command }}
run: |
import os
inputs_command = os.environ["INPUTS_COMMAND"]
output_file = os.environ["GITHUB_OUTPUT"]
files_to_parse = []
if inputs_command == "integer_zk":
files_to_parse.append("pke_zk_crs_sizes.csv")
elif inputs_command == "hlapi_erc7984":
files_to_parse.append("erc7984_pbs_count.csv")
elif inputs_command == "hlapi_dex":
files_to_parse.extend(
[
"dex_swap_request_update_dex_balance_pbs_count.csv",
"dex_swap_request_finalize_pbs_count.csv",
"dex_swap_claim_prepare_pbs_count.csv",
"dex_swap_claim_update_dex_balance_pbs_count.csv"
]
)
with open(output_file, "a") as f:
f.write(f"""additional_file_to_parse={",".join(files_to_parse)}\n""")
run-benchmarks:
name: benchmark_gpu/run-benchmarks
needs: parse-inputs
uses: ./.github/workflows/benchmark_gpu_common.yml
with:
profile: ${{ needs.parse-inputs.outputs.profile }}
hardware_name: ${{ needs.parse-inputs.outputs.hardware_name }}
command: ${{ inputs.command }}
op_flavor: ${{ inputs.op_flavor }}
bench_type: ${{ inputs.bench_type }}
params_type: ${{ inputs.params_type }}
precisions_set: ${{ inputs.precisions_set }}
additional_file_to_parse: ${{ needs.parse-inputs.outputs.additional_file_to_parse }}
secrets:
BOT_USERNAME: ${{ secrets.BOT_USERNAME }}
SLACK_CHANNEL: ${{ secrets.SLACK_CHANNEL }}
SLACK_WEBHOOK: ${{ secrets.SLACK_WEBHOOK }}
REPO_CHECKOUT_TOKEN: ${{ secrets.REPO_CHECKOUT_TOKEN }}
JOB_SECRET: ${{ secrets.JOB_SECRET }}
SLAB_ACTION_TOKEN: ${{ secrets.SLAB_ACTION_TOKEN }}
SLAB_URL: ${{ secrets.SLAB_URL }}
SLAB_BASE_URL: ${{ secrets.SLAB_BASE_URL }}

View File

@@ -1,5 +1,5 @@
# Run benchmarks on an RTX 4090 machine and return parsed results to Slab CI bot.
name: benchmark_gpu_4090
name: TFHE Cuda Backend - 4090 benchmarks
env:
CARGO_TERM_COLOR: always
@@ -11,64 +11,59 @@ env:
SLACK_ICON: https://pbs.twimg.com/profile_images/1274014582265298945/OjBKP9kn_400x400.png
SLACK_USERNAME: ${{ secrets.BOT_USERNAME }}
SLACK_WEBHOOK: ${{ secrets.SLACK_WEBHOOK }}
BIT_SIZES_SET: FAST
FAST_BENCH: TRUE
on:
# Allows you to run this workflow manually from the Actions tab as an alternative.
workflow_dispatch:
pull_request:
types: [ labeled ]
types: [labeled]
schedule:
# Weekly benchmarks will be triggered each Friday at 9p.m.
- cron: "0 21 * * 5"
permissions:
contents: read
# zizmor: ignore[concurrency-limits] each job manage its concurrency
jobs:
cuda-integer-benchmarks:
name: benchmark_gpu_4090/cuda-integer-benchmarks
name: Cuda integer benchmarks (RTX 4090)
if: ${{ github.event_name == 'workflow_dispatch' ||
github.event_name == 'schedule' && github.repository == 'zama-ai/tfhe-rs' ||
contains(github.event.label.name, '4090_bench') }}
concurrency:
group: ${{ github.workflow_ref }}_cuda_integer_bench
group: ${{ github.workflow }}_${{ github.ref }}_cuda_integer_bench
cancel-in-progress: ${{ github.ref != 'refs/heads/main' }}
runs-on: ["self-hosted", "4090-desktop"]
timeout-minutes: 1440 # 24 hours
strategy:
fail-fast: false
max-parallel: 1
steps:
- name: Checkout tfhe-rs
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd
uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683
with:
fetch-depth: 0
persist-credentials: 'false'
token: ${{ secrets.REPO_CHECKOUT_TOKEN }}
token: ${{ secrets.FHE_ACTIONS_TOKEN }}
- name: Get benchmark details
run: |
COMMIT_DATE=$(git --no-pager show -s --format=%cd --date=iso8601-strict "${SHA}");
{
echo "BENCH_DATE=$(date --iso-8601=seconds)";
echo "COMMIT_DATE=${COMMIT_DATE}";
echo "COMMIT_DATE=$(git --no-pager show -s --format=%cd --date=iso8601-strict ${{ github.sha }})";
echo "COMMIT_HASH=$(git describe --tags --dirty)";
} >> "${GITHUB_ENV}"
env:
SHA: ${{ github.sha }}
echo "FAST_BENCH=TRUE" >> "${GITHUB_ENV}"
- name: Install rust
uses: dtolnay/rust-toolchain@e97e2d8cc328f1b50210efc529dca0028893a2d9 # zizmor: ignore[stale-action-refs] this action doesn't create releases
uses: dtolnay/rust-toolchain@315e265cd78dad1e1dcf3a5074f6d6c47029d5aa
with:
toolchain: nightly
- name: Checkout Slab repo
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd
uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683
with:
repository: zama-ai/slab
path: slab
persist-credentials: 'false'
token: ${{ secrets.REPO_CHECKOUT_TOKEN }}
token: ${{ secrets.FHE_ACTIONS_TOKEN }}
- name: Run integer benchmarks
run: |
@@ -76,20 +71,18 @@ jobs:
- name: Parse results
run: |
python3 ./ci/benchmark_parser.py target/criterion "${RESULTS_FILENAME}" \
python3 ./ci/benchmark_parser.py target/criterion ${{ env.RESULTS_FILENAME }} \
--database tfhe_rs \
--hardware "rtx4090" \
--backend gpu \
--project-version "${COMMIT_HASH}" \
--branch "${REF_NAME}" \
--commit-date "${COMMIT_DATE}" \
--bench-date "${BENCH_DATE}" \
--project-version "${{ env.COMMIT_HASH }}" \
--branch ${{ github.ref_name }} \
--commit-date "${{ env.COMMIT_DATE }}" \
--bench-date "${{ env.BENCH_DATE }}" \
--walk-subdirs
env:
REF_NAME: ${{ github.ref_name }}
- name: Upload parsed results artifact
uses: actions/upload-artifact@043fb46d1a93c77aae656e7c1c64a875d1fc6a0a
uses: actions/upload-artifact@b4b15b8c7c6ac21ea08fcf65892d2ee8f75cf882
with:
name: ${{ github.sha }}_integer_multi_bit_gpu_default
path: ${{ env.RESULTS_FILENAME }}
@@ -97,61 +90,53 @@ jobs:
- name: Send data to Slab
shell: bash
run: |
python3 slab/scripts/data_sender.py "${RESULTS_FILENAME}" "${JOB_SECRET}" \
--slab-url "${SLAB_URL}"
env:
JOB_SECRET: ${{ secrets.JOB_SECRET }}
SLAB_URL: ${{ secrets.SLAB_URL }}
python3 slab/scripts/data_sender.py ${{ env.RESULTS_FILENAME }} "${{ secrets.JOB_SECRET }}" \
--slab-url "${{ secrets.SLAB_URL }}"
- name: Slack Notification
if: ${{ failure() || (cancelled() && github.event_name != 'pull_request') }}
if: ${{ failure() }}
continue-on-error: true
uses: rtCamp/action-slack-notify@e31e87e03dd19038e411e38ae27cbad084a90661
uses: rtCamp/action-slack-notify@c33737706dea87cd7784c687dadc9adf1be59990
env:
SLACK_COLOR: ${{ job.status }}
SLACK_MESSAGE: "Integer RTX 4090 full benchmarks finished with status: ${{ job.status }}. (${{ env.ACTION_RUN_URL }})"
cuda-core-crypto-benchmarks:
name: benchmark_gpu_4090/cuda-core-crypto-benchmarks
name: Cuda core crypto benchmarks (RTX 4090)
if: ${{ github.event_name == 'workflow_dispatch' || github.event_name == 'schedule' || contains(github.event.label.name, '4090_bench') }}
needs: cuda-integer-benchmarks
concurrency:
group: ${{ github.workflow_ref }}_cuda_core_crypto_bench
group: ${{ github.workflow }}_${{ github.ref }}_cuda_core_crypto_bench
cancel-in-progress: ${{ github.ref != 'refs/heads/main' }}
runs-on: ["self-hosted", "4090-desktop"]
timeout-minutes: 1440 # 24 hours
steps:
- name: Checkout tfhe-rs
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd
uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683
with:
fetch-depth: 0
persist-credentials: 'false'
token: ${{ secrets.REPO_CHECKOUT_TOKEN }}
token: ${{ secrets.FHE_ACTIONS_TOKEN }}
- name: Get benchmark details
run: |
COMMIT_DATE=$(git --no-pager show -s --format=%cd --date=iso8601-strict "${SHA}");
{
echo "BENCH_DATE=$(date --iso-8601=seconds)";
echo "COMMIT_DATE=${COMMIT_DATE}";
echo "COMMIT_DATE=$(git --no-pager show -s --format=%cd --date=iso8601-strict ${{ github.sha }})";
echo "COMMIT_HASH=$(git describe --tags --dirty)";
} >> "${GITHUB_ENV}"
env:
SHA: ${{ github.sha }}
- name: Install rust
uses: dtolnay/rust-toolchain@e97e2d8cc328f1b50210efc529dca0028893a2d9 # zizmor: ignore[stale-action-refs] this action doesn't create releases
uses: dtolnay/rust-toolchain@315e265cd78dad1e1dcf3a5074f6d6c47029d5aa
with:
toolchain: nightly
- name: Checkout Slab repo
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd
uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683
with:
repository: zama-ai/slab
path: slab
persist-credentials: 'false'
token: ${{ secrets.REPO_CHECKOUT_TOKEN }}
token: ${{ secrets.FHE_ACTIONS_TOKEN }}
- name: Run core crypto benchmarks
run: |
@@ -160,20 +145,19 @@ jobs:
- name: Parse results
run: |
python3 ./ci/benchmark_parser.py target/criterion "${RESULTS_FILENAME}" \
python3 ./ci/benchmark_parser.py target/criterion ${{ env.RESULTS_FILENAME }} \
--database tfhe_rs \
--hardware "rtx4090" \
--backend gpu \
--project-version "${COMMIT_HASH}" \
--branch "${REF_NAME}" \
--commit-date "${COMMIT_DATE}" \
--bench-date "${BENCH_DATE}" \
--project-version "${{ env.COMMIT_HASH }}" \
--branch ${{ github.ref_name }} \
--commit-date "${{ env.COMMIT_DATE }}" \
--bench-date "${{ env.BENCH_DATE }}" \
--walk-subdirs \
env:
REF_NAME: ${{ github.ref_name }}
- name: Upload parsed results artifact
uses: actions/upload-artifact@043fb46d1a93c77aae656e7c1c64a875d1fc6a0a
uses: actions/upload-artifact@b4b15b8c7c6ac21ea08fcf65892d2ee8f75cf882
with:
name: ${{ github.sha }}_core_crypto
path: ${{ env.RESULTS_FILENAME }}
@@ -181,22 +165,27 @@ jobs:
- name: Send data to Slab
shell: bash
run: |
python3 slab/scripts/data_sender.py "${RESULTS_FILENAME}" "${JOB_SECRET}" \
--slab-url "${SLAB_URL}"
env:
JOB_SECRET: ${{ secrets.JOB_SECRET }}
SLAB_URL: ${{ secrets.SLAB_URL }}
echo "Computing HMac on results file"
SIGNATURE="$(slab/scripts/hmac_calculator.sh ${{ env.RESULTS_FILENAME }} '${{ secrets.JOB_SECRET }}')"
echo "Sending results to Slab..."
curl -v -k \
-H "Content-Type: application/json" \
-H "X-Slab-Repository: ${{ github.repository }}" \
-H "X-Slab-Command: store_data_v2" \
-H "X-Hub-Signature-256: sha256=${SIGNATURE}" \
-d @${{ env.RESULTS_FILENAME }} \
${{ secrets.SLAB_URL }}
- name: Slack Notification
if: ${{ failure() || (cancelled() && github.event_name != 'pull_request') }}
if: ${{ failure() }}
continue-on-error: true
uses: rtCamp/action-slack-notify@e31e87e03dd19038e411e38ae27cbad084a90661
uses: rtCamp/action-slack-notify@c33737706dea87cd7784c687dadc9adf1be59990
env:
SLACK_COLOR: ${{ job.status }}
SLACK_MESSAGE: "Core crypto RTX 4090 full benchmarks finished with status: ${{ job.status }}. (${{ env.ACTION_RUN_URL }})"
remove_github_label:
name: benchmark_gpu_4090/remove_github_label
name: Remove 4090 bench label
if: ${{ always() && github.event_name == 'pull_request' }}
needs: [cuda-integer-benchmarks, cuda-core-crypto-benchmarks]
runs-on: ubuntu-latest

View File

@@ -1,329 +0,0 @@
# Run benchmarks on CUDA instance and return parsed results to Slab CI bot.
name: benchmark_gpu_common
on:
workflow_call:
inputs:
backend:
type: string
default: hyperstack
profile:
type: string
required: true
hardware_name:
type: string
required: true
command: # Use a comma separated values to generate an array
type: string
required: true
op_flavor: # Use a comma separated values to generate an array
type: string
default: default
bench_type:
type: string
default: latency
params_type:
type: string
default: multi_bit
precisions_set:
type: string
default: fast
additional_file_to_parse: # Other files to parse, located under tfhe-benchmark/ directory
type: string # Use comma separated values to generate an array
secrets:
REPO_CHECKOUT_TOKEN:
required: true
SLAB_ACTION_TOKEN:
required: true
SLAB_BASE_URL:
required: true
SLAB_URL:
required: true
JOB_SECRET:
required: true
SLACK_CHANNEL:
required: true
BOT_USERNAME:
required: true
SLACK_WEBHOOK:
required: true
env:
CARGO_TERM_COLOR: always
RESULTS_FILENAME: parsed_benchmark_results_${{ github.sha }}.json
ACTION_RUN_URL: ${{ github.server_url }}/${{ github.repository }}/actions/runs/${{ github.run_id }}
RUST_BACKTRACE: "full"
RUST_MIN_STACK: "8388608"
SLACK_CHANNEL: ${{ secrets.SLACK_CHANNEL }}
SLACK_ICON: https://pbs.twimg.com/profile_images/1274014582265298945/OjBKP9kn_400x400.png
SLACK_USERNAME: ${{ secrets.BOT_USERNAME }}
SLACK_WEBHOOK: ${{ secrets.SLACK_WEBHOOK }}
permissions: {}
# zizmor: ignore[concurrency-limits] caller workflow is responsible for the concurrency
jobs:
prepare-matrix:
name: benchmark_gpu_common/prepare-matrix
runs-on: ubuntu-latest
outputs:
command: ${{ steps.set_matrix_args.outputs.command }}
op_flavor: ${{ steps.set_matrix_args.outputs.op_flavor }}
bench_type: ${{ steps.set_matrix_args.outputs.bench_type }}
params_type: ${{ steps.set_matrix_args.outputs.params_type }}
steps:
- name: Parse user inputs
shell: python
env:
INPUTS_COMMAND: ${{ inputs.command }}
INPUTS_OP_FLAVOR: ${{ inputs.op_flavor }}
INPUTS_BENCH_TYPE: ${{ inputs.bench_type }}
INPUTS_PARAMS_TYPE: ${{ inputs.params_type }}
run: |
import os
inputs_command = os.environ["INPUTS_COMMAND"]
inputs_op_flavor = os.environ["INPUTS_OP_FLAVOR"]
inputs_bench_type = os.environ["INPUTS_BENCH_TYPE"]
inputs_params_type = os.environ["INPUTS_PARAMS_TYPE"]
env_file = os.environ["GITHUB_ENV"]
split_command = inputs_command.replace(" ", "").split(",")
split_op_flavor = inputs_op_flavor.replace(" ", "").split(",")
if inputs_bench_type == "both":
bench_type = ["latency", "throughput"]
else:
bench_type = [inputs_bench_type, ]
if "+" in inputs_params_type:
split_params_type= inputs_params_type.replace(" ", "").split("+")
else:
split_params_type = [inputs_params_type, ]
with open(env_file, "a") as f:
for env_name, values_to_join in [
("COMMAND", split_command),
("OP_FLAVOR", split_op_flavor),
("BENCH_TYPE", bench_type),
("PARAMS_TYPE", split_params_type),
]:
f.write(f"""{env_name}=["{'", "'.join(values_to_join)}"]\n""")
- name: Set matrix arguments outputs
id: set_matrix_args
run: | # zizmor: ignore[template-injection] these env variable are safe
{
echo "command=${{ toJSON(env.COMMAND) }}";
echo "op_flavor=${{ toJSON(env.OP_FLAVOR) }}";
echo "bench_type=${{ toJSON(env.BENCH_TYPE) }}";
echo "params_type=${{ toJSON(env.PARAMS_TYPE) }}";
} >> "${GITHUB_OUTPUT}"
setup-instance:
name: benchmark_gpu_common/setup-instance
needs: prepare-matrix
runs-on: ubuntu-latest
outputs:
runner-name: ${{ steps.start-instance.outputs.label }}
steps:
- name: Start instance
id: start-instance
uses: zama-ai/slab-github-runner@5aee5d157f4a0201e5eaefc9cc648e5f9f5472a5 # v1.6.0
with:
mode: start
github-token: ${{ secrets.SLAB_ACTION_TOKEN }}
slab-url: ${{ secrets.SLAB_BASE_URL }}
job-secret: ${{ secrets.JOB_SECRET }}
backend: ${{ inputs.backend }}
profile: ${{ inputs.profile }}
# Install dependencies only once since cuda-benchmarks uses a matrix strategy, thus running multiple times.
install-dependencies:
name: benchmark_gpu_common/install-dependencies
needs: [ setup-instance ]
runs-on: ${{ needs.setup-instance.outputs.runner-name }}
strategy:
matrix:
# explicit include-based build matrix, of known valid options
include:
- cuda: "12.8"
gcc: 11
steps:
- name: Checkout tfhe-rs repo
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd
with:
fetch-depth: 0
persist-credentials: 'false'
token: ${{ secrets.REPO_CHECKOUT_TOKEN }}
- name: Setup Hyperstack dependencies
uses: ./.github/actions/gpu_setup
with:
cuda-version: ${{ matrix.cuda }}
gcc-version: ${{ matrix.gcc }}
cuda-benchmarks:
name: benchmark_gpu_common/cuda-benchmarks
needs: [ prepare-matrix, setup-instance, install-dependencies ]
runs-on: ${{ needs.setup-instance.outputs.runner-name }}
timeout-minutes: 1440 # 24 hours
strategy:
fail-fast: false
max-parallel: 1
matrix:
command: ${{ fromJSON(needs.prepare-matrix.outputs.command) }}
op_flavor: ${{ fromJSON(needs.prepare-matrix.outputs.op_flavor) }}
bench_type: ${{ fromJSON(needs.prepare-matrix.outputs.bench_type) }}
params_type: ${{ fromJSON(needs.prepare-matrix.outputs.params_type) }}
# explicit include-based build matrix, of known valid options
include:
- cuda: "12.8"
gcc: 11
env:
CUDA_PATH: /usr/local/cuda-${{ matrix.cuda }}
steps:
- name: Checkout tfhe-rs repo with tags
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd
with:
fetch-depth: 0
persist-credentials: 'false'
token: ${{ secrets.REPO_CHECKOUT_TOKEN }}
- name: Get benchmark details
run: |
COMMIT_DATE=$(git --no-pager show -s --format=%cd --date=iso8601-strict "${SHA}");
{
echo "BENCH_DATE=$(date --iso-8601=seconds)";
echo "COMMIT_DATE=${COMMIT_DATE}";
echo "COMMIT_HASH=$(git describe --tags --dirty)";
} >> "${GITHUB_ENV}"
env:
SHA: ${{ github.sha }}
# Re-export environment variables as dependencies setup perform this task in the previous job.
# Local env variables are cleaned at the end of each job.
- name: Export CUDA variables
shell: bash
run: |
echo "CUDA_PATH=$CUDA_PATH" >> "${GITHUB_ENV}"
echo "PATH=$PATH:$CUDA_PATH/bin" >> "${GITHUB_PATH}"
echo "LD_LIBRARY_PATH=$CUDA_PATH/lib64:$LD_LIBRARY_PATH" >> "${GITHUB_ENV}"
echo "CUDA_MODULE_LOADER=EAGER" >> "${GITHUB_ENV}"
- name: Export gcc and g++ variables
shell: bash
run: |
{
echo "CC=/usr/bin/gcc-${GCC_VERSION}";
echo "CXX=/usr/bin/g++-${GCC_VERSION}";
echo "CUDAHOSTCXX=/usr/bin/g++-${GCC_VERSION}";
} >> "${GITHUB_ENV}"
env:
GCC_VERSION: ${{ matrix.gcc }}
- name: Install rust
uses: dtolnay/rust-toolchain@e97e2d8cc328f1b50210efc529dca0028893a2d9 # zizmor: ignore[stale-action-refs] this action doesn't create releases
with:
toolchain: nightly
- name: Run benchmarks
run: |
make BIT_SIZES_SET="${PRECISIONS_SET}" BENCH_OP_FLAVOR="${OP_FLAVOR}" BENCH_TYPE="${BENCH_TYPE}" BENCH_PARAM_TYPE="${BENCH_PARAMS_TYPE}" bench_"${BENCH_COMMAND}"_gpu
env:
OP_FLAVOR: ${{ matrix.op_flavor }}
BENCH_TYPE: ${{ matrix.bench_type }}
BENCH_PARAMS_TYPE: ${{ matrix.params_type }}
BENCH_COMMAND: ${{ matrix.command }}
PRECISIONS_SET: ${{ inputs.precisions_set }}
- name: Parse results
run: |
python3 ./ci/benchmark_parser.py target/criterion "${RESULTS_FILENAME}" \
--database tfhe_rs \
--hardware "${INPUTS_HARDWARE_NAME}" \
--backend gpu \
--project-version "${COMMIT_HASH}" \
--branch "${REF_NAME}" \
--commit-date "${COMMIT_DATE}" \
--bench-date "${BENCH_DATE}" \
--walk-subdirs \
--name-suffix avx512 \
--bench-type "${BENCH_TYPE}"
env:
INPUTS_HARDWARE_NAME: ${{ inputs.hardware_name }}
REF_NAME: ${{ github.ref_name }}
BENCH_TYPE: ${{ matrix.bench_type }}
- name: Parse additional benchmarks results files
if: ${{ inputs.additional_file_to_parse }}
run: |
filenames_list="${filenames}"
IFS=','
for filename in $filenames_list; do
python3 ./ci/benchmark_parser.py "tfhe-benchmark/${filename}" "${RESULTS_FILENAME}" \
--object-size \
--append-results
done
env:
filenames: ${{ inputs.additional_file_to_parse }}
- name: Upload parsed results artifact
uses: actions/upload-artifact@043fb46d1a93c77aae656e7c1c64a875d1fc6a0a
with:
name: ${{ github.sha }}_${{ matrix.command }}_${{ matrix.op_flavor }}_${{ inputs.profile }}_${{ matrix.bench_type }}_${{ matrix.params_type }}
path: ${{ env.RESULTS_FILENAME }}
- name: Checkout Slab repo
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd
with:
repository: zama-ai/slab
path: slab
persist-credentials: 'false'
token: ${{ secrets.REPO_CHECKOUT_TOKEN }}
- name: Send data to Slab
shell: bash
run: |
python3 slab/scripts/data_sender.py "${RESULTS_FILENAME}" "${JOB_SECRET}" \
--slab-url "${SLAB_URL}"
env:
JOB_SECRET: ${{ secrets.JOB_SECRET }}
SLAB_URL: ${{ secrets.SLAB_URL }}
slack-notify:
name: benchmark_gpu_common/slack-notify
needs: [ setup-instance, cuda-benchmarks ]
runs-on: ubuntu-latest
if: ${{ always() && needs.cuda-benchmarks.result != 'skipped' && failure() }}
continue-on-error: true
steps:
- name: Send message
uses: rtCamp/action-slack-notify@e31e87e03dd19038e411e38ae27cbad084a90661
env:
SLACK_COLOR: ${{ needs.cuda-benchmarks.result }}
SLACK_MESSAGE: "Cuda benchmarks (${{ inputs.profile }}) finished with status: ${{ needs.cuda-benchmarks.result }}. (${{ env.ACTION_RUN_URL }})"
teardown-instance:
name: benchmark_gpu_common/teardown-instance
if: ${{ always() && needs.setup-instance.result == 'success' }}
needs: [ setup-instance, cuda-benchmarks, slack-notify ]
runs-on: ubuntu-latest
steps:
- name: Stop instance
id: stop-instance
uses: zama-ai/slab-github-runner@5aee5d157f4a0201e5eaefc9cc648e5f9f5472a5 # v1.6.0
with:
mode: stop
github-token: ${{ secrets.SLAB_ACTION_TOKEN }}
slab-url: ${{ secrets.SLAB_BASE_URL }}
job-secret: ${{ secrets.JOB_SECRET }}
label: ${{ needs.setup-instance.outputs.runner-name }}
- name: Slack Notification
if: ${{ failure() }}
uses: rtCamp/action-slack-notify@e31e87e03dd19038e411e38ae27cbad084a90661
env:
SLACK_COLOR: ${{ job.status }}
SLACK_MESSAGE: "Instance teardown (cuda-${{ inputs.profile }}-benchmarks) finished with status: ${{ job.status }}. (${{ env.ACTION_RUN_URL }})"

View File

@@ -1,342 +0,0 @@
# Run all fhevm coprocessor benchmarks on a GPU instance on Hyperstack and return parsed results to Slab CI bot.
name: benchmark_gpu_coprocessor
on:
workflow_dispatch:
inputs:
profile:
description: "Instance type"
required: true
type: choice
options:
- "l40 (n3-L40x1)"
- "4-l40 (n3-L40x4)"
- "single-h100 (n3-H100x1)"
- "2-h100 (n3-H100x2)"
- "4-h100 (n3-H100x4)"
- "multi-h100 (n3-H100x8)"
- "multi-h100-nvlink (n3-H100x8-NVLink)"
- "multi-h100-sxm5 (n3-H100-SXM5x8)"
- "multi-h100-sxm5_fallback (n3-H100-SXM5x8)"
schedule:
# Weekly tests @ 1AM
- cron: "0 1 * * 6"
permissions:
contents: read
# zizmor: ignore[concurrency-limits] concurrency is managed after instance setup to ensure safe provisioning
env:
CARGO_TERM_COLOR: always
RESULTS_FILENAME: parsed_benchmark_results_${{ github.sha }}.json
ACTION_RUN_URL: ${{ github.server_url }}/${{ github.repository }}/actions/runs/${{ github.run_id }}
RUST_BACKTRACE: "full"
RUST_MIN_STACK: "8388608"
CHECKOUT_TOKEN: ${{ secrets.REPO_CHECKOUT_TOKEN || secrets.GITHUB_TOKEN }}
PROFILE_SCHEDULED_RUN: "multi-h100-sxm5 (n3-H100-SXM5x8)"
PROFILE_MANUAL_RUN: ${{ inputs.profile }}
IS_MANUAL_RUN: ${{ github.event_name == 'workflow_dispatch' }}
BENCHMARK_TYPE: "ALL"
OPTIMIZATION_TARGET: "throughput"
BATCH_SIZE: "5000"
SCHEDULING_POLICY: "MAX_PARALLELISM"
BENCHMARKS: "erc7984"
BRANCH_NAME: ${{ github.ref_name }}
COMMIT_SHA: ${{ github.sha }}
SLAB_SECRET: ${{ secrets.JOB_SECRET }}
jobs:
parse-inputs:
name: benchmark_gpu_coprocessor/parse-inputs
if: github.event_name == 'workflow_dispatch' ||
(github.event_name == 'schedule' && github.repository == 'zama-ai/tfhe-rs')
runs-on: ubuntu-latest
permissions:
contents: 'read'
outputs:
profile: ${{ steps.parse_profile.outputs.profile }}
hardware_name: ${{ steps.parse_hardware_name.outputs.name }}
steps:
- name: Parse profile
id: parse_profile
run: |
if [[ ${IS_MANUAL_RUN} == true ]]; then
PROFILE_RAW="${PROFILE_MANUAL_RUN}"
else
PROFILE_RAW="${PROFILE_SCHEDULED_RUN}"
fi
# shellcheck disable=SC2001
PROFILE_VAL=$(echo "${PROFILE_RAW}" | sed 's|\(.*\)[[:space:]](.*)|\1|')
echo "profile=$PROFILE_VAL" >> "${GITHUB_OUTPUT}"
- name: Parse hardware name
id: parse_hardware_name
run: |
if [[ ${IS_MANUAL_RUN} == true ]]; then
PROFILE_RAW="${PROFILE_MANUAL_RUN}"
else
PROFILE_RAW="${PROFILE_SCHEDULED_RUN}"
fi
# shellcheck disable=SC2001
PROFILE_VAL=$(echo "${PROFILE_RAW}" | sed 's|.*[[:space:]](\(.*\))|\1|')
echo "name=$PROFILE_VAL" >> "${GITHUB_OUTPUT}"
setup-instance:
name: benchmark_gpu_coprocessor/setup-instance
needs: parse-inputs
runs-on: ubuntu-latest
permissions:
contents: 'read'
outputs:
runner-name: ${{ steps.start-remote-instance.outputs.label }}
steps:
- name: Start remote instance
id: start-remote-instance
uses: zama-ai/slab-github-runner@5aee5d157f4a0201e5eaefc9cc648e5f9f5472a5 # v1.6.0
with:
mode: start
github-token: ${{ secrets.SLAB_ACTION_TOKEN }}
slab-url: ${{ secrets.SLAB_BASE_URL }}
job-secret: ${{ secrets.JOB_SECRET }}
backend: hyperstack
profile: ${{ needs.parse-inputs.outputs.profile }}
benchmark-gpu:
name: benchmark_gpu_coprocessor/benchmark-gpu (bpr)
needs: [ parse-inputs, setup-instance ]
runs-on: ${{ needs.setup-instance.outputs.runner-name }}
continue-on-error: true
timeout-minutes: 720 # 12 hours
permissions:
contents: 'read' # Needed to read repositories contents
packages: 'read' # Needed to get fhevm packages
strategy:
fail-fast: false
# explicit include-based build matrix, of known valid options
matrix:
include:
- os: ubuntu-22.04
cuda: "12.8"
gcc: 11
env:
HW_NAME: "${{ needs.parse-inputs.outputs.hardware_name }}"
steps:
- name: Install git LFS
run: |
# Disable unattended-upgrades to avoid lock issues
sudo systemctl mask --now unattended-upgrades
sudo systemctl stop --now unattended-upgrades
sudo apt-get clean
sudo rm -rf /var/lib/apt/lists/*
sudo apt purge -y unattended-upgrades
sudo apt-get update
sudo apt-get install -y git-lfs protobuf-compiler
git lfs install
- name: Checkout tfhe-rs
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # v6.0.2
with:
path: tfhe-rs
persist-credentials: false
- name: Check fhEVM and TFHE-rs repos
run: |
pwd
ls
- name: Checkout fhevm
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # v6.0.2
with:
repository: zama-ai/fhevm
persist-credentials: 'false'
fetch-depth: 0
lfs: true
ref: antoniu/use-tfhe-main-benches
path: fhevm
- name: Get benchmark details
run: |
COMMIT_DATE_ENV=$(git --no-pager show -s --format=%cd --date=iso8601-strict "${COMMIT_SHA}")
{
echo "BENCH_DATE=$(date --iso-8601=seconds)";
echo "COMMIT_DATE=$COMMIT_DATE_ENV";
echo "COMMIT_HASH=$(git rev-parse HEAD)";
} >> "${GITHUB_ENV}"
working-directory: tfhe-rs/
- name: Setup Hyperstack dependencies
uses: ./tfhe-rs/.github/actions/gpu_setup
with:
cuda-version: ${{ matrix.cuda }}
gcc-version: ${{ matrix.gcc }}
github-instance: ${{ env.SECRETS_AVAILABLE == 'false' }}
- name: Check fhEVM and TFHE-rs repos
run: |
pwd
ls
mv tfhe-rs fhevm/coprocessor/
- name: Checkout LFS objects
run: git lfs checkout
working-directory: fhevm/
- name: Install rust
uses: dtolnay/rust-toolchain@e97e2d8cc328f1b50210efc529dca0028893a2d9 # zizmor: ignore[stale-action-refs] this action doesn't create releases
with:
toolchain: nightly
- name: Install cargo dependencies
run: |
sudo apt-get install -y protobuf-compiler pkg-config libssl-dev \
libclang-dev docker-compose-v2 docker.io acl
sudo usermod -aG docker "$USER"
newgrp docker
sudo setfacl --modify user:"$USER":rw /var/run/docker.sock
cargo install sqlx-cli
- name: Install foundry
uses: foundry-rs/foundry-toolchain@8789b3e21e6c11b2697f5eb56eddae542f746c10
- name: Cache cargo
uses: actions/cache@27d5ce7f107fe9357f9df03efb73ab90386fccae # v5.0.5
with:
path: |
~/.cargo/registry
~/.cargo/git
target
key: ${{ runner.os }}-cargo-${{ hashFiles('**/Cargo.lock') }}
restore-keys: ${{ runner.os }}-cargo-
- name: Login to GitHub Container Registry
uses: docker/login-action@4907a6ddec9925e35a0a9e82d7399ccc52663121 # v4.1.0
with:
registry: ghcr.io
username: ${{ github.actor }}
password: ${{ secrets.GITHUB_TOKEN }}
- name: Login to Chainguard Registry
uses: docker/login-action@4907a6ddec9925e35a0a9e82d7399ccc52663121 # v4.1.0
with:
registry: cgr.dev
username: ${{ secrets.CGR_USERNAME }}
password: ${{ secrets.CGR_PASSWORD }}
- name: Init database
run: make init_db
working-directory: fhevm/coprocessor/fhevm-engine/tfhe-worker
- name: Use Node.js
uses: actions/setup-node@48b55a011bda9f5d6aeb4c2d9c7362e8dae4041e # v6.4.0
with:
node-version: 20.x
- name: Build contracts
env:
HARDHAT_NETWORK: hardhat
run: |
ls
pwd
cp ./host-contracts/.env.example ./host-contracts/.env
cd ./host-contracts
npm ci --include=optional
npm install && npm run deploy:emptyProxies && npx hardhat compile
working-directory: fhevm/
- name: Profile erc7984 no-cmux benchmark on GPU
run: |
BENCHMARK_BATCH_SIZE="${BATCH_SIZE}" \
FHEVM_DF_SCHEDULE="${SCHEDULING_POLICY}" \
BENCHMARK_TYPE="THROUGHPUT_200" \
OPTIMIZATION_TARGET="${OPTIMIZATION_TARGET}" \
make -e "profile_erc7984_gpu"
working-directory: fhevm/coprocessor/fhevm-engine/tfhe-worker
- name: Get nsys profile name
id: nsys_profile_name
run: echo "profile=coprocessor_profile_$(date +"%Y-%m-%d-%Hh").nsys-rep" >> "$GITHUB_OUTPUT"
- name: Timestamp nsys profile # zizmor: ignore[template-injection]
env:
REPORT_NAME: ${{ steps.nsys_profile_name.outputs.profile }}
run: |
mv report1.nsys-rep ${{ env.REPORT_NAME }}
working-directory: fhevm/coprocessor/fhevm-engine/tfhe-worker
- name: Upload profile artifact
env:
REPORT_NAME: ${{ steps.nsys_profile_name.outputs.profile }}
uses: actions/upload-artifact@043fb46d1a93c77aae656e7c1c64a875d1fc6a0a
with:
name: ${{ env.REPORT_NAME }}
path: fhevm/coprocessor/fhevm-engine/tfhe-worker/${{ env.REPORT_NAME }}
- name: Run latency benchmark on GPU
run: |
BENCHMARK_BATCH_SIZE="${BATCH_SIZE}" FHEVM_DF_SCHEDULE="${SCHEDULING_POLICY}" BENCHMARK_TYPE="LATENCY" OPTIMIZATION_TARGET="${OPTIMIZATION_TARGET}" make -e "benchmark_${BENCHMARKS}_gpu"
working-directory: fhevm/coprocessor/fhevm-engine/tfhe-worker
- name: Run throughput benchmarks on GPU
run: |
BENCHMARK_BATCH_SIZE="${BATCH_SIZE}" FHEVM_DF_SCHEDULE="${SCHEDULING_POLICY}" BENCHMARK_TYPE="THROUGHPUT_200" OPTIMIZATION_TARGET="${OPTIMIZATION_TARGET}" make -e "benchmark_${BENCHMARKS}_gpu"
working-directory: fhevm/coprocessor/fhevm-engine/tfhe-worker
- name: Parse results
run: |
python3 ./ci/benchmark_parser.py coprocessor/fhevm-engine/target/criterion "${RESULTS_FILENAME}" \
--database coprocessor \
--hardware "${HW_NAME}" \
--backend gpu \
--project-version "${COMMIT_HASH}" \
--branch "${BRANCH_NAME}" \
--commit-date "${COMMIT_DATE}" \
--bench-date "${BENCH_DATE}" \
--walk-subdirs \
--crate "coprocessor/fhevm-engine/tfhe-worker" \
--name-suffix "operation_batch_size_${BATCH_SIZE}-schedule_${SCHEDULING_POLICY}-optimization_target_${OPTIMIZATION_TARGET}"
working-directory: fhevm/
- name: Upload parsed results artifact
uses: actions/upload-artifact@043fb46d1a93c77aae656e7c1c64a875d1fc6a0a
with:
name: ${COMMIT_SHA}_${BENCHMARKS}_${{ needs.parse-inputs.outputs.profile }}
path: fhevm/$${{ env.RESULTS_FILENAME }}
- name: Checkout Slab repo
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd
with:
repository: zama-ai/slab
path: slab
persist-credentials: 'false'
token: ${{ secrets.REPO_CHECKOUT_TOKEN }}
- name: Send data to Slab
shell: bash
env:
SLAB_URL: ${{ secrets.SLAB_URL }}
run: |
python3 slab/scripts/data_sender.py fhevm/"${RESULTS_FILENAME}" "${SLAB_SECRET}" \
--slab-url "${SLAB_URL}"
teardown-instance:
name: benchmark_gpu_coprocessor/teardown-instance
if: ${{ always() && needs.setup-instance.result == 'success' }}
needs: [ setup-instance, benchmark-gpu ]
runs-on: ubuntu-latest
permissions:
contents: 'read'
steps:
- name: Stop remote instance
id: stop-instance
uses: zama-ai/slab-github-runner@5aee5d157f4a0201e5eaefc9cc648e5f9f5472a5 # v1.6.0
with:
mode: stop
github-token: ${{ secrets.SLAB_ACTION_TOKEN }}
slab-url: ${{ secrets.SLAB_BASE_URL }}
job-secret: ${{ secrets.JOB_SECRET }}
label: ${{ needs.setup-instance.outputs.runner-name }}

View File

@@ -0,0 +1,154 @@
# Run core crypto benchmarks on an instance with CUDA and return parsed results to Slab CI bot.
name: Core crypto GPU benchmarks
on:
workflow_dispatch:
schedule:
# Weekly benchmarks will be triggered each Saturday at 1a.m.
- cron: '0 1 * * 6'
env:
CARGO_TERM_COLOR: always
RESULTS_FILENAME: parsed_benchmark_results_${{ github.sha }}.json
ACTION_RUN_URL: ${{ github.server_url }}/${{ github.repository }}/actions/runs/${{ github.run_id }}
SLACK_CHANNEL: ${{ secrets.SLACK_CHANNEL }}
SLACK_ICON: https://pbs.twimg.com/profile_images/1274014582265298945/OjBKP9kn_400x400.png
SLACK_USERNAME: ${{ secrets.BOT_USERNAME }}
SLACK_WEBHOOK: ${{ secrets.SLACK_WEBHOOK }}
jobs:
setup-instance:
name: Setup instance (cuda-core-crypto-benchmarks)
runs-on: ubuntu-latest
if: github.event_name != 'schedule' ||
(github.event_name == 'schedule' && github.repository == 'zama-ai/tfhe-rs')
outputs:
runner-name: ${{ steps.start-instance.outputs.label }}
steps:
- name: Start instance
id: start-instance
uses: zama-ai/slab-github-runner@98f0788261a7323d5d695a883e20df36591a92b7
with:
mode: start
github-token: ${{ secrets.SLAB_ACTION_TOKEN }}
slab-url: ${{ secrets.SLAB_BASE_URL }}
job-secret: ${{ secrets.JOB_SECRET }}
backend: hyperstack
profile: single-h100
cuda-core-crypto-benchmarks:
name: Execute GPU core crypto benchmarks
needs: setup-instance
runs-on: ${{ needs.setup-instance.outputs.runner-name }}
strategy:
fail-fast: false
# explicit include-based build matrix, of known valid options
matrix:
include:
- os: ubuntu-22.04
cuda: "12.2"
gcc: 11
steps:
- name: Checkout tfhe-rs repo with tags
uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683
with:
fetch-depth: 0
token: ${{ secrets.FHE_ACTIONS_TOKEN }}
- name: Setup Hyperstack dependencies
uses: ./.github/actions/hyperstack_setup
with:
cuda-version: ${{ matrix.cuda }}
gcc-version: ${{ matrix.gcc }}
- name: Get benchmark details
run: |
{
echo "BENCH_DATE=$(date --iso-8601=seconds)";
echo "COMMIT_DATE=$(git --no-pager show -s --format=%cd --date=iso8601-strict ${{ github.sha }})";
echo "COMMIT_HASH=$(git describe --tags --dirty)";
} >> "${GITHUB_ENV}"
- name: Set up home
# "Install rust" step require root user to have a HOME directory which is not set.
run: |
echo "HOME=/home/ubuntu" >> "${GITHUB_ENV}"
- name: Install rust
uses: dtolnay/rust-toolchain@315e265cd78dad1e1dcf3a5074f6d6c47029d5aa
with:
toolchain: nightly
- name: Run benchmarks with AVX512
run: |
make bench_pbs_gpu
make bench_ks_gpu
- name: Parse results
run: |
python3 ./ci/benchmark_parser.py target/criterion ${{ env.RESULTS_FILENAME }} \
--database tfhe_rs \
--hardware "n3-H100x1" \
--backend gpu \
--project-version "${{ env.COMMIT_HASH }}" \
--branch ${{ github.ref_name }} \
--commit-date "${{ env.COMMIT_DATE }}" \
--bench-date "${{ env.BENCH_DATE }}" \
--name-suffix avx512 \
--walk-subdirs
- name: Upload parsed results artifact
uses: actions/upload-artifact@b4b15b8c7c6ac21ea08fcf65892d2ee8f75cf882
with:
name: ${{ github.sha }}_core_crypto
path: ${{ env.RESULTS_FILENAME }}
- name: Checkout Slab repo
uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683
with:
repository: zama-ai/slab
path: slab
token: ${{ secrets.FHE_ACTIONS_TOKEN }}
- name: Send data to Slab
shell: bash
run: |
python3 slab/scripts/data_sender.py ${{ env.RESULTS_FILENAME }} "${{ secrets.JOB_SECRET }}" \
--slab-url "${{ secrets.SLAB_URL }}"
slack-notify:
name: Slack Notification
needs: [ setup-instance, cuda-core-crypto-benchmarks ]
runs-on: ubuntu-latest
if: ${{ !success() && !cancelled() }}
continue-on-error: true
steps:
- name: Send message
uses: rtCamp/action-slack-notify@c33737706dea87cd7784c687dadc9adf1be59990
env:
SLACK_COLOR: ${{ needs.cuda-core-crypto-benchmarks.result }}
SLACK_MESSAGE: "PBS GPU benchmarks finished with status: ${{ needs.cuda-core-crypto-benchmarks.result }}. (${{ env.ACTION_RUN_URL }})"
teardown-instance:
name: Teardown instance (cuda-integer-full-benchmarks)
if: ${{ always() && needs.setup-instance.result != 'skipped' }}
needs: [ setup-instance, cuda-core-crypto-benchmarks, slack-notify ]
runs-on: ubuntu-latest
steps:
- name: Stop instance
id: stop-instance
uses: zama-ai/slab-github-runner@98f0788261a7323d5d695a883e20df36591a92b7
with:
mode: stop
github-token: ${{ secrets.SLAB_ACTION_TOKEN }}
slab-url: ${{ secrets.SLAB_BASE_URL }}
job-secret: ${{ secrets.JOB_SECRET }}
label: ${{ needs.setup-instance.outputs.runner-name }}
- name: Slack Notification
if: ${{ failure() }}
continue-on-error: true
uses: rtCamp/action-slack-notify@c33737706dea87cd7784c687dadc9adf1be59990
env:
SLACK_COLOR: ${{ job.status }}
SLACK_MESSAGE: "Instance teardown (cuda-core-crypto-benchmarks) finished with status: ${{ job.status }}. (${{ env.ACTION_RUN_URL }})"

View File

@@ -0,0 +1,44 @@
# Run CUDA ERC20 benchmarks on a Hyperstack VM and return parsed results to Slab CI bot.
name: Cuda ERC20 benchmarks
on:
workflow_dispatch:
inputs:
profile:
description: "Instance type"
required: true
type: choice
options:
- "l40 (n3-L40x1)"
- "single-h100 (n3-H100x1)"
- "2-h100 (n3-H100x2)"
- "4-h100 (n3-H100x4)"
- "multi-h100 (n3-H100x8)"
- "multi-h100-nvlink (n3-H100x8-NVLink)"
- "multi-h100-sxm5 (n3-H100x8-SXM5)"
jobs:
parse-inputs:
runs-on: ubuntu-latest
outputs:
profile: ${{ steps.parse_profile.outputs.profile }}
hardware_name: ${{ steps.parse_hardware_name.outputs.name }}
steps:
- name: Parse profile
id: parse_profile
run: |
echo "profile=$(echo '${{ inputs.profile }}' | sed 's|\(.*\)[[:space:]](.*)|\1|')" >> "${GITHUB_OUTPUT}"
- name: Parse hardware name
id: parse_hardware_name
run: |
echo "name=$(echo '${{ inputs.profile }}' | sed 's|.*[[:space:]](\(.*\))|\1|')" >> "${GITHUB_OUTPUT}"
run-benchmarks:
name: Run benchmarks
needs: parse-inputs
uses: ./.github/workflows/benchmark_gpu_erc20_common.yml
with:
profile: ${{ needs.parse-inputs.outputs.profile }}
hardware_name: ${{ needs.parse-inputs.outputs.hardware_name }}
secrets: inherit

View File

@@ -0,0 +1,180 @@
# Run ERC20 benchmarks on an instance with CUDA and return parsed results to Slab CI bot.
name: Cuda ERC20 benchmarks - common
on:
workflow_call:
inputs:
backend:
type: string
default: hyperstack
profile:
type: string
required: true
hardware_name:
type: string
required: true
secrets:
FHE_ACTIONS_TOKEN:
required: true
SLAB_ACTION_TOKEN:
required: true
SLAB_BASE_URL:
required: true
SLAB_URL:
required: true
JOB_SECRET:
required: true
SLACK_CHANNEL:
required: true
BOT_USERNAME:
required: true
SLACK_WEBHOOK:
required: true
env:
CARGO_TERM_COLOR: always
RESULTS_FILENAME: parsed_benchmark_results_${{ github.sha }}.json
PARSE_INTEGER_BENCH_CSV_FILE: tfhe_rs_integer_benches_${{ github.sha }}.csv
ACTION_RUN_URL: ${{ github.server_url }}/${{ github.repository }}/actions/runs/${{ github.run_id }}
RUST_BACKTRACE: "full"
RUST_MIN_STACK: "8388608"
SLACK_CHANNEL: ${{ secrets.SLACK_CHANNEL }}
SLACK_ICON: https://pbs.twimg.com/profile_images/1274014582265298945/OjBKP9kn_400x400.png
SLACK_USERNAME: ${{ secrets.BOT_USERNAME }}
SLACK_WEBHOOK: ${{ secrets.SLACK_WEBHOOK }}
jobs:
setup-instance:
name: Setup instance (cuda-erc20-benchmarks)
runs-on: ubuntu-latest
if: github.event_name == 'workflow_dispatch' ||
(github.event_name == 'schedule' && github.repository == 'zama-ai/tfhe-rs')
outputs:
runner-name: ${{ steps.start-instance.outputs.label }}
steps:
- name: Start instance
id: start-instance
uses: zama-ai/slab-github-runner@98f0788261a7323d5d695a883e20df36591a92b7
with:
mode: start
github-token: ${{ secrets.SLAB_ACTION_TOKEN }}
slab-url: ${{ secrets.SLAB_BASE_URL }}
job-secret: ${{ secrets.JOB_SECRET }}
backend: ${{ inputs.backend }}
profile: ${{ inputs.profile }}
cuda-erc20-benchmarks:
name: Cuda ERC20 benchmarks (${{ inputs.profile }})
needs: setup-instance
runs-on: ${{ needs.setup-instance.outputs.runner-name }}
strategy:
fail-fast: false
# explicit include-based build matrix, of known valid options
matrix:
include:
- os: ubuntu-22.04
cuda: "12.2"
gcc: 11
steps:
- name: Checkout tfhe-rs repo with tags
uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683
with:
fetch-depth: 0
token: ${{ secrets.FHE_ACTIONS_TOKEN }}
- name: Setup Hyperstack dependencies
uses: ./.github/actions/hyperstack_setup
with:
cuda-version: ${{ matrix.cuda }}
gcc-version: ${{ matrix.gcc }}
- name: Get benchmark details
run: |
{
echo "BENCH_DATE=$(date --iso-8601=seconds)";
echo "COMMIT_DATE=$(git --no-pager show -s --format=%cd --date=iso8601-strict ${{ github.sha }})";
echo "COMMIT_HASH=$(git describe --tags --dirty)";
} >> "${GITHUB_ENV}"
- name: Set up home
# "Install rust" step require root user to have a HOME directory which is not set.
run: |
echo "HOME=/home/ubuntu" >> "${GITHUB_ENV}"
- name: Install rust
uses: dtolnay/rust-toolchain@315e265cd78dad1e1dcf3a5074f6d6c47029d5aa
with:
toolchain: nightly
- name: Run benchmarks
run: |
make bench_hlapi_erc20_gpu
- name: Parse results
run: |
python3 ./ci/benchmark_parser.py target/criterion ${{ env.RESULTS_FILENAME }} \
--database tfhe_rs \
--hardware "${{ inputs.hardware_name }}" \
--backend gpu \
--project-version "${{ env.COMMIT_HASH }}" \
--branch ${{ github.ref_name }} \
--commit-date "${{ env.COMMIT_DATE }}" \
--bench-date "${{ env.BENCH_DATE }}" \
--walk-subdirs \
--name-suffix avx512
- name: Upload parsed results artifact
uses: actions/upload-artifact@b4b15b8c7c6ac21ea08fcf65892d2ee8f75cf882
with:
name: ${{ github.sha }}_erc20_${{ inputs.profile }}
path: ${{ env.RESULTS_FILENAME }}
- name: Checkout Slab repo
uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683
with:
repository: zama-ai/slab
path: slab
token: ${{ secrets.FHE_ACTIONS_TOKEN }}
- name: Send data to Slab
shell: bash
run: |
python3 slab/scripts/data_sender.py ${{ env.RESULTS_FILENAME }} "${{ secrets.JOB_SECRET }}" \
--slab-url "${{ secrets.SLAB_URL }}"
slack-notify:
name: Slack Notification
needs: [ setup-instance, cuda-erc20-benchmarks ]
runs-on: ubuntu-latest
if: ${{ always() && needs.cuda-erc20-benchmarks.result != 'skipped' && failure() }}
continue-on-error: true
steps:
- name: Send message
uses: rtCamp/action-slack-notify@c33737706dea87cd7784c687dadc9adf1be59990
env:
SLACK_COLOR: ${{ needs.cuda-erc20-benchmarks.result }}
SLACK_MESSAGE: "Cuda ERC20 benchmarks (${{ inputs.profile }}) finished with status: ${{ needs.cuda-erc20-benchmarks.result }}. (${{ env.ACTION_RUN_URL }})"
teardown-instance:
name: Teardown instance (cuda-erc20-${{ inputs.profile }}-benchmarks)
if: ${{ always() && needs.setup-instance.result != 'skipped' }}
needs: [ setup-instance, cuda-erc20-benchmarks, slack-notify ]
runs-on: ubuntu-latest
steps:
- name: Stop instance
id: stop-instance
uses: zama-ai/slab-github-runner@98f0788261a7323d5d695a883e20df36591a92b7
with:
mode: stop
github-token: ${{ secrets.SLAB_ACTION_TOKEN }}
slab-url: ${{ secrets.SLAB_BASE_URL }}
job-secret: ${{ secrets.JOB_SECRET }}
label: ${{ needs.setup-instance.outputs.runner-name }}
- name: Slack Notification
if: ${{ failure() }}
continue-on-error: true
uses: rtCamp/action-slack-notify@c33737706dea87cd7784c687dadc9adf1be59990
env:
SLACK_COLOR: ${{ job.status }}
SLACK_MESSAGE: "Instance teardown (cuda-erc20-${{ inputs.profile }}-benchmarks) finished with status: ${{ job.status }}. (${{ env.ACTION_RUN_URL }})"

View File

@@ -0,0 +1,35 @@
# Run CUDA ERC20 benchmarks on multiple Hyperstack VMs and return parsed results to Slab CI bot.
name: Cuda ERC20 weekly benchmarks
on:
schedule:
# Weekly benchmarks will be triggered each Saturday at 5a.m.
- cron: '0 5 * * 6'
jobs:
run-benchmarks-1-h100:
name: Run benchmarks (1xH100)
if: github.repository == 'zama-ai/tfhe-rs'
uses: ./.github/workflows/benchmark_gpu_erc20_common.yml
with:
profile: single-h100
hardware_name: n3-H100x1
secrets: inherit
run-benchmarks-2-h100:
name: Run benchmarks (2xH100)
if: github.repository == 'zama-ai/tfhe-rs'
uses: ./.github/workflows/benchmark_gpu_erc20_common.yml
with:
profile: 2-h100
hardware_name: n3-H100x2
secrets: inherit
run-benchmarks-8-h100:
name: Run benchmarks (8xH100)
if: github.repository == 'zama-ai/tfhe-rs'
uses: ./.github/workflows/benchmark_gpu_erc20_common.yml
with:
profile: multi-h100
hardware_name: n3-H100x8
secrets: inherit

View File

@@ -0,0 +1,79 @@
# Run CUDA benchmarks on a Hyperstack VM and return parsed results to Slab CI bot.
name: Cuda benchmarks
on:
workflow_dispatch:
inputs:
profile:
description: "Instance type"
required: true
type: choice
options:
- "l40 (n3-L40x1)"
- "single-h100 (n3-H100x1)"
- "2-h100 (n3-H100x2)"
- "4-h100 (n3-H100x4)"
- "multi-h100 (n3-H100x8)"
- "multi-h100-nvlink (n3-H100x8-NVLink)"
- "multi-h100-sxm5 (n3-H100x8-SXM5)"
- "multi-a100-nvlink (n3-A100x8-NVLink)"
command:
description: "Benchmark command to run"
type: choice
default: integer_multi_bit
options:
- integer
- integer_multi_bit
- integer_compression
- pbs
- ks
op_flavor:
description: "Operations set to run"
type: choice
default: default
options:
- default
- fast_default
- unchecked
all_precisions:
description: "Run all precisions"
type: boolean
default: false
bench_type:
description: "Benchmarks type"
type: choice
default: latency
options:
- latency
- throughput
- both
jobs:
parse-inputs:
runs-on: ubuntu-latest
outputs:
profile: ${{ steps.parse_profile.outputs.profile }}
hardware_name: ${{ steps.parse_hardware_name.outputs.name }}
steps:
- name: Parse profile
id: parse_profile
run: |
echo "profile=$(echo '${{ inputs.profile }}' | sed 's|\(.*\)[[:space:]](.*)|\1|')" >> "${GITHUB_OUTPUT}"
- name: Parse hardware name
id: parse_hardware_name
run: |
echo "name=$(echo '${{ inputs.profile }}' | sed 's|.*[[:space:]](\(.*\))|\1|')" >> "${GITHUB_OUTPUT}"
run-benchmarks:
name: Run benchmarks
needs: parse-inputs
uses: ./.github/workflows/benchmark_gpu_integer_common.yml
with:
profile: ${{ needs.parse-inputs.outputs.profile }}
hardware_name: ${{ needs.parse-inputs.outputs.hardware_name }}
command: ${{ inputs.command }}
op_flavor: ${{ inputs.op_flavor }}
bench_type: ${{ inputs.bench_type }}
all_precisions: ${{ inputs.all_precisions }}
secrets: inherit

View File

@@ -0,0 +1,256 @@
# Run integer benchmarks on CUDA instance and return parsed results to Slab CI bot.
name: Cuda benchmarks - common
on:
workflow_call:
inputs:
backend:
type: string
default: hyperstack
profile:
type: string
required: true
hardware_name:
type: string
required: true
command: # Use a comma separated values to generate an array
type: string
required: true
op_flavor: # Use a comma separated values to generate an array
type: string
required: true
bench_type:
type: string
default: latency
all_precisions:
type: boolean
default: false
secrets:
FHE_ACTIONS_TOKEN:
required: true
SLAB_ACTION_TOKEN:
required: true
SLAB_BASE_URL:
required: true
SLAB_URL:
required: true
JOB_SECRET:
required: true
SLACK_CHANNEL:
required: true
BOT_USERNAME:
required: true
SLACK_WEBHOOK:
required: true
env:
CARGO_TERM_COLOR: always
RESULTS_FILENAME: parsed_benchmark_results_${{ github.sha }}.json
ACTION_RUN_URL: ${{ github.server_url }}/${{ github.repository }}/actions/runs/${{ github.run_id }}
RUST_BACKTRACE: "full"
RUST_MIN_STACK: "8388608"
SLACK_CHANNEL: ${{ secrets.SLACK_CHANNEL }}
SLACK_ICON: https://pbs.twimg.com/profile_images/1274014582265298945/OjBKP9kn_400x400.png
SLACK_USERNAME: ${{ secrets.BOT_USERNAME }}
SLACK_WEBHOOK: ${{ secrets.SLACK_WEBHOOK }}
FAST_BENCH: TRUE
jobs:
prepare-matrix:
name: Prepare operations matrix
runs-on: ubuntu-latest
outputs:
command: ${{ steps.set_command.outputs.command }}
op_flavor: ${{ steps.set_op_flavor.outputs.op_flavor }}
bench_type: ${{ steps.set_bench_type.outputs.bench_type }}
steps:
- name: Set single command
if: ${{ !contains(inputs.command, ',')}}
run: |
echo "COMMAND=[\"${{ inputs.command }}\"]" >> "${GITHUB_ENV}"
- name: Set multiple commands
if: ${{ contains(inputs.command, ',')}}
run: |
PARSED_COMMAND=$(echo "${{ inputs.command }}" | sed 's/[[:space:]]*,[[:space:]]*/\\", \\"/g')
echo "COMMAND=[\"${PARSED_COMMAND}\"]" >> "${GITHUB_ENV}"
- name: Set single operations flavor
if: ${{ !contains(inputs.op_flavor, ',')}}
run: |
echo "OP_FLAVOR=[\"${{ inputs.op_flavor }}\"]" >> "${GITHUB_ENV}"
- name: Set multiple operations flavors
if: ${{ contains(inputs.op_flavor, ',')}}
run: |
PARSED_OP_FLAVOR=$(echo "${{ inputs.op_flavor }}" | sed 's/[[:space:]]*,[[:space:]]*/", "/g')
echo "OP_FLAVOR=[\"${PARSED_OP_FLAVOR}\"]" >> "${GITHUB_ENV}"
- name: Set benchmark types
run: |
if [[ "${{ inputs.bench_type }}" == "both" ]]; then
echo "BENCH_TYPE=[\"latency\", \"throughput\"]" >> "${GITHUB_ENV}"
else
echo "BENCH_TYPE=[\"${{ inputs.bench_type }}\"]" >> "${GITHUB_ENV}"
fi
- name: Set command output
id: set_command
run: |
echo "command=${{ toJSON(env.COMMAND) }}" >> "${GITHUB_OUTPUT}"
- name: Set operation flavor output
id: set_op_flavor
run: |
echo "op_flavor=${{ toJSON(env.OP_FLAVOR) }}" >> "${GITHUB_OUTPUT}"
- name: Set benchmark types output
id: set_bench_type
run: |
echo "bench_type=${{ toJSON(env.BENCH_TYPE) }}" >> "${GITHUB_OUTPUT}"
setup-instance:
name: Setup instance (cuda-${{ inputs.profile }}-benchmarks)
needs: prepare-matrix
runs-on: ubuntu-latest
outputs:
runner-name: ${{ steps.start-instance.outputs.label }}
steps:
- name: Start instance
id: start-instance
uses: zama-ai/slab-github-runner@98f0788261a7323d5d695a883e20df36591a92b7
with:
mode: start
github-token: ${{ secrets.SLAB_ACTION_TOKEN }}
slab-url: ${{ secrets.SLAB_BASE_URL }}
job-secret: ${{ secrets.JOB_SECRET }}
backend: ${{ inputs.backend }}
profile: ${{ inputs.profile }}
cuda-benchmarks:
name: Cuda benchmarks (${{ inputs.profile }})
needs: [ prepare-matrix, setup-instance ]
runs-on: ${{ needs.setup-instance.outputs.runner-name }}
timeout-minutes: 1440 # 24 hours
continue-on-error: true
strategy:
fail-fast: false
max-parallel: 1
matrix:
command: ${{ fromJSON(needs.prepare-matrix.outputs.command) }}
op_flavor: ${{ fromJSON(needs.prepare-matrix.outputs.op_flavor) }}
bench_type: ${{ fromJSON(needs.prepare-matrix.outputs.bench_type) }}
# explicit include-based build matrix, of known valid options
include:
- os: ubuntu-22.04
cuda: "12.2"
gcc: 11
steps:
- name: Checkout tfhe-rs repo with tags
uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683
with:
fetch-depth: 0
token: ${{ secrets.FHE_ACTIONS_TOKEN }}
- name: Setup Hyperstack dependencies
uses: ./.github/actions/hyperstack_setup
with:
cuda-version: ${{ matrix.cuda }}
gcc-version: ${{ matrix.gcc }}
- name: Get benchmark details
run: |
{
echo "BENCH_DATE=$(date --iso-8601=seconds)";
echo "COMMIT_DATE=$(git --no-pager show -s --format=%cd --date=iso8601-strict ${{ github.sha }})";
echo "COMMIT_HASH=$(git describe --tags --dirty)";
} >> "${GITHUB_ENV}"
- name: Set up home
# "Install rust" step require root user to have a HOME directory which is not set.
run: |
echo "HOME=/home/ubuntu" >> "${GITHUB_ENV}"
- name: Install rust
uses: dtolnay/rust-toolchain@315e265cd78dad1e1dcf3a5074f6d6c47029d5aa
with:
toolchain: nightly
- name: Should run benchmarks with all precisions
if: inputs.all_precisions
run: |
echo "FAST_BENCH=FALSE" >> "${GITHUB_ENV}"
- name: Run benchmarks
run: |
make BENCH_OP_FLAVOR=${{ matrix.op_flavor }} BENCH_TYPE=${{ matrix.bench_type }} bench_${{ matrix.command }}_gpu
- name: Parse results
run: |
python3 ./ci/benchmark_parser.py target/criterion ${{ env.RESULTS_FILENAME }} \
--database tfhe_rs \
--hardware "${{ inputs.hardware_name }}" \
--backend gpu \
--project-version "${{ env.COMMIT_HASH }}" \
--branch ${{ github.ref_name }} \
--commit-date "${{ env.COMMIT_DATE }}" \
--bench-date "${{ env.BENCH_DATE }}" \
--walk-subdirs \
--name-suffix avx512 \
--bench-type ${{ matrix.bench_type }}
- name: Upload parsed results artifact
uses: actions/upload-artifact@b4b15b8c7c6ac21ea08fcf65892d2ee8f75cf882
with:
name: ${{ github.sha }}_${{ matrix.command }}_${{ matrix.op_flavor }}_${{ inputs.profile }}
path: ${{ env.RESULTS_FILENAME }}
- name: Checkout Slab repo
uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683
with:
repository: zama-ai/slab
path: slab
token: ${{ secrets.FHE_ACTIONS_TOKEN }}
- name: Send data to Slab
shell: bash
run: |
python3 slab/scripts/data_sender.py ${{ env.RESULTS_FILENAME }} "${{ secrets.JOB_SECRET }}" \
--slab-url "${{ secrets.SLAB_URL }}"
slack-notify:
name: Slack Notification
needs: [ setup-instance, cuda-benchmarks ]
runs-on: ubuntu-latest
if: ${{ always() && needs.cuda-benchmarks.result != 'skipped' && failure() }}
continue-on-error: true
steps:
- name: Send message
uses: rtCamp/action-slack-notify@c33737706dea87cd7784c687dadc9adf1be59990
env:
SLACK_COLOR: ${{ needs.cuda-benchmarks.result }}
SLACK_MESSAGE: "Cuda benchmarks (${{ inputs.profile }}) finished with status: ${{ needs.cuda-benchmarks.result }}. (${{ env.ACTION_RUN_URL }})"
teardown-instance:
name: Teardown instance (cuda-${{ inputs.profile }}-benchmarks)
if: ${{ always() && needs.setup-instance.result != 'skipped' }}
needs: [ setup-instance, cuda-benchmarks, slack-notify ]
runs-on: ubuntu-latest
steps:
- name: Stop instance
id: stop-instance
uses: zama-ai/slab-github-runner@98f0788261a7323d5d695a883e20df36591a92b7
with:
mode: stop
github-token: ${{ secrets.SLAB_ACTION_TOKEN }}
slab-url: ${{ secrets.SLAB_BASE_URL }}
job-secret: ${{ secrets.JOB_SECRET }}
label: ${{ needs.setup-instance.outputs.runner-name }}
- name: Slack Notification
if: ${{ failure() }}
continue-on-error: true
uses: rtCamp/action-slack-notify@c33737706dea87cd7784c687dadc9adf1be59990
env:
SLACK_COLOR: ${{ job.status }}
SLACK_MESSAGE: "Instance teardown (cuda-${{ inputs.profile }}-benchmarks) finished with status: ${{ job.status }}. (${{ env.ACTION_RUN_URL }})"

View File

@@ -0,0 +1,60 @@
# Run CUDA benchmarks on multiple Hyperstack VMs and return parsed results to Slab CI bot.
name: Cuda weekly benchmarks
on:
schedule:
# Weekly benchmarks will be triggered each Saturday at 1a.m.
- cron: '0 1 * * 6'
jobs:
run-benchmarks-1-h100:
name: Run benchmarks (1xH100)
if: github.repository == 'zama-ai/tfhe-rs'
uses: ./.github/workflows/benchmark_gpu_integer_common.yml
with:
profile: single-h100
hardware_name: n3-H100x1
command: integer,integer_multi_bit
op_flavor: default
bench_type: latency
all_precisions: true
secrets: inherit
run-benchmarks-2-h100:
name: Run benchmarks (2xH100)
if: github.repository == 'zama-ai/tfhe-rs'
uses: ./.github/workflows/benchmark_gpu_integer_common.yml
with:
profile: 2-h100
hardware_name: n3-H100x2
command: integer_multi_bit
op_flavor: default
bench_type: latency
all_precisions: true
secrets: inherit
run-benchmarks-8-h100:
name: Run benchmarks (8xH100)
if: github.repository == 'zama-ai/tfhe-rs'
uses: ./.github/workflows/benchmark_gpu_integer_common.yml
with:
profile: multi-h100
hardware_name: n3-H100x8
command: integer_multi_bit
op_flavor: default
bench_type: latency
all_precisions: true
secrets: inherit
run-benchmarks-l40:
name: Run benchmarks (L40)
if: github.repository == 'zama-ai/tfhe-rs'
uses: ./.github/workflows/benchmark_gpu_integer_common.yml
with:
profile: l40
hardware_name: n3-L40x1
command: integer_multi_bit,integer_compression,pbs,ks
op_flavor: default
bench_type: latency
all_precisions: true
secrets: inherit

View File

@@ -1,34 +0,0 @@
# Run CUDA benchmarks on Hyperstack VM and return parsed results to Slab CI bot.
name: benchmark_gpu_weekly
run-name: GPU weekly benchmarks
on:
schedule:
# Weekly benchmarks will be triggered each Saturday at 1a.m.
- cron: '0 1 * * 6'
permissions: {}
# zizmor: ignore[concurrency-limits] only GitHub can trigger this workflow
jobs:
run-benchmarks-8-h100-sxm5-summary:
name: benchmark_gpu_weekly/run-benchmarks-8-h100-sxm5-summary
if: github.repository == 'zama-ai/tfhe-rs'
uses: ./.github/workflows/benchmark_gpu_common.yml
with:
profile: multi-h100-sxm5
hardware_name: n3-H100-SXM5x8
command: summary
bench_type: both
params_type: classical + multi_bit
secrets:
BOT_USERNAME: ${{ secrets.BOT_USERNAME }}
SLACK_CHANNEL: ${{ secrets.SLACK_CHANNEL }}
SLACK_WEBHOOK: ${{ secrets.SLACK_WEBHOOK }}
REPO_CHECKOUT_TOKEN: ${{ secrets.REPO_CHECKOUT_TOKEN }}
JOB_SECRET: ${{ secrets.JOB_SECRET }}
SLAB_ACTION_TOKEN: ${{ secrets.SLAB_ACTION_TOKEN }}
SLAB_URL: ${{ secrets.SLAB_URL }}
SLAB_BASE_URL: ${{ secrets.SLAB_BASE_URL }}

View File

@@ -1,72 +0,0 @@
# Run benchmarks on a permanent HPU instance and return parsed results to Slab CI bot.
name: benchmark_hpu
run-name: ${{ inputs.command }}::${{ inputs.bench_type}} (${{ inputs.op_flavor }}, ${{ inputs.precisions_set }})
on:
workflow_dispatch:
inputs:
command:
description: "Benchmark command to run"
type: choice
default: integer
options:
- integer
- hlapi_unsigned
- hlapi_signed
- hlapi_erc7984
op_flavor:
description: "Operations set to run"
type: choice
default: default
options:
- default
- fast_default
precisions_set:
description: "Bit precisions set"
type: choice
default: fast
options:
- fast
- all
- documentation
bench_type:
description: "Benchmarks type"
type: choice
default: latency
options:
- latency
- throughput
- both
v80_pcie_dev:
description: "V80 PCIe device number"
default: 24
v80_serial_number:
description: "V80 serial number"
default: XFL12NWY3ZKG
permissions: {}
# zizmor: ignore[concurrency-limits] only Zama organization members can trigger this workflow
jobs:
run-benchmarks:
name: benchmark_hpu/run-benchmarks
uses: ./.github/workflows/benchmark_hpu_common.yml
with:
command: ${{ inputs.command }}
op_flavor: ${{ inputs.op_flavor }}
bench_type: ${{ inputs.bench_type }}
precisions_set: ${{ inputs.precisions_set }}
v80_pcie_dev: ${{ inputs.v80_pcie_dev }}
v80_serial_number: ${{ inputs.v80_serial_number }}
secrets:
BOT_USERNAME: ${{ secrets.BOT_USERNAME }}
SLACK_CHANNEL: ${{ secrets.SLACK_CHANNEL }}
SLACK_WEBHOOK: ${{ secrets.SLACK_WEBHOOK }}
REPO_CHECKOUT_TOKEN: ${{ secrets.REPO_CHECKOUT_TOKEN }}
JOB_SECRET: ${{ secrets.JOB_SECRET }}
SLAB_ACTION_TOKEN: ${{ secrets.SLAB_ACTION_TOKEN }}
SLAB_URL: ${{ secrets.SLAB_URL }}
SLAB_BASE_URL: ${{ secrets.SLAB_BASE_URL }}
SSH_PRIVATE_KEY: ${{ secrets.SSH_PRIVATE_KEY }}

View File

@@ -1,208 +0,0 @@
# Run benchmarks on a permanent HPU instance and return parsed results to Slab CI bot.
name: benchmark_hpu_common
on:
workflow_call:
inputs:
command: # Use a comma separated values to generate an array
type: string
required: true
op_flavor: # Use a comma separated values to generate an array
type: string
default: default
bench_type:
type: string
default: latency
precisions_set:
type: string
default: fast
v80_pcie_dev:
type: string
default: 24
v80_serial_number:
type: string
default: XFL12NWY3ZKG
secrets:
REPO_CHECKOUT_TOKEN:
required: true
SLAB_ACTION_TOKEN:
required: true
SLAB_BASE_URL:
required: true
SLAB_URL:
required: true
JOB_SECRET:
required: true
SLACK_CHANNEL:
required: true
BOT_USERNAME:
required: true
SLACK_WEBHOOK:
required: true
SSH_PRIVATE_KEY:
required: true
env:
CARGO_TERM_COLOR: always
RESULTS_FILENAME: parsed_benchmark_results_${{ github.sha }}.json
ACTION_RUN_URL: ${{ github.server_url }}/${{ github.repository }}/actions/runs/${{ github.run_id }}
RUST_BACKTRACE: "full"
RUST_MIN_STACK: "8388608"
permissions: {}
# zizmor: ignore[concurrency-limits] caller workflow is responsible for the concurrency
jobs:
prepare-matrix:
name: benchmark_hpu_common/prepare-matrix
runs-on: ubuntu-latest
outputs:
command: ${{ steps.set_matrix_args.outputs.command }}
op_flavor: ${{ steps.set_matrix_args.outputs.op_flavor }}
bench_type: ${{ steps.set_matrix_args.outputs.bench_type }}
env:
INPUTS_COMMAND: ${{ inputs.command }}
INPUTS_OP_FLAVOR: ${{ inputs.op_flavor }}
steps:
- name: Parse user inputs
shell: python
env:
INPUTS_COMMAND: ${{ inputs.command }}
INPUTS_OP_FLAVOR: ${{ inputs.op_flavor }}
INPUTS_BENCH_TYPE: ${{ inputs.bench_type }}
run: |
import os
inputs_command = os.environ["INPUTS_COMMAND"]
inputs_op_flavor = os.environ["INPUTS_OP_FLAVOR"]
inputs_bench_type = os.environ["INPUTS_BENCH_TYPE"]
env_file = os.environ["GITHUB_ENV"]
split_command = inputs_command.replace(" ", "").split(",")
split_op_flavor = inputs_op_flavor.replace(" ", "").split(",")
if inputs_bench_type == "both":
bench_type = ["latency", "throughput"]
else:
bench_type = [inputs_bench_type, ]
with open(env_file, "a") as f:
for env_name, values_to_join in [
("COMMAND", split_command),
("OP_FLAVOR", split_op_flavor),
("BENCH_TYPE", bench_type),
]:
f.write(f"""{env_name}=["{'", "'.join(values_to_join)}"]\n""")
- name: Set matrix arguments outputs
id: set_matrix_args
run: | # zizmor: ignore[template-injection] these env variable are safe
{
echo "command=${{ toJSON(env.COMMAND) }}";
echo "op_flavor=${{ toJSON(env.OP_FLAVOR) }}";
echo "bench_type=${{ toJSON(env.BENCH_TYPE) }}";
} >> "${GITHUB_OUTPUT}"
hpu-benchmarks:
name: benchmark_hpu_common/hpu-benchmarks
needs: prepare-matrix
runs-on: v80-marais
concurrency:
group: ${{ github.workflow }}_${{ github.ref }}
cancel-in-progress: ${{ github.ref != 'refs/heads/main' }}
timeout-minutes: 1440 # 24 hours
strategy:
max-parallel: 1
matrix:
command: ${{ fromJSON(needs.prepare-matrix.outputs.command) }}
op_flavor: ${{ fromJSON(needs.prepare-matrix.outputs.op_flavor) }}
bench_type: ${{ fromJSON(needs.prepare-matrix.outputs.bench_type) }}
steps:
# Needed as long as hw_regmap repository is private
- name: Configure SSH
uses: webfactory/ssh-agent@e83874834305fe9a4a2997156cb26c5de65a8555 # v0.10.0
with:
ssh-private-key: ${{ secrets.SSH_PRIVATE_KEY }}
- name: Checkout tfhe-rs repo with tags
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd
with:
fetch-depth: 0
persist-credentials: 'false'
lfs: true
token: ${{ secrets.REPO_CHECKOUT_TOKEN }}
- name: Get benchmark details
run: |
COMMIT_DATE=$(git --no-pager show -s --format=%cd --date=iso8601-strict "${SHA}");
{
echo "BENCH_DATE=$(date --iso-8601=seconds)";
echo "COMMIT_DATE=${COMMIT_DATE}";
echo "COMMIT_HASH=$(git describe --tags --dirty)";
} >> "${GITHUB_ENV}"
env:
SHA: ${{ github.sha }}
- name: Install rust
uses: dtolnay/rust-toolchain@e97e2d8cc328f1b50210efc529dca0028893a2d9 # zizmor: ignore[stale-action-refs] this action doesn't create releases
with:
toolchain: nightly
- name: Select HPU board
run: |
echo "V80_PCIE_DEV=${PCIE_DEV}" >> "${GITHUB_ENV}"
echo "V80_SERIAL_NUMBER=${SERIAL_NUMBER}" >> "${GITHUB_ENV}"
env:
PCIE_DEV: ${{ inputs.v80_pcie_dev }}
SERIAL_NUMBER: ${{ inputs.v80_serial_number }}
- name: Run benchmarks
run: |
echo "${V80_PCIE_DEV} ${V80_SERIAL_NUMBER}"
make pull_hpu_files
make BIT_SIZES_SET="${PRECISIONS_SET}" BENCH_OP_FLAVOR="${OP_FLAVOR}" BENCH_TYPE="${BENCH_TYPE}" BENCH_PARAM_TYPE="${BENCH_PARAMS_TYPE}" bench_"${BENCH_COMMAND}"_hpu
env:
OP_FLAVOR: ${{ matrix.op_flavor }}
BENCH_TYPE: ${{ matrix.bench_type }}
BENCH_COMMAND: ${{ matrix.command }}
PRECISIONS_SET: ${{ inputs.precisions_set }}
- name: Parse results
run: |
python3 ./ci/benchmark_parser.py target/criterion "${RESULTS_FILENAME}" \
--database tfhe_rs \
--hardware "hpu_x1" \
--backend hpu \
--project-version "${COMMIT_HASH}" \
--branch "${REF_NAME}" \
--commit-date "${COMMIT_DATE}" \
--bench-date "${BENCH_DATE}" \
--walk-subdirs \
--bench-type "${BENCH_TYPE}"
env:
REF_NAME: ${{ github.ref_name }}
BENCH_TYPE: ${{ matrix.bench_type }}
- name: Upload parsed results artifact
uses: actions/upload-artifact@043fb46d1a93c77aae656e7c1c64a875d1fc6a0a
with:
name: ${{ github.sha }}_${{ matrix.bench_type }}_${{ matrix.command }}_benchmarks
path: ${{ env.RESULTS_FILENAME }}
- name: Checkout Slab repo
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd
with:
repository: zama-ai/slab
path: slab
persist-credentials: 'false'
token: ${{ secrets.REPO_CHECKOUT_TOKEN }}
- name: Send data to Slab
shell: bash
run: |
python3 slab/scripts/data_sender.py "${RESULTS_FILENAME}" "${JOB_SECRET}" \
--slab-url "${SLAB_URL}"
env:
JOB_SECRET: ${{ secrets.JOB_SECRET }}
SLAB_URL: ${{ secrets.SLAB_URL }}

214
.github/workflows/benchmark_integer.yml vendored Normal file
View File

@@ -0,0 +1,214 @@
# Run all integer benchmarks on an AWS instance and return parsed results to Slab CI bot.
name: Integer benchmarks
on:
workflow_dispatch:
inputs:
all_precisions:
description: "Run all precisions"
type: boolean
default: false
bench_type:
description: "Benchmarks type"
type: choice
default: latency
options:
- latency
- throughput
- both
schedule:
# Weekly benchmarks will be triggered each Saturday at 1a.m.
- cron: '0 1 * * 6'
# Quarterly benchmarks will be triggered right before end of quarter, the 25th of the current month at 4a.m.
# These benchmarks are far longer to execute hence the reason to run them only four time a year.
- cron: '0 4 25 MAR,JUN,SEP,DEC *'
env:
CARGO_TERM_COLOR: always
RESULTS_FILENAME: parsed_benchmark_results_${{ github.sha }}.json
ACTION_RUN_URL: ${{ github.server_url }}/${{ github.repository }}/actions/runs/${{ github.run_id }}
RUST_BACKTRACE: "full"
RUST_MIN_STACK: "8388608"
SLACK_CHANNEL: ${{ secrets.SLACK_CHANNEL }}
SLACK_ICON: https://pbs.twimg.com/profile_images/1274014582265298945/OjBKP9kn_400x400.png
SLACK_USERNAME: ${{ secrets.BOT_USERNAME }}
SLACK_WEBHOOK: ${{ secrets.SLACK_WEBHOOK }}
FAST_BENCH: TRUE
jobs:
prepare-matrix:
name: Prepare operations matrix
runs-on: ubuntu-latest
if: github.event_name != 'schedule' ||
(github.event_name == 'schedule' && github.repository == 'zama-ai/tfhe-rs')
outputs:
op_flavor: ${{ steps.set_op_flavor.outputs.op_flavor }}
bench_type: ${{ steps.set_bench_type.outputs.bench_type }}
steps:
- name: Weekly benchmarks
if: github.event.schedule == '0 1 * * 6'
run: |
echo "OP_FLAVOR=[\"default\"]" >> "${GITHUB_ENV}"
- name: Quarterly benchmarks
if: github.event.schedule == '0 4 25 MAR,JUN,SEP,DEC *'
run: |
echo "OP_FLAVOR=[\"default\", \"smart\", \"unchecked\", \"misc\"]" >> "${GITHUB_ENV}"
- name: Set benchmark types
if: github.event_name == 'workflow_dispatch'
run: |
echo "OP_FLAVOR=[\"default\"]" >> "${GITHUB_ENV}"
if [[ "${{ inputs.bench_type }}" == "both" ]]; then
echo "BENCH_TYPE=[\"latency\", \"throughput\"]" >> "${GITHUB_ENV}"
else
echo "BENCH_TYPE=[\"${{ inputs.bench_type }}\"]" >> "${GITHUB_ENV}"
fi
- name: Default benchmark type
if: github.event_name != 'workflow_dispatch'
run: |
echo "BENCH_TYPE=[\"latency\"]" >> "${GITHUB_ENV}"
- name: Set operation flavor output
id: set_op_flavor
run: |
echo "op_flavor=${{ toJSON(env.OP_FLAVOR) }}" >> "${GITHUB_OUTPUT}"
- name: Set benchmark types output
id: set_bench_type
run: |
echo "bench_type=${{ toJSON(env.BENCH_TYPE) }}" >> "${GITHUB_OUTPUT}"
setup-instance:
name: Setup instance (integer-benchmarks)
needs: prepare-matrix
runs-on: ubuntu-latest
outputs:
runner-name: ${{ steps.start-instance.outputs.label }}
steps:
- name: Start instance
id: start-instance
uses: zama-ai/slab-github-runner@98f0788261a7323d5d695a883e20df36591a92b7
with:
mode: start
github-token: ${{ secrets.SLAB_ACTION_TOKEN }}
slab-url: ${{ secrets.SLAB_BASE_URL }}
job-secret: ${{ secrets.JOB_SECRET }}
backend: aws
profile: bench
integer-benchmarks:
name: Execute integer benchmarks for all operations flavor
needs: [ prepare-matrix, setup-instance ]
runs-on: ${{ needs.setup-instance.outputs.runner-name }}
concurrency:
group: ${{ github.workflow }}_${{ github.ref }}
cancel-in-progress: ${{ github.ref != 'refs/heads/main' }}
continue-on-error: true
timeout-minutes: 1440 # 24 hours
strategy:
max-parallel: 1
matrix:
command: [ integer, integer_multi_bit]
op_flavor: ${{ fromJson(needs.prepare-matrix.outputs.op_flavor) }}
bench_type: ${{ fromJSON(needs.prepare-matrix.outputs.bench_type) }}
steps:
- name: Checkout tfhe-rs repo with tags
uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683
with:
fetch-depth: 0
token: ${{ secrets.FHE_ACTIONS_TOKEN }}
- name: Get benchmark details
run: |
{
echo "BENCH_DATE=$(date --iso-8601=seconds)";
echo "COMMIT_DATE=$(git --no-pager show -s --format=%cd --date=iso8601-strict ${{ github.sha }})";
echo "COMMIT_HASH=$(git describe --tags --dirty)";
} >> "${GITHUB_ENV}"
- name: Install rust
uses: dtolnay/rust-toolchain@315e265cd78dad1e1dcf3a5074f6d6c47029d5aa
with:
toolchain: nightly
- name: Checkout Slab repo
uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683
with:
repository: zama-ai/slab
path: slab
token: ${{ secrets.FHE_ACTIONS_TOKEN }}
- name: Should run benchmarks with all precisions
if: inputs.all_precisions
run: |
echo "FAST_BENCH=FALSE" >> "${GITHUB_ENV}"
- name: Run benchmarks with AVX512
run: |
make BENCH_OP_FLAVOR=${{ matrix.op_flavor }} BENCH_TYPE=${{ matrix.bench_type }} bench_${{ matrix.command }}
# Run these benchmarks only once per benchmark type
- name: Run compression benchmarks with AVX512
if: matrix.op_flavor == 'default' && matrix.command == 'integer'
run: |
make BENCH_TYPE=${{ matrix.bench_type }} bench_integer_compression
- name: Parse results
run: |
python3 ./ci/benchmark_parser.py target/criterion ${{ env.RESULTS_FILENAME }} \
--database tfhe_rs \
--hardware "hpc7a.96xlarge" \
--project-version "${{ env.COMMIT_HASH }}" \
--branch ${{ github.ref_name }} \
--commit-date "${{ env.COMMIT_DATE }}" \
--bench-date "${{ env.BENCH_DATE }}" \
--walk-subdirs \
--name-suffix avx512 \
--bench-type ${{ matrix.bench_type }}
- name: Upload parsed results artifact
uses: actions/upload-artifact@b4b15b8c7c6ac21ea08fcf65892d2ee8f75cf882
with:
name: ${{ github.sha }}_${{ matrix.command }}_${{ matrix.op_flavor }}_${{ matrix.bench_type }}
path: ${{ env.RESULTS_FILENAME }}
- name: Send data to Slab
shell: bash
run: |
python3 slab/scripts/data_sender.py ${{ env.RESULTS_FILENAME }} "${{ secrets.JOB_SECRET }}" \
--slab-url "${{ secrets.SLAB_URL }}"
- name: Slack Notification
if: ${{ failure() }}
continue-on-error: true
uses: rtCamp/action-slack-notify@c33737706dea87cd7784c687dadc9adf1be59990
env:
SLACK_COLOR: ${{ job.status }}
SLACK_MESSAGE: "Integer full benchmarks finished with status: ${{ job.status }}. (${{ env.ACTION_RUN_URL }})"
teardown-instance:
name: Teardown instance (integer-benchmarks)
if: ${{ always() && needs.setup-instance.result != 'skipped' }}
needs: [ setup-instance, integer-benchmarks ]
runs-on: ubuntu-latest
steps:
- name: Stop instance
id: stop-instance
uses: zama-ai/slab-github-runner@98f0788261a7323d5d695a883e20df36591a92b7
with:
mode: stop
github-token: ${{ secrets.SLAB_ACTION_TOKEN }}
slab-url: ${{ secrets.SLAB_BASE_URL }}
job-secret: ${{ secrets.JOB_SECRET }}
label: ${{ needs.setup-instance.outputs.runner-name }}
- name: Slack Notification
if: ${{ failure() }}
continue-on-error: true
uses: rtCamp/action-slack-notify@c33737706dea87cd7784c687dadc9adf1be59990
env:
SLACK_COLOR: ${{ job.status }}
SLACK_MESSAGE: "Instance teardown (integer-benchmarks) finished with status: ${{ job.status }}. (${{ env.ACTION_RUN_URL }})"

View File

@@ -1,403 +0,0 @@
# Run performance regression benchmarks and return parsed results to associated pull-request.
name: benchmark_perf_regression
on:
issue_comment:
types: [ created ]
pull_request:
types: [ labeled ]
env:
CARGO_TERM_COLOR: always
RESULTS_FILENAME: parsed_benchmark_results_${{ github.sha }}.json
ACTION_RUN_URL: ${{ github.server_url }}/${{ github.repository }}/actions/runs/${{ github.run_id }}
RUST_BACKTRACE: "full"
RUST_MIN_STACK: "8388608"
SLACK_CHANNEL: ${{ secrets.SLACK_CHANNEL }}
SLACK_ICON: https://pbs.twimg.com/profile_images/1274014582265298945/OjBKP9kn_400x400.png
SLACK_USERNAME: ${{ secrets.BOT_USERNAME }}
SLACK_WEBHOOK: ${{ secrets.SLACK_WEBHOOK }}
permissions: { }
# zizmor: ignore[concurrency-limits] only Zama organization members can trigger this workflow
jobs:
verify-triggering-actor:
name: benchmark_perf_regression/verify-actor
if: (github.event_name == 'pull_request' &&
(contains(github.event.label.name, 'bench-perfs-cpu') ||
contains(github.event.label.name, 'bench-perfs-gpu'))) ||
(github.event.issue.pull_request && startsWith(github.event.comment.body, '/bench'))
uses: ./.github/workflows/verify_triggering_actor.yml
secrets:
ALLOWED_TEAM: ${{ secrets.RELEASE_TEAM }}
READ_ORG_TOKEN: ${{ secrets.READ_ORG_TOKEN }}
prepare-benchmarks:
name: benchmark_perf_regression/prepare-benchmarks
needs: verify-triggering-actor
runs-on: ubuntu-latest
outputs:
commands: ${{ steps.set_commands.outputs.commands }}
slab-backend: ${{ steps.set_slab_details.outputs.backend }}
slab-profile: ${{ steps.set_slab_details.outputs.profile }}
hardware-name: ${{ steps.get_hardware_name.outputs.name }}
tfhe-backend: ${{ steps.set_regression_details.outputs.tfhe-backend }}
selected-regression-profile: ${{ steps.set_regression_details.outputs.selected-profile }}
custom-env: ${{ steps.get_custom_env.outputs.custom_env }}
permissions:
pull-requests: write # Needed to write a comment in a pull-request
steps:
- name: Checkout tfhe-rs repo
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd
with:
persist-credentials: 'false'
token: ${{ secrets.REPO_CHECKOUT_TOKEN }}
- name: Acknowledge issue comment
if: github.event_name == 'issue_comment'
uses: peter-evans/create-or-update-comment@e8674b075228eee787fea43ef493e45ece1004c9 # v5.0.0
with:
comment-id: ${{ github.event.comment.id }}
reactions: '+1'
- name: Display workflow run URL
if: github.event_name == 'issue_comment'
uses: peter-evans/create-or-update-comment@e8674b075228eee787fea43ef493e45ece1004c9 # v5.0.0
with:
issue-number: ${{ github.event.issue.number }}
body: |
User triggered performance regression benchmark.
Workflow run URL: ${{ env.ACTION_RUN_URL }}
- name: Generate CPU benchmarks command from label
if: (github.event_name == 'pull_request' && contains(github.event.label.name, 'bench-perfs-cpu'))
run: |
echo "DEFAULT_BENCH_OPTIONS=--backend cpu" >> "${GITHUB_ENV}"
- name: Generate GPU benchmarks command from label
if: (github.event_name == 'pull_request' && contains(github.event.label.name, 'bench-perfs-gpu'))
run: |
echo "DEFAULT_BENCH_OPTIONS=--backend gpu" >> "${GITHUB_ENV}"
# TODO add support for HPU backend
- name: Install Python requirements
run: |
python3 -m pip install -r ci/perf_regression/requirements.txt
- name: Generate cargo commands and env from label
if: github.event_name == 'pull_request'
run: |
python3 ci/perf_regression/perf_regression.py parse_profile --issue-comment "/bench ${DEFAULT_BENCH_OPTIONS}"
echo "COMMANDS=$(cat ci/perf_regression/perf_regression_generated_commands.json)" >> "${GITHUB_ENV}"
- name: Dump issue comment into file # To avoid possible code-injection
if: github.event_name == 'issue_comment'
run: |
echo "${COMMENT_BODY}" >> dumped_comment.txt
env:
COMMENT_BODY: ${{ github.event.comment.body }}
- name: Generate cargo commands and env
if: github.event_name == 'issue_comment'
run: |
python3 ci/perf_regression/perf_regression.py parse_profile --issue-comment "$(cat dumped_comment.txt)"
echo "COMMANDS=$(cat ci/perf_regression/perf_regression_generated_commands.json)" >> "${GITHUB_ENV}"
- name: Set commands output
id: set_commands
run: | # zizmor: ignore[template-injection] this env variable is safe
echo "commands=${{ toJSON(env.COMMANDS) }}" >> "${GITHUB_OUTPUT}"
- name: Set Slab details outputs
id: set_slab_details
run: |
echo "backend=$(cat ci/perf_regression/perf_regression_slab_backend_config.txt)" >> "${GITHUB_OUTPUT}"
echo "profile=$(cat ci/perf_regression/perf_regression_slab_profile_config.txt)" >> "${GITHUB_OUTPUT}"
- name: Get hardware name
id: get_hardware_name
run: | # zizmor: ignore[template-injection] these interpolations are safe
HARDWARE_NAME=$(python3 ci/hardware_finder.py "${{ steps.set_slab_details.outputs.backend }}" "${{ steps.set_slab_details.outputs.profile }}");
echo "name=${HARDWARE_NAME}" >> "${GITHUB_OUTPUT}"
- name: Set regression details outputs
id: set_regression_details
run: |
echo "tfhe-backend=$(cat ci/perf_regression/perf_regression_tfhe_rs_backend_config.txt)" >> "${GITHUB_OUTPUT}"
echo "selected-profile=$(cat ci/perf_regression/perf_regression_selected_profile_config.txt)" >> "${GITHUB_OUTPUT}"
- name: Get custom env vars
id: get_custom_env
run: |
echo "custom_env=$(cat ci/perf_regression/perf_regression_custom_env.sh)" >> "${GITHUB_OUTPUT}"
setup-instance:
name: benchmark_perf_regression/setup-instance
needs: prepare-benchmarks
runs-on: ubuntu-latest
outputs:
runner-name: ${{ steps.start-instance.outputs.label }}
steps:
- name: Start instance
id: start-instance
uses: zama-ai/slab-github-runner@5aee5d157f4a0201e5eaefc9cc648e5f9f5472a5 # v1.6.0
with:
mode: start
github-token: ${{ secrets.SLAB_ACTION_TOKEN }}
slab-url: ${{ secrets.SLAB_BASE_URL }}
job-secret: ${{ secrets.JOB_SECRET }}
backend: ${{ needs.prepare-benchmarks.outputs.slab-backend }}
profile: ${{ needs.prepare-benchmarks.outputs.slab-profile }}
install-cuda-dependencies-if-required:
name: benchmark_perf_regression/install-cuda-dependencies-if-required
needs: [ prepare-benchmarks, setup-instance ]
runs-on: ${{ needs.setup-instance.outputs.runner-name }}
strategy:
matrix:
# explicit include-based build matrix, of known valid options
include:
- cuda: "12.8"
gcc: 11
steps:
- name: Checkout tfhe-rs repo
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd
with:
persist-credentials: 'false'
token: ${{ secrets.REPO_CHECKOUT_TOKEN }}
- name: Setup Hyperstack dependencies
if: needs.prepare-benchmarks.outputs.slab-backend == 'hyperstack'
uses: ./.github/actions/gpu_setup
with:
cuda-version: ${{ matrix.cuda }}
gcc-version: ${{ matrix.gcc }}
regression-benchmarks:
name: benchmark_perf_regression/regression-benchmarks
needs: [ prepare-benchmarks, setup-instance, install-cuda-dependencies-if-required ]
runs-on: ${{ needs.setup-instance.outputs.runner-name }}
concurrency:
group: ${{ github.workflow_ref }}_${{ needs.prepare-benchmarks.outputs.slab-backend }}_${{ needs.prepare-benchmarks.outputs.slab-profile }}
cancel-in-progress: true
timeout-minutes: 720 # 12 hours
strategy:
fail-fast: false
max-parallel: 1
matrix:
command: ${{ fromJson(needs.prepare-benchmarks.outputs.commands) }}
steps:
- name: Checkout tfhe-rs repo
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd
with:
fetch-depth: 0 # Needed to get commit hash
persist-credentials: 'false'
token: ${{ secrets.REPO_CHECKOUT_TOKEN }}
- name: Get benchmark details
run: |
COMMIT_DATE=$(git --no-pager show -s --format=%cd --date=iso8601-strict "${SHA}");
{
echo "BENCH_DATE=$(date --iso-8601=seconds)";
echo "COMMIT_DATE=${COMMIT_DATE}";
echo "COMMIT_HASH=$(git describe --tags --dirty)";
} >> "${GITHUB_ENV}"
env:
SHA: ${{ github.sha }}
- name: Export custom env variables
run: | # zizmor: ignore[template-injection] this env variable is safe
{
${{ needs.prepare-benchmarks.outputs.custom-env }}
} >> "$GITHUB_ENV"
# Re-export environment variables as dependencies setup perform this task in the previous job.
# Local env variables are cleaned at the end of each job.
- name: Export CUDA variables
if: needs.prepare-benchmarks.outputs.slab-backend == 'hyperstack'
shell: bash
run: |
echo "CUDA_PATH=$CUDA_PATH" >> "${GITHUB_ENV}"
echo "PATH=$PATH:$CUDA_PATH/bin" >> "${GITHUB_PATH}"
echo "LD_LIBRARY_PATH=$CUDA_PATH/lib64:$LD_LIBRARY_PATH" >> "${GITHUB_ENV}"
echo "CUDA_MODULE_LOADER=EAGER" >> "${GITHUB_ENV}"
env:
CUDA_PATH: /usr/local/cuda-12.8
- name: Export gcc and g++ variables
if: needs.prepare-benchmarks.outputs.slab-backend == 'hyperstack'
shell: bash
run: |
{
echo "CC=/usr/bin/gcc-${GCC_VERSION}";
echo "CXX=/usr/bin/g++-${GCC_VERSION}";
echo "CUDAHOSTCXX=/usr/bin/g++-${GCC_VERSION}";
} >> "${GITHUB_ENV}"
env:
GCC_VERSION: 11
- name: Install rust
uses: dtolnay/rust-toolchain@b3b07ba8b418998c39fb20f53e8b695cdcc8de1b # zizmor: ignore[stale-action-refs] this action doesn't create releases
with:
toolchain: nightly
- name: Checkout Slab repo
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd
with:
repository: zama-ai/slab
path: slab
persist-credentials: 'false'
token: ${{ secrets.REPO_CHECKOUT_TOKEN }}
- name: Run regression benchmarks
run: |
make BENCH_CUSTOM_COMMAND="${BENCH_COMMAND}" bench_custom
env:
BENCH_COMMAND: ${{ matrix.command }}
- name: Parse results
run: |
python3 ./ci/benchmark_parser.py target/criterion "${RESULTS_FILENAME}" \
--database tfhe_rs \
--hardware "${HARDWARE_NAME}" \
--backend "${TFHE_BACKEND}" \
--project-version "${COMMIT_HASH}" \
--branch "${REF_NAME}" \
--commit-date "${COMMIT_DATE}" \
--bench-date "${BENCH_DATE}" \
--walk-subdirs \
--name-suffix regression \
--bench-type "${BENCH_TYPE}"
echo "RESULTS_FILE_SHA=$(sha256sum "${RESULTS_FILENAME}" | cut -d " " -f1)" >> "${GITHUB_ENV}"
env:
HARDWARE_NAME: ${{ needs.prepare-benchmarks.outputs.hardware-name }}
TFHE_BACKEND: ${{ needs.prepare-benchmarks.outputs.tfhe-backend }}
REF_NAME: ${{ github.head_ref || github.ref_name }}
BENCH_TYPE: ${{ env.__TFHE_RS_BENCH_TYPE }}
- name: Upload parsed results artifact
uses: actions/upload-artifact@043fb46d1a93c77aae656e7c1c64a875d1fc6a0a
with:
name: ${{ github.sha }}_regression_${{ env.RESULTS_FILE_SHA }} # RESULT_FILE_SHA is needed to avoid collision between matrix.command runs
path: ${{ env.RESULTS_FILENAME }}
- name: Send data to Slab
shell: bash
run: |
python3 slab/scripts/data_sender.py "${RESULTS_FILENAME}" "${JOB_SECRET}" \
--slab-url "${SLAB_URL}"
env:
JOB_SECRET: ${{ secrets.JOB_SECRET }}
SLAB_URL: ${{ secrets.SLAB_URL }}
check-regressions:
name: benchmark_perf_regression/check-regressions
needs: [ prepare-benchmarks, regression-benchmarks ]
runs-on: ubuntu-latest
permissions:
pull-requests: write # Needed to write a comment in a pull-request
contents: read # Needed to set up Python dependencies
env:
REF_NAME: ${{ github.head_ref || github.ref_name }}
steps:
- name: Checkout tfhe-rs repo
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd
with:
persist-credentials: 'false'
token: ${{ secrets.REPO_CHECKOUT_TOKEN }}
- name: Install recent Python
uses: actions/setup-python@a309ff8b426b58ec0e2a45f0f869d46889d02405 # v6.2.0
with:
python-version: '3.12'
- name: Install requirements
run: |
python3.12 -m venv venv
venv/bin/pip3 install -r ci/data_extractor/requirements.txt -r ci/perf_regression/requirements.txt
- name: Fetch data
run: |
venv/bin/python3 ci/data_extractor/src/data_extractor.py regression_data \
--generate-regression-json \
--regression-profiles ci/regression.toml \
--regression-selected-profile "${REGRESSION_PROFILE}" \
--backend "${TFHE_BACKEND}" \
--hardware "${HARDWARE_NAME}" \
--branch "${REF_NAME}" \
--time-span-days 60
env:
REGRESSION_PROFILE: ${{ needs.prepare-benchmarks.outputs.selected-regression-profile }}
TFHE_BACKEND: ${{ needs.prepare-benchmarks.outputs.tfhe-backend }}
HARDWARE_NAME: ${{ needs.prepare-benchmarks.outputs.hardware-name }}
DATA_EXTRACTOR_DATABASE_HOST: ${{ secrets.DATA_EXTRACTOR_DATABASE_HOST }}
DATA_EXTRACTOR_DATABASE_USER: ${{ secrets.DATA_EXTRACTOR_DATABASE_USER }}
DATA_EXTRACTOR_DATABASE_PASSWORD: ${{ secrets.DATA_EXTRACTOR_DATABASE_PASSWORD }}
- name: Generate regression report
run: |
venv/bin/python3 ci/perf_regression/perf_regression.py check_regression \
--results-file regression_data.json \
--generate-report
- name: Write report in pull-request
uses: peter-evans/create-or-update-comment@e8674b075228eee787fea43ef493e45ece1004c9 # v5.0.0
with:
issue-number: ${{ github.event.pull_request.number || github.event.issue.number }}
body-path: ci/perf_regression/regression_report.md
comment-on-failure:
name: benchmark_perf_regression/comment-on-failure
needs: [ prepare-benchmarks, setup-instance, regression-benchmarks, check-regressions ]
runs-on: ubuntu-latest
if: ${{ failure() && github.event_name == 'issue_comment' }}
continue-on-error: true
permissions:
pull-requests: write # Needed to write a comment in a pull-request
steps:
- name: Write failure message
uses: peter-evans/create-or-update-comment@e8674b075228eee787fea43ef493e45ece1004c9 # v5.0.0
with:
issue-number: ${{ github.event.issue.number }}
body: |
:x: Performance regression benchmark failed ([workflow run](${{ env.ACTION_RUN_URL }}))
slack-notify:
name: benchmark_perf_regression/slack-notify
needs: [ prepare-benchmarks, setup-instance, regression-benchmarks, check-regressions ]
runs-on: ubuntu-latest
if: ${{ failure() }}
continue-on-error: true
steps:
- name: Send message
uses: rtCamp/action-slack-notify@e31e87e03dd19038e411e38ae27cbad084a90661
env:
SLACK_COLOR: failure
SLACK_MESSAGE: "Performance regression benchmarks failed. (${{ env.ACTION_RUN_URL }})"
teardown-instance:
name: benchmark_perf_regression/teardown-instance
if: ${{ always() && needs.setup-instance.result == 'success' }}
needs: [ setup-instance, regression-benchmarks ]
runs-on: ubuntu-latest
steps:
- name: Stop instance
id: stop-instance
uses: zama-ai/slab-github-runner@5aee5d157f4a0201e5eaefc9cc648e5f9f5472a5 # v1.6.0
with:
mode: stop
github-token: ${{ secrets.SLAB_ACTION_TOKEN }}
slab-url: ${{ secrets.SLAB_BASE_URL }}
job-secret: ${{ secrets.JOB_SECRET }}
label: ${{ needs.setup-instance.outputs.runner-name }}
- name: Slack Notification
if: ${{ failure() }}
uses: rtCamp/action-slack-notify@e31e87e03dd19038e411e38ae27cbad084a90661
env:
SLACK_COLOR: ${{ job.status }}
SLACK_MESSAGE: "Instance teardown (regression-benchmarks) finished with status: ${{ job.status }}. (${{ env.ACTION_RUN_URL }})"

180
.github/workflows/benchmark_shortint.yml vendored Normal file
View File

@@ -0,0 +1,180 @@
# Run all shortint benchmarks on an AWS instance and return parsed results to Slab CI bot.
name: Shortint full benchmarks
on:
workflow_dispatch:
schedule:
# Weekly benchmarks will be triggered each Saturday at 1a.m.
- cron: '0 1 * * 6'
# Quarterly benchmarks will be triggered right before end of quarter, the 25th of the current month at 4a.m.
# These benchmarks are far longer to execute hence the reason to run them only four time a year.
- cron: '0 4 25 MAR,JUN,SEP,DEC *'
env:
CARGO_TERM_COLOR: always
RESULTS_FILENAME: parsed_benchmark_results_${{ github.sha }}.json
ACTION_RUN_URL: ${{ github.server_url }}/${{ github.repository }}/actions/runs/${{ github.run_id }}
RUST_BACKTRACE: "full"
RUST_MIN_STACK: "8388608"
SLACK_CHANNEL: ${{ secrets.SLACK_CHANNEL }}
SLACK_ICON: https://pbs.twimg.com/profile_images/1274014582265298945/OjBKP9kn_400x400.png
SLACK_USERNAME: ${{ secrets.BOT_USERNAME }}
SLACK_WEBHOOK: ${{ secrets.SLACK_WEBHOOK }}
jobs:
prepare-matrix:
name: Prepare operations matrix
runs-on: ubuntu-latest
if: github.event_name != 'schedule' ||
(github.event_name == 'schedule' && github.repository == 'zama-ai/tfhe-rs')
outputs:
op_flavor: ${{ steps.set_op_flavor.outputs.op_flavor }}
steps:
- name: Weekly benchmarks
if: github.event_name == 'workflow_dispatch' ||
github.event.schedule == '0 1 * * 6'
run: |
echo "OP_FLAVOR=[\"default\"]" >> "${GITHUB_ENV}"
- name: Quarterly benchmarks
if: github.event.schedule == '0 4 25 MAR,JUN,SEP,DEC *'
run: |
echo "OP_FLAVOR=[\"default\", \"smart\", \"unchecked\"]" >> "${GITHUB_ENV}"
- name: Set operation flavor output
id: set_op_flavor
run: |
echo "op_flavor=${{ toJSON(env.OP_FLAVOR) }}" >> "${GITHUB_OUTPUT}"
setup-instance:
name: Setup instance (shortint-benchmarks)
needs: prepare-matrix
runs-on: ubuntu-latest
outputs:
runner-name: ${{ steps.start-instance.outputs.label }}
steps:
- name: Start instance
id: start-instance
uses: zama-ai/slab-github-runner@98f0788261a7323d5d695a883e20df36591a92b7
with:
mode: start
github-token: ${{ secrets.SLAB_ACTION_TOKEN }}
slab-url: ${{ secrets.SLAB_BASE_URL }}
job-secret: ${{ secrets.JOB_SECRET }}
backend: aws
profile: bench
shortint-benchmarks:
name: Execute shortint benchmarks for all operations flavor
needs: [ prepare-matrix, setup-instance ]
runs-on: ${{ needs.setup-instance.outputs.runner-name }}
concurrency:
group: ${{ github.workflow }}_${{ github.ref }}
cancel-in-progress: ${{ github.ref != 'refs/heads/main' }}
continue-on-error: true
strategy:
max-parallel: 1
matrix:
op_flavor: ${{ fromJson(needs.prepare-matrix.outputs.op_flavor) }}
steps:
- name: Checkout tfhe-rs repo with tags
uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683
with:
fetch-depth: 0
token: ${{ secrets.FHE_ACTIONS_TOKEN }}
- name: Get benchmark details
run: |
{
echo "BENCH_DATE=$(date --iso-8601=seconds)";
echo "COMMIT_DATE=$(git --no-pager show -s --format=%cd --date=iso8601-strict ${{ github.sha }})";
echo "COMMIT_HASH=$(git describe --tags --dirty)";
} >> "${GITHUB_ENV}"
- name: Install rust
uses: dtolnay/rust-toolchain@315e265cd78dad1e1dcf3a5074f6d6c47029d5aa
with:
toolchain: nightly
- name: Checkout Slab repo
uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683
with:
repository: zama-ai/slab
path: slab
token: ${{ secrets.FHE_ACTIONS_TOKEN }}
- name: Run benchmarks with AVX512
run: |
make BENCH_OP_FLAVOR=${{ matrix.op_flavor }} bench_shortint
- name: Parse results
run: |
COMMIT_DATE="$(git --no-pager show -s --format=%cd --date=iso8601-strict ${{ github.sha }})"
COMMIT_HASH="$(git describe --tags --dirty)"
python3 ./ci/benchmark_parser.py target/criterion ${{ env.RESULTS_FILENAME }} \
--database tfhe_rs \
--hardware "hpc7a.96xlarge" \
--project-version "${COMMIT_HASH}" \
--branch ${{ github.ref_name }} \
--commit-date "${COMMIT_DATE}" \
--bench-date "${{ env.BENCH_DATE }}" \
--walk-subdirs \
--name-suffix avx512
# This small benchmark needs to be executed only once.
- name: Measure key sizes
if: matrix.op_flavor == 'default'
run: |
make measure_shortint_key_sizes
- name: Parse key sizes results
if: matrix.op_flavor == 'default'
run: |
python3 ./ci/benchmark_parser.py tfhe/shortint_key_sizes.csv ${{ env.RESULTS_FILENAME }} \
--object-sizes \
--append-results
- name: Upload parsed results artifact
uses: actions/upload-artifact@b4b15b8c7c6ac21ea08fcf65892d2ee8f75cf882
with:
name: ${{ github.sha }}_shortint_${{ matrix.op_flavor }}
path: ${{ env.RESULTS_FILENAME }}
- name: Send data to Slab
shell: bash
run: |
python3 slab/scripts/data_sender.py ${{ env.RESULTS_FILENAME }} "${{ secrets.JOB_SECRET }}" \
--slab-url "${{ secrets.SLAB_URL }}"
- name: Slack Notification
if: ${{ failure() }}
continue-on-error: true
uses: rtCamp/action-slack-notify@c33737706dea87cd7784c687dadc9adf1be59990
env:
SLACK_COLOR: ${{ job.status }}
SLACK_MESSAGE: "Shortint full benchmarks finished with status: ${{ job.status }}. (${{ env.ACTION_RUN_URL }})"
teardown-instance:
name: Teardown instance (shortint-benchmarks)
if: ${{ always() && needs.setup-instance.result != 'skipped' }}
needs: [ setup-instance, shortint-benchmarks ]
runs-on: ubuntu-latest
steps:
- name: Stop instance
id: stop-instance
uses: zama-ai/slab-github-runner@98f0788261a7323d5d695a883e20df36591a92b7
with:
mode: stop
github-token: ${{ secrets.SLAB_ACTION_TOKEN }}
slab-url: ${{ secrets.SLAB_BASE_URL }}
job-secret: ${{ secrets.JOB_SECRET }}
label: ${{ needs.setup-instance.outputs.runner-name }}
- name: Slack Notification
if: ${{ failure() }}
continue-on-error: true
uses: rtCamp/action-slack-notify@c33737706dea87cd7784c687dadc9adf1be59990
env:
SLACK_COLOR: ${{ job.status }}
SLACK_MESSAGE: "Instance teardown (shortint-benchmarks) finished with status: ${{ job.status }}. (${{ env.ACTION_RUN_URL }})"

View File

@@ -0,0 +1,208 @@
# Run all signed integer benchmarks on an AWS instance and return parsed results to Slab CI bot.
name: Signed Integer full benchmarks
on:
workflow_dispatch:
inputs:
all_precisions:
description: "Run all precisions"
type: boolean
default: false
bench_type:
description: "Benchmarks type"
type: choice
default: latency
options:
- latency
- throughput
- both
schedule:
# Weekly benchmarks will be triggered each Saturday at 1a.m.
- cron: '0 1 * * 6'
# Quarterly benchmarks will be triggered right before end of quarter, the 25th of the current month at 4a.m.
# These benchmarks are far longer to execute hence the reason to run them only four time a year.
- cron: '0 4 25 MAR,JUN,SEP,DEC *'
env:
CARGO_TERM_COLOR: always
RESULTS_FILENAME: parsed_benchmark_results_${{ github.sha }}.json
ACTION_RUN_URL: ${{ github.server_url }}/${{ github.repository }}/actions/runs/${{ github.run_id }}
RUST_BACKTRACE: "full"
RUST_MIN_STACK: "8388608"
SLACK_CHANNEL: ${{ secrets.SLACK_CHANNEL }}
SLACK_ICON: https://pbs.twimg.com/profile_images/1274014582265298945/OjBKP9kn_400x400.png
SLACK_USERNAME: ${{ secrets.BOT_USERNAME }}
SLACK_WEBHOOK: ${{ secrets.SLACK_WEBHOOK }}
FAST_BENCH: TRUE
jobs:
prepare-matrix:
name: Prepare operations matrix
runs-on: ubuntu-latest
if: github.event_name != 'schedule' ||
(github.event_name == 'schedule' && github.repository == 'zama-ai/tfhe-rs')
outputs:
op_flavor: ${{ steps.set_op_flavor.outputs.op_flavor }}
bench_type: ${{ steps.set_bench_type.outputs.bench_type }}
steps:
- name: Weekly benchmarks
if: github.event.schedule == '0 1 * * 6'
run: |
echo "OP_FLAVOR=[\"default\"]" >> "${GITHUB_ENV}"
- name: Quarterly benchmarks
if: github.event.schedule == '0 4 25 MAR,JUN,SEP,DEC *'
run: |
echo "OP_FLAVOR=[\"default\", \"unchecked\"]" >> "${GITHUB_ENV}"
- name: Set benchmark types
if: github.event_name == 'workflow_dispatch'
run: |
echo "OP_FLAVOR=[\"default\"]" >> "${GITHUB_ENV}"
if [[ "${{ inputs.bench_type }}" == "both" ]]; then
echo "BENCH_TYPE=[\"latency\", \"throughput\"]" >> "${GITHUB_ENV}"
else
echo "BENCH_TYPE=[\"${{ inputs.bench_type }}\"]" >> "${GITHUB_ENV}"
fi
- name: Default benchmark type
if: github.event_name != 'workflow_dispatch'
run: |
echo "BENCH_TYPE=[\"latency\"]" >> "${GITHUB_ENV}"
- name: Set operation flavor output
id: set_op_flavor
run: |
echo "op_flavor=${{ toJSON(env.OP_FLAVOR) }}" >> "${GITHUB_OUTPUT}"
- name: Set benchmark types output
id: set_bench_type
run: |
echo "bench_type=${{ toJSON(env.BENCH_TYPE) }}" >> "${GITHUB_OUTPUT}"
setup-instance:
name: Setup instance (signed-integer-benchmarks)
needs: prepare-matrix
runs-on: ubuntu-latest
outputs:
runner-name: ${{ steps.start-instance.outputs.label }}
steps:
- name: Start instance
id: start-instance
uses: zama-ai/slab-github-runner@98f0788261a7323d5d695a883e20df36591a92b7
with:
mode: start
github-token: ${{ secrets.SLAB_ACTION_TOKEN }}
slab-url: ${{ secrets.SLAB_BASE_URL }}
job-secret: ${{ secrets.JOB_SECRET }}
backend: aws
profile: bench
signed-integer-benchmarks:
name: Execute signed integer benchmarks for all operations flavor
needs: [ prepare-matrix, setup-instance ]
runs-on: ${{ needs.setup-instance.outputs.runner-name }}
concurrency:
group: ${{ github.workflow }}_${{ github.ref }}
cancel-in-progress: ${{ github.ref != 'refs/heads/main' }}
continue-on-error: true
timeout-minutes: 1440 # 24 hours
strategy:
max-parallel: 1
matrix:
command: [ integer, integer_multi_bit ]
op_flavor: ${{ fromJSON(needs.prepare-matrix.outputs.op_flavor) }}
bench_type: ${{ fromJSON(needs.prepare-matrix.outputs.bench_type) }}
steps:
- name: Checkout tfhe-rs repo with tags
uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683
with:
fetch-depth: 0
token: ${{ secrets.FHE_ACTIONS_TOKEN }}
- name: Get benchmark details
run: |
{
echo "BENCH_DATE=$(date --iso-8601=seconds)";
echo "COMMIT_DATE=$(git --no-pager show -s --format=%cd --date=iso8601-strict ${{ github.sha }})";
echo "COMMIT_HASH=$(git describe --tags --dirty)";
} >> "${GITHUB_ENV}"
- name: Install rust
uses: dtolnay/rust-toolchain@315e265cd78dad1e1dcf3a5074f6d6c47029d5aa
with:
toolchain: nightly
- name: Checkout Slab repo
uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683
with:
repository: zama-ai/slab
path: slab
token: ${{ secrets.FHE_ACTIONS_TOKEN }}
- name: Should run benchmarks with all precisions
if: inputs.all_precisions
run: |
echo "FAST_BENCH=FALSE" >> "${GITHUB_ENV}"
- name: Run benchmarks with AVX512
run: |
make BENCH_OP_FLAVOR=${{ matrix.op_flavor }} BENCH_TYPE=${{ matrix.bench_type }} bench_signed_${{ matrix.command }}
- name: Parse results
run: |
python3 ./ci/benchmark_parser.py target/criterion ${{ env.RESULTS_FILENAME }} \
--database tfhe_rs \
--hardware "hpc7a.96xlarge" \
--project-version "${{ env.COMMIT_HASH }}" \
--branch ${{ github.ref_name }} \
--commit-date "${{ env.COMMIT_DATE }}" \
--bench-date "${{ env.BENCH_DATE }}" \
--walk-subdirs \
--name-suffix avx512 \
--bench-type ${{ matrix.bench_type }}
- name: Upload parsed results artifact
uses: actions/upload-artifact@b4b15b8c7c6ac21ea08fcf65892d2ee8f75cf882
with:
name: ${{ github.sha }}_${{ matrix.command }}_${{ matrix.op_flavor }}_${{ matrix.bench_type }}
path: ${{ env.RESULTS_FILENAME }}
- name: Send data to Slab
shell: bash
run: |
python3 slab/scripts/data_sender.py ${{ env.RESULTS_FILENAME }} "${{ secrets.JOB_SECRET }}" \
--slab-url "${{ secrets.SLAB_URL }}"
- name: Slack Notification
if: ${{ failure() }}
continue-on-error: true
uses: rtCamp/action-slack-notify@c33737706dea87cd7784c687dadc9adf1be59990
env:
SLACK_COLOR: ${{ job.status }}
SLACK_MESSAGE: "Signed integer full benchmarks finished with status: ${{ job.status }}. (${{ env.ACTION_RUN_URL }})"
teardown-instance:
name: Teardown instance (integer-benchmarks)
if: ${{ always() && needs.setup-instance.result != 'skipped' }}
needs: [ setup-instance, signed-integer-benchmarks ]
runs-on: ubuntu-latest
steps:
- name: Stop instance
id: stop-instance
uses: zama-ai/slab-github-runner@98f0788261a7323d5d695a883e20df36591a92b7
with:
mode: stop
github-token: ${{ secrets.SLAB_ACTION_TOKEN }}
slab-url: ${{ secrets.SLAB_BASE_URL }}
job-secret: ${{ secrets.JOB_SECRET }}
label: ${{ needs.setup-instance.outputs.runner-name }}
- name: Slack Notification
if: ${{ failure() }}
continue-on-error: true
uses: rtCamp/action-slack-notify@c33737706dea87cd7784c687dadc9adf1be59990
env:
SLACK_COLOR: ${{ job.status }}
SLACK_MESSAGE: "Instance teardown (signed-integer-benchmarks) finished with status: ${{ job.status }}. (${{ env.ACTION_RUN_URL }})"

View File

@@ -1,136 +0,0 @@
# Run all benchmarks displayed in the internal documentation.
name: benchmark_summary
run-name: Benchmark Summary
on:
workflow_dispatch:
inputs:
run-cpu-benchmarks:
description: "Run CPU benchmarks"
type: boolean
default: true
run-gpu-benchmarks:
description: "Run GPU benchmarks"
type: boolean
default: true
gpu-profile:
description: "GPU Instance type"
required: true
default: "multi-h100-sxm5 (n3-H100-SXM5x8)"
type: choice
options:
- "l40 (n3-L40x1)"
- "4-l40 (n3-L40x4)"
- "8-l40 (n3-L40x8)"
- "multi-a100-nvlink (n3-A100x8-NVLink)"
- "single-h100 (n3-H100x1)"
- "2-h100 (n3-H100x2)"
- "4-h100 (n3-H100x4)"
- "multi-h100 (n3-H100x8)"
- "multi-h100-nvlink (n3-H100x8-NVLink)"
- "multi-h100-sxm5 (n3-H100-SXM5x8)"
bench_type:
description: "Benchmarks type"
type: choice
default: both
options:
- latency
- throughput
- both
run-hpu-benchmarks:
description: "Run HPU benchmarks"
type: boolean
default: true
permissions: {}
# zizmor: ignore[concurrency-limits] only Zama organization members can trigger this workflow
jobs:
parse-gpu-inputs:
name: benchmark_summary/parse-gpu-inputs
if: inputs.run-gpu-benchmarks
runs-on: ubuntu-latest
outputs:
profile: ${{ steps.parse_profile.outputs.profile }}
hardware_name: ${{ steps.parse_hardware_name.outputs.name }}
env:
INPUTS_PROFILE: ${{ inputs.gpu-profile }}
steps:
- name: Parse profile
id: parse_profile
run: |
# Use Sed to extract a value from a string, this cannot be done with the ${variable//search/replace} pattern.
# shellcheck disable=SC2001
PROFILE=$(echo "${INPUTS_PROFILE}" | sed 's|\(.*\)[[:space:]](.*)|\1|')
echo "profile=${PROFILE}" >> "${GITHUB_OUTPUT}"
- name: Parse hardware name
id: parse_hardware_name
run: |
# Use Sed to extract a value from a string, this cannot be done with the ${variable//search/replace} pattern.
# shellcheck disable=SC2001
NAME=$(echo "${INPUTS_PROFILE}" | sed 's|.*[[:space:]](\(.*\))|\1|')
echo "name=${NAME}" >> "${GITHUB_OUTPUT}"
run-benchmarks-cpu:
name: benchmark_documentation/run-benchmarks-cpu-integer
uses: ./.github/workflows/benchmark_cpu_common.yml
if: inputs.run-cpu-benchmarks
with:
command: summary
bench_type: ${{ inputs.bench_type }}
params_type: classical + multi_bit
secrets:
BOT_USERNAME: ${{ secrets.BOT_USERNAME }}
SLACK_CHANNEL: ${{ secrets.SLACK_CHANNEL }}
SLACK_WEBHOOK: ${{ secrets.SLACK_WEBHOOK }}
REPO_CHECKOUT_TOKEN: ${{ secrets.REPO_CHECKOUT_TOKEN }}
JOB_SECRET: ${{ secrets.JOB_SECRET }}
SLAB_ACTION_TOKEN: ${{ secrets.SLAB_ACTION_TOKEN }}
SLAB_URL: ${{ secrets.SLAB_URL }}
SLAB_BASE_URL: ${{ secrets.SLAB_BASE_URL }}
run-benchmarks-gpu:
name: benchmark_documentation/run-benchmarks-gpu
uses: ./.github/workflows/benchmark_gpu_common.yml
if: inputs.run-gpu-benchmarks
needs: parse-gpu-inputs
with:
profile: ${{ needs.parse-gpu-inputs.outputs.profile }}
hardware_name: ${{ needs.parse-gpu-inputs.outputs.hardware_name }}
command: summary
bench_type: ${{ inputs.bench_type }}
params_type: classical + multi_bit
secrets:
BOT_USERNAME: ${{ secrets.BOT_USERNAME }}
SLACK_CHANNEL: ${{ secrets.SLACK_CHANNEL }}
SLACK_WEBHOOK: ${{ secrets.SLACK_WEBHOOK }}
REPO_CHECKOUT_TOKEN: ${{ secrets.REPO_CHECKOUT_TOKEN }}
JOB_SECRET: ${{ secrets.JOB_SECRET }}
SLAB_ACTION_TOKEN: ${{ secrets.SLAB_ACTION_TOKEN }}
SLAB_URL: ${{ secrets.SLAB_URL }}
SLAB_BASE_URL: ${{ secrets.SLAB_BASE_URL }}
# TODO add make recipe for HPU benchmarks
# run-benchmarks-hpu:
# name: benchmark_documentation/run-benchmarks-hpu
# uses: ./.github/workflows/benchmark_hpu_common.yml
# if: inputs.run-hpu-benchmarks
# with:
# command: summary
# bench_type: ${{ inputs.bench_type }}
# v80_pcie_dev: 24
# v80_serial_number: XFL12NWY3ZKG
# secrets:
# BOT_USERNAME: ${{ secrets.BOT_USERNAME }}
# SLACK_CHANNEL: ${{ secrets.SLACK_CHANNEL }}
# SLACK_WEBHOOK: ${{ secrets.SLACK_WEBHOOK }}
# REPO_CHECKOUT_TOKEN: ${{ secrets.REPO_CHECKOUT_TOKEN }}
# JOB_SECRET: ${{ secrets.JOB_SECRET }}
# SLAB_ACTION_TOKEN: ${{ secrets.SLAB_ACTION_TOKEN }}
# SLAB_URL: ${{ secrets.SLAB_URL }}
# SLAB_BASE_URL: ${{ secrets.SLAB_BASE_URL }}
# SSH_PRIVATE_KEY: ${{ secrets.SSH_PRIVATE_KEY }}

View File

@@ -1,5 +1,5 @@
# Run FFT benchmarks on an AWS instance and return parsed results to Slab CI bot.
name: benchmark_tfhe_fft
name: FFT benchmarks
env:
CARGO_TERM_COLOR: always
@@ -23,24 +23,16 @@ on:
# Job will be triggered each Thursday at 11p.m.
- cron: '0 23 * * 4'
permissions: {}
# zizmor: ignore[concurrency-limits] concurrency is managed after instance setup to ensure safe provisioning
jobs:
setup-instance:
name: benchmark_tfhe_fft/setup-instance
if:
(github.event_name != 'workflow_dispatch' && github.repository == 'zama-ai/tfhe-rs') ||
github.event_name == 'workflow_dispatch'
setup-ec2:
name: Setup EC2 instance (fft-benchmarks)
runs-on: ubuntu-latest
outputs:
runner-name: ${{ steps.start-instance.outputs.label }}
steps:
- name: Start instance
id: start-instance
uses: zama-ai/slab-github-runner@5aee5d157f4a0201e5eaefc9cc648e5f9f5472a5 # v1.6.0
uses: zama-ai/slab-github-runner@98f0788261a7323d5d695a883e20df36591a92b7
with:
mode: start
github-token: ${{ secrets.SLAB_ACTION_TOKEN }}
@@ -50,33 +42,28 @@ jobs:
profile: bench
fft-benchmarks:
name: benchmark_tfhe_fft/fft-benchmarks
needs: setup-instance
name: Execute FFT benchmarks in EC2
needs: setup-ec2
concurrency:
group: ${{ github.workflow_ref }}${{ github.ref == 'refs/heads/main' && github.sha || '' }}
group: ${{ github.workflow }}_${{ github.ref }}
cancel-in-progress: true
runs-on: ${{ needs.setup-instance.outputs.runner-name }}
runs-on: ${{ needs.setup-ec2.outputs.runner-name }}
steps:
- name: Checkout tfhe-rs repo with tags
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd
uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683
with:
fetch-depth: 0
persist-credentials: 'false'
token: ${{ secrets.REPO_CHECKOUT_TOKEN }}
- name: Get benchmark details
run: |
COMMIT_DATE=$(git --no-pager show -s --format=%cd --date=iso8601-strict "${SHA}");
{
echo "BENCH_DATE=$(date --iso-8601=seconds)";
echo "COMMIT_DATE=${COMMIT_DATE}";
echo "COMMIT_DATE=$(git --no-pager show -s --format=%cd --date=iso8601-strict ${{ github.sha }})";
echo "COMMIT_HASH=$(git describe --tags --dirty)";
} >> "${GITHUB_ENV}"
env:
SHA: ${{ github.sha }}
- name: Install rust
uses: dtolnay/rust-toolchain@e97e2d8cc328f1b50210efc529dca0028893a2d9 # zizmor: ignore[stale-action-refs] this action doesn't create releases
uses: actions-rs/toolchain@16499b5e05bf2e26879000db0c1d13f7e13fa3af
with:
toolchain: nightly
override: true
@@ -87,67 +74,70 @@ jobs:
- name: Parse AVX512 results
run: |
python3 ./ci/fft_benchmark_parser.py target/criterion "${RESULTS_FILENAME}" \
python3 ./ci/fft_benchmark_parser.py target/criterion ${{ env.RESULTS_FILENAME }} \
--database concrete_fft \
--hardware "hpc7a.96xlarge" \
--project-version "${COMMIT_HASH}" \
--branch "${REF_NAME}" \
--commit-date "${COMMIT_DATE}" \
--bench-date "${BENCH_DATE}" \
--project-version "${{ env.COMMIT_HASH }}" \
--branch ${{ github.ref_name }} \
--commit-date "${{ env.COMMIT_DATE }}" \
--bench-date "${{ env.BENCH_DATE }}" \
--name-suffix avx512
env:
REF_NAME: ${{ github.ref_name }}
- name: Upload parsed results artifact
uses: actions/upload-artifact@043fb46d1a93c77aae656e7c1c64a875d1fc6a0a
uses: actions/upload-artifact@b4b15b8c7c6ac21ea08fcf65892d2ee8f75cf882
with:
name: ${{ github.sha }}_fft
path: ${{ env.RESULTS_FILENAME }}
- name: Checkout Slab repo
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd
uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683
with:
repository: zama-ai/slab
path: slab
persist-credentials: 'false'
token: ${{ secrets.REPO_CHECKOUT_TOKEN }}
token: ${{ secrets.FHE_ACTIONS_TOKEN }}
- name: Send data to Slab
shell: bash
run: |
python3 slab/scripts/data_sender.py "${RESULTS_FILENAME}" "${JOB_SECRET}" \
--slab-url "${SLAB_URL}"
env:
JOB_SECRET: ${{ secrets.JOB_SECRET }}
SLAB_URL: ${{ secrets.SLAB_URL }}
echo "Computing HMac on downloaded artifact"
SIGNATURE="$(slab/scripts/hmac_calculator.sh ${{ env.RESULTS_FILENAME }} '${{ secrets.JOB_SECRET }}')"
echo "Sending results to Slab..."
curl -v -k \
-H "Content-Type: application/json" \
-H "X-Slab-Repository: ${{ github.repository }}" \
-H "X-Slab-Command: store_data_v2" \
-H "X-Hub-Signature-256: sha256=${SIGNATURE}" \
-d @${{ env.RESULTS_FILENAME }} \
${{ secrets.SLAB_URL }}
- name: Slack Notification
if: ${{ failure() || (cancelled() && github.event_name != 'pull_request') }}
if: ${{ failure() }}
continue-on-error: true
uses: rtCamp/action-slack-notify@e31e87e03dd19038e411e38ae27cbad084a90661
uses: rtCamp/action-slack-notify@c33737706dea87cd7784c687dadc9adf1be59990
env:
SLACK_COLOR: ${{ job.status }}
SLACK_MESSAGE: "tfhe-fft benchmarks failed. (${{ env.ACTION_RUN_URL }})"
teardown-instance:
name: benchmark_tfhe_fft/teardown-instance
if: ${{ always() && needs.setup-instance.result != 'skipped' }}
needs: [ setup-instance, fft-benchmarks ]
teardown-ec2:
name: Teardown EC2 instance (fft-benchmarks)
if: ${{ always() && needs.setup-ec2.result != 'skipped' }}
needs: [ setup-ec2, fft-benchmarks ]
runs-on: ubuntu-latest
steps:
- name: Stop instance
id: stop-instance
uses: zama-ai/slab-github-runner@5aee5d157f4a0201e5eaefc9cc648e5f9f5472a5 # v1.6.0
uses: zama-ai/slab-github-runner@98f0788261a7323d5d695a883e20df36591a92b7
with:
mode: stop
github-token: ${{ secrets.SLAB_ACTION_TOKEN }}
slab-url: ${{ secrets.SLAB_BASE_URL }}
job-secret: ${{ secrets.JOB_SECRET }}
label: ${{ needs.setup-instance.outputs.runner-name }}
label: ${{ needs.setup-ec2.outputs.runner-name }}
- name: Slack Notification
if: ${{ failure() }}
uses: rtCamp/action-slack-notify@e31e87e03dd19038e411e38ae27cbad084a90661
continue-on-error: true
uses: rtCamp/action-slack-notify@c33737706dea87cd7784c687dadc9adf1be59990
env:
SLACK_COLOR: ${{ job.status }}
SLACK_MESSAGE: "Instance teardown (fft-benchmarks) finished with status: ${{ job.status }}. (${{ env.ACTION_RUN_URL }})"
SLACK_MESSAGE: "EC2 teardown (fft-benchmarks) failed. (${{ env.ACTION_RUN_URL }})"

View File

@@ -1,5 +1,5 @@
# Run NTT benchmarks on an AWS instance and return parsed results to Slab CI bot.
name: benchmark_tfhe_ntt
name: NTT benchmarks
env:
CARGO_TERM_COLOR: always
@@ -23,24 +23,16 @@ on:
# Job will be triggered each Friday at 11p.m.
- cron: "0 23 * * 5"
permissions: {}
# zizmor: ignore[concurrency-limits] concurrency is managed after instance setup to ensure safe provisioning
jobs:
setup-instance:
name: benchmark_tfhe_ntt/setup-instance
if:
(github.event_name != 'workflow_dispatch' && github.repository == 'zama-ai/tfhe-rs') ||
github.event_name == 'workflow_dispatch'
setup-ec2:
name: Setup EC2 instance (ntt-benchmarks)
runs-on: ubuntu-latest
outputs:
runner-name: ${{ steps.start-instance.outputs.label }}
steps:
- name: Start instance
id: start-instance
uses: zama-ai/slab-github-runner@5aee5d157f4a0201e5eaefc9cc648e5f9f5472a5 # v1.6.0
uses: zama-ai/slab-github-runner@98f0788261a7323d5d695a883e20df36591a92b7
with:
mode: start
github-token: ${{ secrets.SLAB_ACTION_TOKEN }}
@@ -50,33 +42,28 @@ jobs:
profile: bench
ntt-benchmarks:
name: benchmark_tfhe_ntt/ntt-benchmarks
needs: setup-instance
name: Execute NTT benchmarks in EC2
needs: setup-ec2
concurrency:
group: ${{ github.workflow_ref }}${{ github.ref == 'refs/heads/main' && github.sha || '' }}
group: ${{ github.workflow }}_${{ github.ref }}
cancel-in-progress: true
runs-on: ${{ needs.setup-instance.outputs.runner-name }}
runs-on: ${{ needs.setup-ec2.outputs.runner-name }}
steps:
- name: Checkout tfhe-rs repo with tags
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd
uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683
with:
fetch-depth: 0
persist-credentials: 'false'
token: ${{ secrets.REPO_CHECKOUT_TOKEN }}
- name: Get benchmark details
run: |
COMMIT_DATE=$(git --no-pager show -s --format=%cd --date=iso8601-strict "${SHA}");
{
echo "BENCH_DATE=$(date --iso-8601=seconds)";
echo "COMMIT_DATE=${COMMIT_DATE}";
echo "COMMIT_DATE=$(git --no-pager show -s --format=%cd --date=iso8601-strict ${{ github.sha }})";
echo "COMMIT_HASH=$(git describe --tags --dirty)";
} >> "${GITHUB_ENV}"
env:
SHA: ${{ github.sha }}
- name: Install rust
uses: dtolnay/rust-toolchain@e97e2d8cc328f1b50210efc529dca0028893a2d9 # zizmor: ignore[stale-action-refs] this action doesn't create releases
uses: actions-rs/toolchain@16499b5e05bf2e26879000db0c1d13f7e13fa3af
with:
toolchain: nightly
override: true
@@ -87,67 +74,70 @@ jobs:
- name: Parse results
run: |
python3 ./ci/ntt_benchmark_parser.py target/criterion "${RESULTS_FILENAME}" \
python3 ./ci/ntt_benchmark_parser.py target/criterion ${{ env.RESULTS_FILENAME }} \
--database concrete_ntt \
--hardware "hpc7a.96xlarge" \
--project-version "${COMMIT_HASH}" \
--branch "${REF_NAME}" \
--commit-date "${COMMIT_DATE}" \
--bench-date "${BENCH_DATE}" \
--project-version "${{ env.COMMIT_HASH }}" \
--branch ${{ github.ref_name }} \
--commit-date "${{ env.COMMIT_DATE }}" \
--bench-date "${{ env.BENCH_DATE }}" \
--name-suffix avx512
env:
REF_NAME: ${{ github.ref_name }}
- name: Upload parsed results artifact
uses: actions/upload-artifact@043fb46d1a93c77aae656e7c1c64a875d1fc6a0a
uses: actions/upload-artifact@b4b15b8c7c6ac21ea08fcf65892d2ee8f75cf882
with:
name: ${{ github.sha }}_ntt
path: ${{ env.RESULTS_FILENAME }}
- name: Checkout Slab repo
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd
uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683
with:
repository: zama-ai/slab
path: slab
persist-credentials: 'false'
token: ${{ secrets.REPO_CHECKOUT_TOKEN }}
token: ${{ secrets.FHE_ACTIONS_TOKEN }}
- name: Send data to Slab
shell: bash
run: |
python3 slab/scripts/data_sender.py "${RESULTS_FILENAME}" "${JOB_SECRET}" \
--slab-url "${SLAB_URL}"
env:
JOB_SECRET: ${{ secrets.JOB_SECRET }}
SLAB_URL: ${{ secrets.SLAB_URL }}
echo "Computing HMac on downloaded artifact"
SIGNATURE="$(slab/scripts/hmac_calculator.sh ${{ env.RESULTS_FILENAME }} '${{ secrets.JOB_SECRET }}')"
echo "Sending results to Slab..."
curl -v -k \
-H "Content-Type: application/json" \
-H "X-Slab-Repository: ${{ github.repository }}" \
-H "X-Slab-Command: store_data_v2" \
-H "X-Hub-Signature-256: sha256=${SIGNATURE}" \
-d @${{ env.RESULTS_FILENAME }} \
${{ secrets.SLAB_URL }}
- name: Slack Notification
if: ${{ failure() || (cancelled() && github.event_name != 'pull_request') }}
if: ${{ failure() }}
continue-on-error: true
uses: rtCamp/action-slack-notify@e31e87e03dd19038e411e38ae27cbad084a90661
uses: rtCamp/action-slack-notify@c33737706dea87cd7784c687dadc9adf1be59990
env:
SLACK_COLOR: ${{ job.status }}
SLACK_MESSAGE: "tfhe-ntt benchmarks failed. (${{ env.ACTION_RUN_URL }})"
teardown-instance:
name: benchmark_tfhe_ntt/teardown-instance
if: ${{ always() && needs.setup-instance.result != 'skipped' }}
needs: [setup-instance, ntt-benchmarks]
teardown-ec2:
name: Teardown EC2 instance (ntt-benchmarks)
if: ${{ always() && needs.setup-ec2.result != 'skipped' }}
needs: [setup-ec2, ntt-benchmarks]
runs-on: ubuntu-latest
steps:
- name: Stop instance
id: stop-instance
uses: zama-ai/slab-github-runner@5aee5d157f4a0201e5eaefc9cc648e5f9f5472a5 # v1.6.0
uses: zama-ai/slab-github-runner@98f0788261a7323d5d695a883e20df36591a92b7
with:
mode: stop
github-token: ${{ secrets.SLAB_ACTION_TOKEN }}
slab-url: ${{ secrets.SLAB_BASE_URL }}
job-secret: ${{ secrets.JOB_SECRET }}
label: ${{ needs.setup-instance.outputs.runner-name }}
label: ${{ needs.setup-ec2.outputs.runner-name }}
- name: Slack Notification
if: ${{ failure() }}
uses: rtCamp/action-slack-notify@e31e87e03dd19038e411e38ae27cbad084a90661
continue-on-error: true
uses: rtCamp/action-slack-notify@c33737706dea87cd7784c687dadc9adf1be59990
env:
SLACK_COLOR: ${{ job.status }}
SLACK_MESSAGE: "EC2 teardown (ntt-benchmarks) finished with status: ${{ job.status }}. (${{ env.ACTION_RUN_URL }})"
SLACK_MESSAGE: "EC2 teardown (ntt-benchmarks) failed. (${{ env.ACTION_RUN_URL }})"

View File

@@ -0,0 +1,172 @@
# Run benchmarks of the tfhe-zk-pok crate on an instance and return parsed results to Slab CI bot.
name: tfhe-zk-pok benchmarks
on:
workflow_dispatch:
push:
branches:
- main
schedule:
# Weekly benchmarks will be triggered each Saturday at 3a.m.
- cron: '0 3 * * 6'
env:
CARGO_TERM_COLOR: always
RESULTS_FILENAME: parsed_benchmark_results_${{ github.sha }}.json
PARSE_INTEGER_BENCH_CSV_FILE: tfhe_rs_integer_benches_${{ github.sha }}.csv
ACTION_RUN_URL: ${{ github.server_url }}/${{ github.repository }}/actions/runs/${{ github.run_id }}
RUST_BACKTRACE: "full"
RUST_MIN_STACK: "8388608"
SLACK_CHANNEL: ${{ secrets.SLACK_CHANNEL }}
SLACK_ICON: https://pbs.twimg.com/profile_images/1274014582265298945/OjBKP9kn_400x400.png
SLACK_USERNAME: ${{ secrets.BOT_USERNAME }}
SLACK_WEBHOOK: ${{ secrets.SLACK_WEBHOOK }}
jobs:
should-run:
runs-on: ubuntu-latest
if: github.event_name == 'workflow_dispatch' ||
((github.event_name == 'push' || github.event_name == 'schedule') && github.repository == 'zama-ai/tfhe-rs')
outputs:
zk_pok_changed: ${{ steps.changed-files.outputs.zk_pok_any_changed }}
steps:
- name: Checkout tfhe-rs
uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683
with:
fetch-depth: 0
- name: Check for file changes
id: changed-files
uses: tj-actions/changed-files@bab30c2299617f6615ec02a68b9a40d10bd21366
with:
since_last_remote_commit: true
files_yaml: |
zk_pok:
- tfhe-zk-pok/**
- .github/workflows/benchmark_tfhe_zk_pok.yml
setup-instance:
name: Setup instance (tfhe-zk-pok-benchmarks)
runs-on: ubuntu-latest
needs: should-run
if: github.event_name == 'workflow_dispatch' ||
(github.event_name == 'schedule' && github.repository == 'zama-ai/tfhe-rs') ||
(github.event_name == 'push' &&
github.repository == 'zama-ai/tfhe-rs' &&
needs.should-run.outputs.zk_pok_changed == 'true')
outputs:
runner-name: ${{ steps.start-instance.outputs.label }}
steps:
- name: Start instance
id: start-instance
uses: zama-ai/slab-github-runner@98f0788261a7323d5d695a883e20df36591a92b7
with:
mode: start
github-token: ${{ secrets.SLAB_ACTION_TOKEN }}
slab-url: ${{ secrets.SLAB_BASE_URL }}
job-secret: ${{ secrets.JOB_SECRET }}
backend: aws
profile: bench
tfhe-zk-pok-benchmarks:
name: Execute tfhe-zk-pok benchmarks
if: needs.setup-instance.result != 'skipped'
needs: setup-instance
concurrency:
group: ${{ github.workflow }}_${{github.event_name}}_${{ github.ref }}${{ github.ref == 'refs/heads/main' && github.sha || '' }}
cancel-in-progress: ${{ github.ref != 'refs/heads/main' }}
runs-on: ${{ needs.setup-instance.outputs.runner-name }}
steps:
- name: Checkout tfhe-rs repo with tags
uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683
with:
fetch-depth: 0
token: ${{ secrets.FHE_ACTIONS_TOKEN }}
- name: Get benchmark details
run: |
{
echo "BENCH_DATE=$(date --iso-8601=seconds)";
echo "COMMIT_DATE=$(git --no-pager show -s --format=%cd --date=iso8601-strict ${{ github.sha }})";
echo "COMMIT_HASH=$(git describe --tags --dirty)";
} >> "${GITHUB_ENV}"
- name: Install rust
uses: dtolnay/rust-toolchain@315e265cd78dad1e1dcf3a5074f6d6c47029d5aa
with:
toolchain: nightly
- name: Checkout Slab repo
uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683
with:
repository: zama-ai/slab
path: slab
token: ${{ secrets.FHE_ACTIONS_TOKEN }}
- name: Run benchmarks
run: |
make bench_tfhe_zk_pok
- name: Parse results
run: |
python3 ./ci/benchmark_parser.py target/criterion ${{ env.RESULTS_FILENAME }} \
--database tfhe_rs \
--crate tfhe-zk-pok \
--hardware "hpc7a.96xlarge" \
--backend cpu \
--project-version "${{ env.COMMIT_HASH }}" \
--branch ${{ github.ref_name }} \
--commit-date "${{ env.COMMIT_DATE }}" \
--bench-date "${{ env.BENCH_DATE }}" \
--walk-subdirs \
--name-suffix avx512
- name: Upload parsed results artifact
uses: actions/upload-artifact@b4b15b8c7c6ac21ea08fcf65892d2ee8f75cf882
with:
name: ${{ github.sha }}_tfhe_zk_pok
path: ${{ env.RESULTS_FILENAME }}
- name: Checkout Slab repo
uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683
with:
repository: zama-ai/slab
path: slab
token: ${{ secrets.FHE_ACTIONS_TOKEN }}
- name: Send data to Slab
shell: bash
run: |
python3 slab/scripts/data_sender.py ${{ env.RESULTS_FILENAME }} "${{ secrets.JOB_SECRET }}" \
--slab-url "${{ secrets.SLAB_URL }}"
- name: Slack Notification
if: ${{ failure() }}
continue-on-error: true
uses: rtCamp/action-slack-notify@c33737706dea87cd7784c687dadc9adf1be59990
env:
SLACK_COLOR: ${{ job.status }}
SLACK_MESSAGE: "tfhe-zk-pok benchmarks finished with status: ${{ job.status }}. (${{ env.ACTION_RUN_URL }})"
teardown-instance:
name: Teardown instance (tfhe-zk-pok-benchmarks)
if: ${{ always() && needs.setup-instance.result != 'skipped' }}
needs: [ setup-instance, tfhe-zk-pok-benchmarks ]
runs-on: ubuntu-latest
steps:
- name: Stop instance
id: stop-instance
uses: zama-ai/slab-github-runner@98f0788261a7323d5d695a883e20df36591a92b7
with:
mode: stop
github-token: ${{ secrets.SLAB_ACTION_TOKEN }}
slab-url: ${{ secrets.SLAB_BASE_URL }}
job-secret: ${{ secrets.JOB_SECRET }}
label: ${{ needs.setup-instance.outputs.runner-name }}
- name: Slack Notification
if: ${{ failure() }}
continue-on-error: true
uses: rtCamp/action-slack-notify@c33737706dea87cd7784c687dadc9adf1be59990
env:
SLACK_COLOR: ${{ job.status }}
SLACK_MESSAGE: "Instance teardown (tfhe-zk-pok-benchmarks) finished with status: ${{ job.status }}. (${{ env.ACTION_RUN_URL }})"

View File

@@ -1,5 +1,5 @@
# Run WASM client benchmarks on an instance and return parsed results to Slab CI bot.
name: benchmark_wasm_client
name: WASM client benchmarks
on:
workflow_dispatch:
@@ -21,33 +21,28 @@ env:
SLACK_USERNAME: ${{ secrets.BOT_USERNAME }}
SLACK_WEBHOOK: ${{ secrets.SLACK_WEBHOOK }}
permissions: {}
# zizmor: ignore[concurrency-limits] only Zama organization members and GitHub can trigger this workflow
jobs:
should-run:
name: benchmark_wasm_client/should-run
runs-on: ubuntu-latest
if: github.event_name == 'workflow_dispatch' ||
(github.event_name != 'workflow_dispatch' && github.repository == 'zama-ai/tfhe-rs')
(github.event_name == 'schedule' && github.repository == 'zama-ai/tfhe-rs') ||
(github.event_name == 'push' && github.repository == 'zama-ai/tfhe-rs')
permissions:
pull-requests: read # Needed to check for file change
pull-requests: write
outputs:
wasm_bench: ${{ steps.changed-files.outputs.wasm_bench_any_changed }}
steps:
- name: Checkout tfhe-rs
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd
uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683
with:
fetch-depth: 0
persist-credentials: 'false'
token: ${{ secrets.REPO_CHECKOUT_TOKEN }}
token: ${{ secrets.FHE_ACTIONS_TOKEN }}
- name: Check for file changes
id: changed-files
uses: tj-actions/changed-files@9426d40962ed5378910ee2e21d5f8c6fcbf2dd96 # v47.0.6
uses: tj-actions/changed-files@bab30c2299617f6615ec02a68b9a40d10bd21366
with:
since_last_remote_commit: true
files_yaml: |
wasm_bench:
- tfhe/Cargo.toml
@@ -58,19 +53,165 @@ jobs:
- tfhe/web_wasm_parallel_tests/**
- .github/workflows/wasm_client_benchmark.yml
run-benchmarks-cpu-zk-client:
name: benchmark_documentation/run-benchmarks-cpu-zk-client
uses: ./.github/workflows/benchmark_wasm_client_common.yml
needs: should-run
setup-instance:
name: Setup instance (wasm-client-benchmarks)
if: github.event_name == 'workflow_dispatch' ||
(github.event_name == 'schedule' && github.repository == 'zama-ai/tfhe-rs') ||
(github.event_name == 'push' && github.repository == 'zama-ai/tfhe-rs' && needs.should-run.outputs.wasm_bench)
secrets:
BOT_USERNAME: ${{ secrets.BOT_USERNAME }}
SLACK_CHANNEL: ${{ secrets.SLACK_CHANNEL }}
SLACK_WEBHOOK: ${{ secrets.SLACK_WEBHOOK }}
REPO_CHECKOUT_TOKEN: ${{ secrets.REPO_CHECKOUT_TOKEN }}
JOB_SECRET: ${{ secrets.JOB_SECRET }}
SLAB_ACTION_TOKEN: ${{ secrets.SLAB_ACTION_TOKEN }}
SLAB_URL: ${{ secrets.SLAB_URL }}
SLAB_BASE_URL: ${{ secrets.SLAB_BASE_URL }}
needs: should-run
runs-on: ubuntu-latest
outputs:
runner-name: ${{ steps.start-instance.outputs.label }}
steps:
- name: Start instance
id: start-instance
uses: zama-ai/slab-github-runner@98f0788261a7323d5d695a883e20df36591a92b7
with:
mode: start
github-token: ${{ secrets.SLAB_ACTION_TOKEN }}
slab-url: ${{ secrets.SLAB_BASE_URL }}
job-secret: ${{ secrets.JOB_SECRET }}
backend: aws
profile: cpu-small
wasm-client-benchmarks:
name: Execute WASM client benchmarks
needs: setup-instance
if: needs.setup-instance.result != 'skipped'
runs-on: ${{ needs.setup-instance.outputs.runner-name }}
strategy:
max-parallel: 1
matrix:
browser: [ chrome, firefox ]
steps:
- name: Checkout tfhe-rs repo with tags
uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683
with:
fetch-depth: 0
token: ${{ secrets.FHE_ACTIONS_TOKEN }}
- name: Get benchmark details
run: |
{
echo "BENCH_DATE=$(date --iso-8601=seconds)";
echo "COMMIT_DATE=$(git --no-pager show -s --format=%cd --date=iso8601-strict ${{ github.sha }})";
echo "COMMIT_HASH=$(git describe --tags --dirty)";
} >> "${GITHUB_ENV}"
- name: Install rust
uses: dtolnay/rust-toolchain@315e265cd78dad1e1dcf3a5074f6d6c47029d5aa
with:
toolchain: nightly
- name: Get Node version
run: |
echo "NODE_VERSION=$(make node_version)" >> "${GITHUB_ENV}"
- name: Node cache restoration
id: node-cache
uses: actions/cache/restore@1bd1e32a3bdc45362d1e726936510720a7c30a57 #v4.2.0
with:
path: |
~/.nvm
~/.npm
key: node-${{ env.NODE_VERSION }}
- name: Install Node
if: steps.node-cache.outputs.cache-hit != 'true'
run: |
make install_node
- name: Node cache save
uses: actions/cache/save@1bd1e32a3bdc45362d1e726936510720a7c30a57 #v4.2.0
if: steps.node-cache.outputs.cache-hit != 'true'
with:
path: |
~/.nvm
~/.npm
key: node-${{ env.NODE_VERSION }}
- name: Install web resources
run: |
make install_${{ matrix.browser }}_browser
make install_${{ matrix.browser }}_web_driver
- name: Run benchmarks
run: |
make bench_web_js_api_parallel_${{ matrix.browser }}_ci
- name: Parse results
run: |
make parse_wasm_benchmarks
python3 ./ci/benchmark_parser.py tfhe/wasm_pk_gen.csv ${{ env.RESULTS_FILENAME }} \
--database tfhe_rs \
--hardware "m6i.4xlarge" \
--project-version "${{ env.COMMIT_HASH }}" \
--branch ${{ github.ref_name }} \
--commit-date "${{ env.COMMIT_DATE }}" \
--bench-date "${{ env.BENCH_DATE }}" \
--key-gen
rm tfhe/wasm_pk_gen.csv
# Run these benchmarks only once
- name: Measure public key and ciphertext sizes in HL Api
if: matrix.browser == 'chrome'
run: |
make measure_hlapi_compact_pk_ct_sizes
- name: Parse key and ciphertext sizes results
if: matrix.browser == 'chrome'
run: |
python3 ./ci/benchmark_parser.py tfhe/hlapi_cpk_and_cctl_sizes.csv ${{ env.RESULTS_FILENAME }} \
--key-gen \
--append-results
- name: Upload parsed results artifact
uses: actions/upload-artifact@b4b15b8c7c6ac21ea08fcf65892d2ee8f75cf882
with:
name: ${{ github.sha }}_wasm_${{ matrix.browser }}
path: ${{ env.RESULTS_FILENAME }}
- name: Checkout Slab repo
uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683
with:
repository: zama-ai/slab
path: slab
token: ${{ secrets.FHE_ACTIONS_TOKEN }}
- name: Send data to Slab
shell: bash
run: |
python3 slab/scripts/data_sender.py ${{ env.RESULTS_FILENAME }} "${{ secrets.JOB_SECRET }}" \
--slab-url "${{ secrets.SLAB_URL }}"
- name: Slack Notification
if: ${{ failure() }}
continue-on-error: true
uses: rtCamp/action-slack-notify@c33737706dea87cd7784c687dadc9adf1be59990
env:
SLACK_COLOR: ${{ job.status }}
SLACK_MESSAGE: "WASM benchmarks (${{ matrix.browser }}) finished with status: ${{ job.status }}. (${{ env.ACTION_RUN_URL }})"
teardown-instance:
name: Teardown instance (wasm-client-benchmarks)
if: ${{ always() && needs.setup-instance.result != 'skipped' }}
needs: [ setup-instance, wasm-client-benchmarks ]
runs-on: ubuntu-latest
steps:
- name: Stop instance
id: stop-instance
uses: zama-ai/slab-github-runner@98f0788261a7323d5d695a883e20df36591a92b7
with:
mode: stop
github-token: ${{ secrets.SLAB_ACTION_TOKEN }}
slab-url: ${{ secrets.SLAB_BASE_URL }}
job-secret: ${{ secrets.JOB_SECRET }}
label: ${{ needs.setup-instance.outputs.runner-name }}
- name: Slack Notification
if: ${{ failure() }}
continue-on-error: true
uses: rtCamp/action-slack-notify@c33737706dea87cd7784c687dadc9adf1be59990
env:
SLACK_COLOR: ${{ job.status }}
SLACK_MESSAGE: "Instance teardown (wasm-client-benchmarks) finished with status: ${{ job.status }}. (${{ env.ACTION_RUN_URL }})"

View File

@@ -1,234 +0,0 @@
# Run WASM client benchmarks on an instance and return parsed results to Slab CI bot.
name: benchmark_wasm_client_common
on:
workflow_call:
inputs:
browser:
type: string # Use comma separated values to generate an array
default: chrome,firefox
secrets:
REPO_CHECKOUT_TOKEN:
required: true
SLAB_ACTION_TOKEN:
required: true
SLAB_BASE_URL:
required: true
SLAB_URL:
required: true
JOB_SECRET:
required: true
SLACK_CHANNEL:
required: true
BOT_USERNAME:
required: true
SLACK_WEBHOOK:
required: true
env:
CARGO_TERM_COLOR: always
RESULTS_FILENAME: parsed_benchmark_results_${{ github.sha }}.json
ACTION_RUN_URL: ${{ github.server_url }}/${{ github.repository }}/actions/runs/${{ github.run_id }}
RUST_BACKTRACE: "full"
RUST_MIN_STACK: "8388608"
SLACK_CHANNEL: ${{ secrets.SLACK_CHANNEL }}
SLACK_ICON: https://pbs.twimg.com/profile_images/1274014582265298945/OjBKP9kn_400x400.png
SLACK_USERNAME: ${{ secrets.BOT_USERNAME }}
SLACK_WEBHOOK: ${{ secrets.SLACK_WEBHOOK }}
permissions: {}
# zizmor: ignore[concurrency-limits] only Zama organization members and GitHub can trigger this workflow
jobs:
prepare-matrix:
name: benchmark_wasm_client_common/prepare-matrix
runs-on: ubuntu-latest
outputs:
browser: ${{ steps.set_matrix_arg.outputs.browser }}
steps:
- name: Parse user inputs
shell: python
env:
INPUTS_BROWSER: ${{ inputs.browser }}
run: |
import os
inputs_browser = os.environ["INPUTS_BROWSER"]
env_file = os.environ["GITHUB_ENV"]
split_browser = inputs_browser.replace(" ", "").split(",")
with open(env_file, "a") as f:
f.write(f"""BROWSER=["{'", "'.join(split_browser)}"]\n""")
- name: Set matrix arguments output
id: set_matrix_arg
run: | # zizmor: ignore[template-injection] this env variable is safe
echo "browser=${{ toJSON(env.BROWSER) }}" >> "${GITHUB_OUTPUT}"
setup-instance:
name: benchmark_wasm_client_common/setup-instance
needs: prepare-matrix
runs-on: ubuntu-latest
outputs:
runner-name: ${{ steps.start-instance.outputs.label }}
steps:
- name: Start instance
id: start-instance
uses: zama-ai/slab-github-runner@5aee5d157f4a0201e5eaefc9cc648e5f9f5472a5 # v1.6.0
with:
mode: start
github-token: ${{ secrets.SLAB_ACTION_TOKEN }}
slab-url: ${{ secrets.SLAB_BASE_URL }}
job-secret: ${{ secrets.JOB_SECRET }}
backend: aws
profile: cpu-small
wasm-client-benchmarks:
name: benchmark_wasm_client_common/wasm-client-benchmarks
needs: [ prepare-matrix, setup-instance ]
runs-on: ${{ needs.setup-instance.outputs.runner-name }}
strategy:
max-parallel: 1
matrix:
browser: ${{ fromJSON(needs.prepare-matrix.outputs.browser) }}
steps:
- name: Checkout tfhe-rs repo with tags
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd
with:
fetch-depth: 0
persist-credentials: 'false'
token: ${{ secrets.REPO_CHECKOUT_TOKEN }}
- name: Get benchmark details
run: |
COMMIT_DATE=$(git --no-pager show -s --format=%cd --date=iso8601-strict "${SHA}");
{
echo "BENCH_DATE=$(date --iso-8601=seconds)";
echo "COMMIT_DATE=${COMMIT_DATE}";
echo "COMMIT_HASH=$(git describe --tags --dirty)";
} >> "${GITHUB_ENV}"
env:
SHA: ${{ github.sha }}
- name: Install rust
uses: dtolnay/rust-toolchain@e97e2d8cc328f1b50210efc529dca0028893a2d9 # zizmor: ignore[stale-action-refs] this action doesn't create releases
with:
toolchain: nightly
- name: Get Node version
run: |
echo "NODE_VERSION=$(make node_version)" >> "${GITHUB_ENV}"
- name: Node cache restoration
id: node-cache
uses: actions/cache/restore@27d5ce7f107fe9357f9df03efb73ab90386fccae #v5.0.5
with:
path: |
~/.nvm
~/.npm
key: node-${{ env.NODE_VERSION }}
- name: Install Node
if: steps.node-cache.outputs.cache-hit != 'true'
run: |
make install_node
- name: Node cache save
uses: actions/cache/save@27d5ce7f107fe9357f9df03efb73ab90386fccae #v5.0.5
if: steps.node-cache.outputs.cache-hit != 'true'
with:
path: |
~/.nvm
~/.npm
key: node-${{ env.NODE_VERSION }}
- name: Install web resources
run: |
make install_"${BROWSER}"_browser
make install_"${BROWSER}"_web_driver
env:
BROWSER: ${{ matrix.browser }}
- name: Run benchmarks
run: |
make bench_web_js_api_parallel_"${BROWSER}"_ci
env:
BROWSER: ${{ matrix.browser }}
- name: Run benchmarks (cross origin)
run: |
make bench_web_js_api_cross_origin_"${BROWSER}"_ci
env:
BROWSER: ${{ matrix.browser }}
- name: Parse results
run: |
make parse_wasm_benchmarks
python3 ./ci/benchmark_parser.py tfhe-benchmark/wasm_pk_gen.csv "${RESULTS_FILENAME}" \
--database tfhe_rs \
--hardware "m6i.4xlarge" \
--project-version "${COMMIT_HASH}" \
--branch "${REF_NAME}" \
--commit-date "${COMMIT_DATE}" \
--bench-date "${BENCH_DATE}" \
--key-gen
rm tfhe-benchmark/wasm_pk_gen.csv
env:
REF_NAME: ${{ github.ref_name }}
- name: Upload parsed results artifact
uses: actions/upload-artifact@043fb46d1a93c77aae656e7c1c64a875d1fc6a0a
with:
name: ${{ github.sha }}_wasm_${{ matrix.browser }}
path: ${{ env.RESULTS_FILENAME }}
- name: Checkout Slab repo
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd
with:
repository: zama-ai/slab
path: slab
persist-credentials: 'false'
token: ${{ secrets.REPO_CHECKOUT_TOKEN }}
- name: Send data to Slab
shell: bash
run: |
python3 slab/scripts/data_sender.py "${RESULTS_FILENAME}" "${JOB_SECRET}" \
--slab-url "${SLAB_URL}"
env:
JOB_SECRET: ${{ secrets.JOB_SECRET }}
SLAB_URL: ${{ secrets.SLAB_URL }}
- name: Slack Notification
if: ${{ failure() || (cancelled() && github.event_name != 'pull_request') }}
continue-on-error: true
uses: rtCamp/action-slack-notify@e31e87e03dd19038e411e38ae27cbad084a90661
env:
SLACK_COLOR: ${{ job.status }}
SLACK_MESSAGE: "WASM benchmarks (${{ matrix.browser }}) finished with status: ${{ job.status }}. (${{ env.ACTION_RUN_URL }})"
teardown-instance:
name: benchmark_wasm_client_common/teardown-instance
if: ${{ always() && needs.setup-instance.result == 'success' }}
needs: [ setup-instance, wasm-client-benchmarks ]
runs-on: ubuntu-latest
steps:
- name: Stop instance
id: stop-instance
uses: zama-ai/slab-github-runner@5aee5d157f4a0201e5eaefc9cc648e5f9f5472a5 # v1.6.0
with:
mode: stop
github-token: ${{ secrets.SLAB_ACTION_TOKEN }}
slab-url: ${{ secrets.SLAB_BASE_URL }}
job-secret: ${{ secrets.JOB_SECRET }}
label: ${{ needs.setup-instance.outputs.runner-name }}
- name: Slack Notification
if: ${{ failure() }}
uses: rtCamp/action-slack-notify@e31e87e03dd19038e411e38ae27cbad084a90661
env:
SLACK_COLOR: ${{ job.status }}
SLACK_MESSAGE: "Instance teardown (wasm-client-benchmarks) finished with status: ${{ job.status }}. (${{ env.ACTION_RUN_URL }})"

View File

@@ -1,18 +0,0 @@
# Run all benchmarks displayed in the tfhe-rs whitepaper.
name: benchmark_whitepaper
on:
workflow_dispatch:
permissions: {}
# zizmor: ignore[concurrency-limits] only Zama organization members can trigger this workflow
jobs:
placeholder-job:
name: benchmark_whitepaper/placeholder-job
runs-on: ubuntu-latest
steps:
- run: |
echo "Hello this is a Placeholder Workflow"

228
.github/workflows/benchmark_zk_pke.yml vendored Normal file
View File

@@ -0,0 +1,228 @@
# Run PKE Zero-Knowledge benchmarks on an instance and return parsed results to Slab CI bot.
name: PKE ZK benchmarks
on:
workflow_dispatch:
inputs:
bench_type:
description: "Benchmarks type"
type: choice
default: latency
options:
- latency
- throughput
- both
push:
branches:
- main
schedule:
# Weekly benchmarks will be triggered each Saturday at 3a.m.
- cron: '0 3 * * 6'
env:
CARGO_TERM_COLOR: always
RESULTS_FILENAME: parsed_benchmark_results_${{ github.sha }}.json
PARSE_INTEGER_BENCH_CSV_FILE: tfhe_rs_integer_benches_${{ github.sha }}.csv
ACTION_RUN_URL: ${{ github.server_url }}/${{ github.repository }}/actions/runs/${{ github.run_id }}
RUST_BACKTRACE: "full"
RUST_MIN_STACK: "8388608"
SLACK_CHANNEL: ${{ secrets.SLACK_CHANNEL }}
SLACK_ICON: https://pbs.twimg.com/profile_images/1274014582265298945/OjBKP9kn_400x400.png
SLACK_USERNAME: ${{ secrets.BOT_USERNAME }}
SLACK_WEBHOOK: ${{ secrets.SLACK_WEBHOOK }}
jobs:
should-run:
runs-on: ubuntu-latest
if: github.event_name == 'workflow_dispatch' ||
((github.event_name == 'push' || github.event_name == 'schedule') && github.repository == 'zama-ai/tfhe-rs')
outputs:
zk_pok_changed: ${{ steps.changed-files.outputs.zk_pok_any_changed }}
steps:
- name: Checkout tfhe-rs
uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683
with:
fetch-depth: 0
token: ${{ secrets.FHE_ACTIONS_TOKEN }}
- name: Check for file changes
id: changed-files
uses: tj-actions/changed-files@bab30c2299617f6615ec02a68b9a40d10bd21366
with:
since_last_remote_commit: true
files_yaml: |
zk_pok:
- tfhe/Cargo.toml
- tfhe-csprng/**
- tfhe-fft/**
- tfhe-zk-pok/**
- tfhe/src/core_crypto/**
- tfhe/src/shortint/**
- tfhe/src/integer/**
- tfhe/src/zk.rs
- tfhe/benches/integer/zk_pke.rs
- .github/workflows/zk_pke_benchmark.yml
prepare-matrix:
name: Prepare operations matrix
runs-on: ubuntu-latest
if: github.event_name != 'schedule' ||
(github.event_name == 'schedule' && github.repository == 'zama-ai/tfhe-rs')
outputs:
bench_type: ${{ steps.set_bench_type.outputs.bench_type }}
steps:
- name: Set benchmark types
if: github.event_name == 'workflow_dispatch'
run: |
if [[ "${{ inputs.bench_type }}" == "both" ]]; then
echo "BENCH_TYPE=[\"latency\", \"throughput\"]" >> "${GITHUB_ENV}"
else
echo "BENCH_TYPE=[\"${{ inputs.bench_type }}\"]" >> "${GITHUB_ENV}"
fi
- name: Default benchmark type
if: github.event_name != 'workflow_dispatch'
run: |
echo "BENCH_TYPE=[\"latency\"]" >> "${GITHUB_ENV}"
- name: Set benchmark types output
id: set_bench_type
run: |
echo "bench_type=${{ toJSON(env.BENCH_TYPE) }}" >> "${GITHUB_OUTPUT}"
setup-instance:
name: Setup instance (pke-zk-benchmarks)
runs-on: ubuntu-latest
needs: [ should-run, prepare-matrix ]
if: github.event_name == 'workflow_dispatch' ||
(github.event_name == 'schedule' && github.repository == 'zama-ai/tfhe-rs') ||
(github.event_name == 'push' &&
github.repository == 'zama-ai/tfhe-rs' &&
needs.should-run.outputs.zk_pok_changed == 'true')
outputs:
runner-name: ${{ steps.start-instance.outputs.label }}
steps:
- name: Start instance
id: start-instance
uses: zama-ai/slab-github-runner@98f0788261a7323d5d695a883e20df36591a92b7
with:
mode: start
github-token: ${{ secrets.SLAB_ACTION_TOKEN }}
slab-url: ${{ secrets.SLAB_BASE_URL }}
job-secret: ${{ secrets.JOB_SECRET }}
backend: aws
profile: bench
pke-zk-benchmarks:
name: Execute PKE ZK benchmarks
if: needs.setup-instance.result != 'skipped'
needs: [ prepare-matrix, setup-instance ]
concurrency:
group: ${{ github.workflow }}_${{github.event_name}}_${{ github.ref }}${{ github.ref == 'refs/heads/main' && github.sha || '' }}
cancel-in-progress: ${{ github.ref != 'refs/heads/main' }}
runs-on: ${{ needs.setup-instance.outputs.runner-name }}
strategy:
max-parallel: 1
matrix:
bench_type: ${{ fromJSON(needs.prepare-matrix.outputs.bench_type) }}
steps:
- name: Checkout tfhe-rs repo with tags
uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683
with:
fetch-depth: 0
token: ${{ secrets.FHE_ACTIONS_TOKEN }}
- name: Get benchmark details
run: |
{
echo "BENCH_DATE=$(date --iso-8601=seconds)";
echo "COMMIT_DATE=$(git --no-pager show -s --format=%cd --date=iso8601-strict ${{ github.sha }})";
echo "COMMIT_HASH=$(git describe --tags --dirty)";
} >> "${GITHUB_ENV}"
- name: Install rust
uses: dtolnay/rust-toolchain@315e265cd78dad1e1dcf3a5074f6d6c47029d5aa
with:
toolchain: nightly
- name: Checkout Slab repo
uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683
with:
repository: zama-ai/slab
path: slab
token: ${{ secrets.FHE_ACTIONS_TOKEN }}
- name: Run benchmarks with AVX512
run: |
make BENCH_TYPE=${{ matrix.bench_type }} bench_integer_zk
- name: Parse results
run: |
python3 ./ci/benchmark_parser.py target/criterion ${{ env.RESULTS_FILENAME }} \
--database tfhe_rs \
--hardware "hpc7a.96xlarge" \
--backend cpu \
--project-version "${{ env.COMMIT_HASH }}" \
--branch ${{ github.ref_name }} \
--commit-date "${{ env.COMMIT_DATE }}" \
--bench-date "${{ env.BENCH_DATE }}" \
--walk-subdirs \
--name-suffix avx512 \
--bench-type ${{ matrix.bench_type }}
- name: Parse CRS sizes results
run: |
python3 ./ci/benchmark_parser.py tfhe/pke_zk_crs_sizes.csv ${{ env.RESULTS_FILENAME }} \
--object-sizes \
--append-results
- name: Upload parsed results artifact
uses: actions/upload-artifact@b4b15b8c7c6ac21ea08fcf65892d2ee8f75cf882
with:
name: ${{ github.sha }}_integer_zk
path: ${{ env.RESULTS_FILENAME }}
- name: Checkout Slab repo
uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683
with:
repository: zama-ai/slab
path: slab
token: ${{ secrets.FHE_ACTIONS_TOKEN }}
- name: Send data to Slab
shell: bash
run: |
python3 slab/scripts/data_sender.py ${{ env.RESULTS_FILENAME }} "${{ secrets.JOB_SECRET }}" \
--slab-url "${{ secrets.SLAB_URL }}"
- name: Slack Notification
if: ${{ failure() }}
continue-on-error: true
uses: rtCamp/action-slack-notify@c33737706dea87cd7784c687dadc9adf1be59990
env:
SLACK_COLOR: ${{ job.status }}
SLACK_MESSAGE: "PKE ZK benchmarks finished with status: ${{ job.status }}. (${{ env.ACTION_RUN_URL }})"
teardown-instance:
name: Teardown instance (pke-zk-benchmarks)
if: ${{ always() && needs.setup-instance.result != 'skipped' }}
needs: [ setup-instance, pke-zk-benchmarks ]
runs-on: ubuntu-latest
steps:
- name: Stop instance
id: stop-instance
uses: zama-ai/slab-github-runner@98f0788261a7323d5d695a883e20df36591a92b7
with:
mode: stop
github-token: ${{ secrets.SLAB_ACTION_TOKEN }}
slab-url: ${{ secrets.SLAB_BASE_URL }}
job-secret: ${{ secrets.JOB_SECRET }}
label: ${{ needs.setup-instance.outputs.runner-name }}
- name: Slack Notification
if: ${{ failure() }}
continue-on-error: true
uses: rtCamp/action-slack-notify@c33737706dea87cd7784c687dadc9adf1be59990
env:
SLACK_COLOR: ${{ job.status }}
SLACK_MESSAGE: "Instance teardown (pke-zk-benchmarks) finished with status: ${{ job.status }}. (${{ env.ACTION_RUN_URL }})"

View File

@@ -1,46 +0,0 @@
# Run cargo audit
name: cargo_audit
on:
workflow_dispatch:
schedule:
# runs every day at 4am UTC
- cron: '0 4 * * *'
env:
CARGO_TERM_COLOR: always
ACTION_RUN_URL: ${{ github.server_url }}/${{ github.repository }}/actions/runs/${{ github.run_id }}
CHECKOUT_TOKEN: ${{ secrets.REPO_CHECKOUT_TOKEN || secrets.GITHUB_TOKEN }}
SLACK_CHANNEL: ${{ secrets.SLACK_CHANNEL }}
SLACK_ICON: https://pbs.twimg.com/profile_images/1274014582265298945/OjBKP9kn_400x400.png
SLACK_USERNAME: ${{ secrets.BOT_USERNAME }}
SLACK_WEBHOOK: ${{ secrets.SLACK_WEBHOOK }}
SLACKIFY_MARKDOWN: true
permissions: {}
# zizmor: ignore[concurrency-limits] only Zama organization members and GitHub can trigger this workflow
jobs:
audit:
name: cargo_audit/audit
if: github.event_name == 'workflow_dispatch' ||
(github.event_name == 'schedule' && github.repository == 'zama-ai/tfhe-rs')
runs-on: ubuntu-latest
steps:
- uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd
with:
persist-credentials: 'false'
token: ${{ env.CHECKOUT_TOKEN }}
- name: Audit dependencies
run: |
make audit_dependencies
- name: Slack Notification
if: ${{ failure() }}
continue-on-error: true
uses: rtCamp/action-slack-notify@e31e87e03dd19038e411e38ae27cbad084a90661
env:
SLACK_COLOR: ${{ job.status }}
SLACK_MESSAGE: "cargo-audit finished with status: ${{ job.status }}. ([action run](${{ env.ACTION_RUN_URL }}))"

View File

@@ -1,4 +1,4 @@
name: cargo_build
name: Cargo Build TFHE-rs
on:
pull_request:
@@ -8,155 +8,84 @@ env:
RUSTFLAGS: "-C target-cpu=native"
RUST_BACKTRACE: "full"
RUST_MIN_STACK: "8388608"
CHECKOUT_TOKEN: ${{ secrets.REPO_CHECKOUT_TOKEN || secrets.GITHUB_TOKEN }}
concurrency:
group: ${{ github.workflow }}-${{ github.head_ref }}
cancel-in-progress: true
permissions:
contents: read
jobs:
prepare-parallel-pcc-matrix:
name: cargo_build/prepare-parallel-pcc-matrix
runs-on: ubuntu-latest
outputs:
matrix_command: ${{ steps.set-pcc-commands-matrix.outputs.commands }}
steps:
- uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd
with:
persist-credentials: "false"
token: ${{ env.CHECKOUT_TOKEN }}
cargo-builds:
runs-on: ${{ matrix.os }}
# Fetch all the Make recipes that start with `pcc_batch_`
- name: Set pcc commands matrix
id: set-pcc-commands-matrix
run: |
COMMANDS=$(grep -oE '^pcc_batch_[^:]*:' Makefile | sed 's/:/\"/; s/^/\"/' | paste -sd,)
echo "commands=[${COMMANDS}]" >> "$GITHUB_OUTPUT"
parallel-pcc-cpu:
name: cargo_build/parallel-pcc-cpu
needs: prepare-parallel-pcc-matrix
strategy:
matrix:
command: ${{ fromJson(needs.prepare-parallel-pcc-matrix.outputs.matrix_command)}}
# GitHub macos-latest are now M1 macs, so use ours, we limit what runs so it will be fast
# even with a few PRs
os: [large_ubuntu_16, macos-latest, windows-latest]
fail-fast: false
uses: ./.github/workflows/cargo_build_common.yml
with:
run-pcc-cpu-batch: ${{ matrix.command }}
secrets:
BOT_USERNAME: ${{ secrets.BOT_USERNAME }}
SLACK_CHANNEL: ${{ secrets.SLACK_CHANNEL }}
SLACK_WEBHOOK: ${{ secrets.SLACK_WEBHOOK }}
REPO_CHECKOUT_TOKEN: ${{ secrets.REPO_CHECKOUT_TOKEN }}
JOB_SECRET: ${{ secrets.JOB_SECRET }}
SLAB_ACTION_TOKEN: ${{ secrets.SLAB_ACTION_TOKEN }}
SLAB_URL: ${{ secrets.SLAB_URL }}
SLAB_BASE_URL: ${{ secrets.SLAB_BASE_URL }}
pcc-hpu:
name: cargo_build/pcc-hpu
uses: ./.github/workflows/cargo_build_common.yml
with:
run-pcc-hpu: true
secrets:
BOT_USERNAME: ${{ secrets.BOT_USERNAME }}
SLACK_CHANNEL: ${{ secrets.SLACK_CHANNEL }}
SLACK_WEBHOOK: ${{ secrets.SLACK_WEBHOOK }}
REPO_CHECKOUT_TOKEN: ${{ secrets.REPO_CHECKOUT_TOKEN }}
JOB_SECRET: ${{ secrets.JOB_SECRET }}
SLAB_ACTION_TOKEN: ${{ secrets.SLAB_ACTION_TOKEN }}
SLAB_URL: ${{ secrets.SLAB_URL }}
SLAB_BASE_URL: ${{ secrets.SLAB_BASE_URL }}
build-tfhe-full:
name: cargo_build/build-tfhe-full
uses: ./.github/workflows/cargo_build_common.yml
with:
run-build-tfhe-full: true
# GitHub macos-latest are now M1 macs, so use ours, we limit what runs so it will be fast
# even with a few PRs
extra-runners-to-use: macos-latest-xlarge,large_windows_16_latest
secrets:
BOT_USERNAME: ${{ secrets.BOT_USERNAME }}
SLACK_CHANNEL: ${{ secrets.SLACK_CHANNEL }}
SLACK_WEBHOOK: ${{ secrets.SLACK_WEBHOOK }}
REPO_CHECKOUT_TOKEN: ${{ secrets.REPO_CHECKOUT_TOKEN }}
JOB_SECRET: ${{ secrets.JOB_SECRET }}
SLAB_ACTION_TOKEN: ${{ secrets.SLAB_ACTION_TOKEN }}
SLAB_URL: ${{ secrets.SLAB_URL }}
SLAB_BASE_URL: ${{ secrets.SLAB_BASE_URL }}
build:
name: cargo_build/build
uses: ./.github/workflows/cargo_build_common.yml
with:
run-build: true
secrets:
BOT_USERNAME: ${{ secrets.BOT_USERNAME }}
SLACK_CHANNEL: ${{ secrets.SLACK_CHANNEL }}
SLACK_WEBHOOK: ${{ secrets.SLACK_WEBHOOK }}
REPO_CHECKOUT_TOKEN: ${{ secrets.REPO_CHECKOUT_TOKEN }}
JOB_SECRET: ${{ secrets.JOB_SECRET }}
SLAB_ACTION_TOKEN: ${{ secrets.SLAB_ACTION_TOKEN }}
SLAB_URL: ${{ secrets.SLAB_URL }}
SLAB_BASE_URL: ${{ secrets.SLAB_BASE_URL }}
build-layers:
name: cargo_build/build-layers
uses: ./.github/workflows/cargo_build_common.yml
with:
run-build-layers: true
secrets:
BOT_USERNAME: ${{ secrets.BOT_USERNAME }}
SLACK_CHANNEL: ${{ secrets.SLACK_CHANNEL }}
SLACK_WEBHOOK: ${{ secrets.SLACK_WEBHOOK }}
REPO_CHECKOUT_TOKEN: ${{ secrets.REPO_CHECKOUT_TOKEN }}
JOB_SECRET: ${{ secrets.JOB_SECRET }}
SLAB_ACTION_TOKEN: ${{ secrets.SLAB_ACTION_TOKEN }}
SLAB_URL: ${{ secrets.SLAB_URL }}
SLAB_BASE_URL: ${{ secrets.SLAB_BASE_URL }}
build-c-api:
name: cargo_build/build-c-api
uses: ./.github/workflows/cargo_build_common.yml
with:
run-build-c-api: true
secrets:
BOT_USERNAME: ${{ secrets.BOT_USERNAME }}
SLACK_CHANNEL: ${{ secrets.SLACK_CHANNEL }}
SLACK_WEBHOOK: ${{ secrets.SLACK_WEBHOOK }}
REPO_CHECKOUT_TOKEN: ${{ secrets.REPO_CHECKOUT_TOKEN }}
JOB_SECRET: ${{ secrets.JOB_SECRET }}
SLAB_ACTION_TOKEN: ${{ secrets.SLAB_ACTION_TOKEN }}
SLAB_URL: ${{ secrets.SLAB_URL }}
SLAB_BASE_URL: ${{ secrets.SLAB_BASE_URL }}
cargo-builds:
name: cargo_build/cargo-builds (bpr)
needs: [ parallel-pcc-cpu, pcc-hpu, build-tfhe-full, build, build-layers, build-c-api ]
if: ${{ always() }}
runs-on: ubuntu-latest
steps:
- name: Check all builds success
if: needs.parallel-pcc-cpu.outputs.builds-result == 'success' &&
needs.pcc-hpu.outputs.builds-result == 'success' &&
needs.build-tfhe-full.outputs.builds-result == 'success' &&
needs.build.outputs.builds-result == 'success' &&
needs.build-layers.outputs.builds-result == 'success' &&
needs.build-c-api.outputs.builds-result == 'success'
run: |
echo "All tfhe-rs build checks passed"
- uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683
- name: Check builds failure
if: needs.parallel-pcc-cpu.outputs.builds-result != 'success' ||
needs.pcc-hpu.outputs.builds-result != 'success' ||
needs.build-tfhe-full.outputs.builds-result != 'success' ||
needs.build.outputs.builds-result != 'success' ||
needs.build-layers.outputs.builds-result != 'success' ||
needs.build-c-api.outputs.builds-result != 'success'
- name: Install latest stable
uses: dtolnay/rust-toolchain@315e265cd78dad1e1dcf3a5074f6d6c47029d5aa
with:
toolchain: stable
- name: Install and run newline linter checks
if: ${{ contains(matrix.os, 'ubuntu') }}
run: |
echo "Some tfhe-rs build checks failed"
exit 1
wget https://github.com/fernandrone/linelint/releases/download/0.0.6/linelint-linux-amd64
echo "16b70fb7b471d6f95cbdc0b4e5dc2b0ac9e84ba9ecdc488f7bdf13df823aca4b linelint-linux-amd64" > checksum
sha256sum -c checksum || exit 1
chmod +x linelint-linux-amd64
mv linelint-linux-amd64 /usr/local/bin/linelint
make check_newline
- name: Run pcc checks
if: ${{ contains(matrix.os, 'ubuntu') }}
run: |
make pcc
- name: Build tfhe-csprng
if: ${{ contains(matrix.os, 'ubuntu') }}
run: |
make build_tfhe_csprng
- name: Build Release core
if: ${{ contains(matrix.os, 'ubuntu') }}
run: |
make build_core AVX512_SUPPORT=ON
make build_core_experimental AVX512_SUPPORT=ON
- name: Build Release boolean
if: ${{ contains(matrix.os, 'ubuntu') }}
run: |
make build_boolean
- name: Build Release shortint
if: ${{ contains(matrix.os, 'ubuntu') }}
run: |
make build_shortint
- name: Build Release integer
if: ${{ contains(matrix.os, 'ubuntu') }}
run: |
make build_integer
- name: Build Release tfhe full
run: |
make build_tfhe_full
- name: Build Release c_api
if: ${{ contains(matrix.os, 'ubuntu') }}
run: |
make build_c_api
- name: Build coverage tests
if: ${{ contains(matrix.os, 'ubuntu') }}
run: |
make build_tfhe_coverage
# The wasm build check is a bit annoying to set-up here and is done during the tests in
# aws_tfhe_tests.yml

View File

@@ -1,232 +0,0 @@
name: cargo_build_common
on:
workflow_call:
inputs:
run-pcc-cpu-batch:
type: string
run-pcc-hpu:
type: boolean
default: false
run-build:
type: boolean
default: false
run-build-layers:
type: boolean
default: false
run-build-c-api:
type: boolean
default: false
run-build-tfhe-full:
type: boolean
default: false
extra-runners-to-use: # Additional runners to run builds command against
type: string # Use comma separated values to generate an array
default: ""
outputs:
builds-result:
description: "Result of builds job"
value: ${{ jobs.builds.outputs.result }}
secrets:
REPO_CHECKOUT_TOKEN:
required: true
SLAB_ACTION_TOKEN:
required: true
SLAB_BASE_URL:
required: true
SLAB_URL:
required: true
JOB_SECRET:
required: true
SLACK_CHANNEL:
required: true
BOT_USERNAME:
required: true
SLACK_WEBHOOK:
required: true
env:
CARGO_TERM_COLOR: always
ACTION_RUN_URL: ${{ github.server_url }}/${{ github.repository }}/actions/runs/${{ github.run_id }}
RUSTFLAGS: "-C target-cpu=native"
RUST_BACKTRACE: "full"
RUST_MIN_STACK: "8388608"
SLACK_CHANNEL: ${{ secrets.SLACK_CHANNEL }}
SLACK_ICON: https://pbs.twimg.com/profile_images/1274014582265298945/OjBKP9kn_400x400.png
SLACK_USERNAME: ${{ secrets.BOT_USERNAME }}
SLACK_WEBHOOK: ${{ secrets.SLACK_WEBHOOK }}
SLACKIFY_MARKDOWN: true
CHECKOUT_TOKEN: ${{ secrets.REPO_CHECKOUT_TOKEN || secrets.GITHUB_TOKEN }}
LINELINT_VERSION: 0.0.6
LINELINT_CHECKSUM: "16b70fb7b471d6f95cbdc0b4e5dc2b0ac9e84ba9ecdc488f7bdf13df823aca4b"
permissions:
contents: read
# zizmor: ignore[concurrency-limits] caller workflow is responsible for the concurrency
jobs:
prepare-matrix:
name: cargo_build_common/prepare-matrix
if: inputs.run-pcc-cpu-batch || inputs.run-pcc-hpu || inputs.run-build || inputs.run-build-layers || inputs.run-build-tfhe-full || inputs.run-build-c-api
runs-on: ubuntu-latest
outputs:
runners: ${{ steps.set_matrix_runners.outputs.runners }}
steps:
- name: Parse runners
shell: python
env:
INPUTS_EXTRA_RUNNERS_TO_USE: ${{ inputs.extra-runners-to-use }}
REMOTE_RUNNER: "runs-on=${{ github.run_id }}/runner=cpu-small"
run: |
import os
inputs_extra_runners = os.environ["INPUTS_EXTRA_RUNNERS_TO_USE"]
remote_runner_label = os.environ["REMOTE_RUNNER"]
env_file = os.environ["GITHUB_ENV"]
runners = [remote_runner_label, ]
if inputs_extra_runners:
split_runners = inputs_extra_runners.replace(" ", "").split(",")
runners.extend(split_runners)
with open(env_file, "a") as f:
f.write(f"""RUNNERS=["{'", "'.join(runners)}"]\n""")
- name: Set matrix runners outputs
id: set_matrix_runners
run: | # zizmor: ignore[template-injection] these env variable are safe
echo "runners=${{ toJSON(env.RUNNERS) }}" >> "${GITHUB_OUTPUT}"
builds:
name: cargo_build_common/builds
needs: prepare-matrix
runs-on: ${{ matrix.runner }}
strategy:
matrix:
runner: ${{ fromJSON(needs.prepare-matrix.outputs.runners) }}
fail-fast: false
outputs:
result: ${{ steps.set_builds_result.outputs.result }}
steps:
- name: Checkout tfhe-rs repo
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd
with:
persist-credentials: 'false'
token: ${{ env.CHECKOUT_TOKEN }}
- name: Install latest stable
uses: dtolnay/rust-toolchain@e97e2d8cc328f1b50210efc529dca0028893a2d9 # zizmor: ignore[stale-action-refs] this action doesn't create releases
with:
toolchain: stable
- name: Install newline linter
if: inputs.run-build && matrix.runner == env.EXTERNAL_CONTRIBUTION_RUNNER
run: |
wget "https://github.com/fernandrone/linelint/releases/download/${LINELINT_VERSION}/linelint-linux-amd64"
echo "${LINELINT_CHECKSUM} linelint-linux-amd64" > checksum
sha256sum -c checksum
chmod +x linelint-linux-amd64
ln -s "$(pwd)/linelint-linux-amd64" /usr/local/bin/linelint
- name: Get Node version
if: inputs.run-pcc-cpu-batch == 'pcc_batch_2'
run: |
echo "NODE_VERSION=$(make node_version)" >> "${GITHUB_ENV}"
- name: Node cache restoration
if: inputs.run-pcc-cpu-batch == 'pcc_batch_2'
id: node-cache
uses: actions/cache/restore@27d5ce7f107fe9357f9df03efb73ab90386fccae #v5.0.5
with:
path: |
~/.nvm
~/.npm
key: node-${{ env.NODE_VERSION }}
- name: Install Node
if: inputs.run-pcc-cpu-batch == 'pcc_batch_2'
run: |
make install_node
- name: Node cache save
uses: actions/cache/save@27d5ce7f107fe9357f9df03efb73ab90386fccae #v5.0.5
if: inputs.run-pcc-cpu-batch == 'pcc_batch_2' && steps.node-cache.outputs.cache-hit != 'true'
with:
path: |
~/.nvm
~/.npm
key: node-${{ env.NODE_VERSION }}
- name: Run pcc checks batch
if: inputs.run-pcc-cpu-batch
run: |
make "${COMMAND}"
env:
COMMAND: ${{ inputs.run-pcc-cpu-batch }}
- name: Run Hpu pcc checks
if: inputs.run-pcc-hpu
run: |
make pcc_hpu
- name: Build Release tfhe full
if: inputs.run-build-tfhe-full
run: |
make build_tfhe_full
- name: Run newline linter checks
if: inputs.run-build
run: |
make check_newline
- name: Build tfhe-csprng
if: inputs.run-build
run: |
make build_tfhe_csprng
- name: Build with MSRV
if: inputs.run-build
run: |
make build_tfhe_msrv
- name: Build coverage tests
if: inputs.run-build
run: |
make build_tfhe_coverage
- name: Build Release core
if: inputs.run-build-layers
run: |
make build_core AVX512_SUPPORT=ON
make build_core_experimental AVX512_SUPPORT=ON
- name: Build Release boolean
if: inputs.run-build-layers
run: |
make build_boolean
- name: Build Release shortint
if: inputs.run-build-layers
run: |
make build_shortint
- name: Build Release integer
if: inputs.run-build-layers
run: |
make build_integer
- name: Build Release c_api
if: inputs.run-build-c-api
run: |
make build_c_api
# The wasm build check is a bit annoying to set-up here and is done during the tests in
# aws_tfhe_tests.yml
- name: Set result output
id: set_builds_result
if: ${{ always() }}
run: | # zizmor: ignore[template-injection] this context variable is safe
echo "result=${{ job.status }}" >> "${GITHUB_OUTPUT}"

View File

@@ -1,23 +1,18 @@
# Build tfhe-fft
name: cargo_build_tfhe_fft
name: Cargo Build tfhe-fft
on:
pull_request:
env:
CARGO_TERM_COLOR: always
CHECKOUT_TOKEN: ${{ secrets.REPO_CHECKOUT_TOKEN || secrets.GITHUB_TOKEN }}
concurrency:
group: ${{ github.workflow }}-${{ github.head_ref }}
cancel-in-progress: true
permissions:
contents: read
jobs:
cargo-builds-fft:
name: cargo_build_tfhe_fft/cargo-builds-fft (bpr)
runs-on: ${{ matrix.runner_type }}
strategy:
@@ -26,13 +21,10 @@ jobs:
fail-fast: false
steps:
- uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd
with:
persist-credentials: 'false'
token: ${{ env.CHECKOUT_TOKEN }}
- uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683
- name: Install Rust
uses: dtolnay/rust-toolchain@e97e2d8cc328f1b50210efc529dca0028893a2d9 # zizmor: ignore[stale-action-refs] this action doesn't create releases
uses: actions-rs/toolchain@16499b5e05bf2e26879000db0c1d13f7e13fa3af
with:
toolchain: stable
override: true

View File

@@ -1,36 +1,28 @@
# Build tfhe-ntt
name: cargo_build_tfhe_ntt
name: Cargo Build tfhe-ntt
on:
pull_request:
env:
CARGO_TERM_COLOR: always
CHECKOUT_TOKEN: ${{ secrets.REPO_CHECKOUT_TOKEN || secrets.GITHUB_TOKEN }}
concurrency:
group: ${{ github.workflow }}-${{ github.head_ref }}
cancel-in-progress: true
permissions:
contents: read
jobs:
cargo-builds-ntt:
name: cargo_build_tfhe_ntt/cargo-builds-ntt (bpr)
cargo-builds:
runs-on: ${{ matrix.os }}
strategy:
matrix:
os: [ubuntu-latest, macos-latest, windows-latest]
fail-fast: false
steps:
- uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd
with:
persist-credentials: 'false'
token: ${{ env.CHECKOUT_TOKEN }}
- uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683
- name: Install Rust
uses: dtolnay/rust-toolchain@e97e2d8cc328f1b50210efc529dca0028893a2d9 # zizmor: ignore[stale-action-refs] this action doesn't create releases
uses: actions-rs/toolchain@16499b5e05bf2e26879000db0c1d13f7e13fa3af
with:
toolchain: stable
override: true

View File

@@ -1,132 +1,71 @@
# Test tfhe-fft
name: cargo_test_fft
name: Cargo Test tfhe-fft
on:
workflow_dispatch:
pull_request:
push:
branches:
- main
env:
CARGO_TERM_COLOR: always
IS_PULL_REQUEST: ${{ github.event_name == 'pull_request' }}
CHECKOUT_TOKEN: ${{ secrets.REPO_CHECKOUT_TOKEN || secrets.GITHUB_TOKEN }}
concurrency:
group: ${{ github.workflow }}-${{ github.head_ref }}${{ github.ref == 'refs/heads/main' && github.sha || '' }}
group: ${{ github.workflow }}-${{ github.head_ref }}
cancel-in-progress: true
permissions:
contents: read
jobs:
should-run:
name: cargo_test_fft/should-run
if: github.event_name != 'push' ||
(github.event_name == 'push' && github.repository == 'zama-ai/tfhe-rs')
runs-on: ubuntu-latest
permissions:
pull-requests: read # Needed to check for file change
outputs:
fft_test: ${{ env.IS_PULL_REQUEST == 'false' || steps.changed-files.outputs.fft_any_changed }}
steps:
- name: Checkout tfhe-rs
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd
with:
fetch-depth: 0
persist-credentials: 'false'
token: ${{ env.CHECKOUT_TOKEN }}
- name: Check for file changes
id: changed-files
uses: tj-actions/changed-files@9426d40962ed5378910ee2e21d5f8c6fcbf2dd96 # v47.0.6
with:
files_yaml: |
fft:
- tfhe/Cargo.toml
- Makefile
- tfhe-fft/**
- '.github/workflows/cargo_test_fft.yml'
cargo-tests-fft:
name: cargo_test_fft/cargo-tests-fft
needs: should-run
if: needs.should-run.outputs.fft_test == 'true'
cargo-tests:
runs-on: ${{ matrix.runner_type }}
strategy:
matrix:
runner_type: [ ubuntu-latest, macos-latest, windows-latest ]
runner_type: [ubuntu-latest, macos-latest, windows-latest]
fail-fast: false
steps:
- uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd
with:
persist-credentials: 'false'
token: ${{ env.CHECKOUT_TOKEN }}
- uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683
- name: Install Rust
uses: dtolnay/rust-toolchain@e97e2d8cc328f1b50210efc529dca0028893a2d9 # zizmor: ignore[stale-action-refs] this action doesn't create releases
uses: actions-rs/toolchain@16499b5e05bf2e26879000db0c1d13f7e13fa3af
with:
toolchain: stable
override: true
- name: Test avx2
- name: Test debug
run: |
make test_fft
- name: Test serialization
run: make test_fft_serde
- name: Test no-std avx2
- name: Test no-std
run: |
make test_fft_no_std
- name: Test avx512
run: |
make test_fft_avx512
- name: Test no-std avx512
run: |
make test_fft_no_std_avx512
cargo-tests-fft-node-js:
name: cargo_test_fft/cargo-tests-fft-node-js
needs: should-run
if: needs.should-run.outputs.fft_test == 'true'
runs-on: ubuntu-latest
cargo-tests-nightly:
runs-on: ${{ matrix.runner_type }}
strategy:
matrix:
runner_type: [ubuntu-latest, macos-latest, windows-latest]
steps:
- uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd
- uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683
- name: Install Rust
uses: actions-rs/toolchain@16499b5e05bf2e26879000db0c1d13f7e13fa3af
with:
persist-credentials: 'false'
token: ${{ env.CHECKOUT_TOKEN }}
toolchain: nightly
override: true
- name: Test nightly
run: |
make test_fft_nightly
- name: Test no-std nightly
run: |
make test_fft_no_std_nightly
cargo-tests-node-js:
runs-on: "ubuntu-latest"
steps:
- uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683
- name: Test node js
run: |
make install_node
make test_fft_node_js_ci
cargo-tests-fft-successful:
name: cargo_test_fft/cargo-tests-fft-successful (bpr)
needs: [ should-run, cargo-tests-fft, cargo-tests-fft-node-js ]
if: ${{ always() }}
runs-on: ubuntu-latest
steps:
- name: Tests do not need to run
if: needs.should-run.outputs.fft_test == 'false'
run: |
echo "tfhe-fft files haven't changed tests don't need to run"
- name: Check all tests passed
if: needs.should-run.outputs.fft_test == 'true' &&
needs.cargo-tests-fft.result == 'success' &&
needs.cargo-tests-fft-node-js.result == 'success'
run: |
echo "All tfhe-fft test passed"
- name: Check tests failure
if: needs.should-run.outputs.fft_test == 'true' &&
(needs.cargo-tests-fft.result != 'success' ||
needs.cargo-tests-fft-node-js.result != 'success')
run: |
echo "Some tfhe-fft tests failed"
exit 1

View File

@@ -1,163 +1,54 @@
# Test tfhe-ntt
name: cargo_test_ntt
name: Cargo Test tfhe-ntt
on:
workflow_dispatch:
pull_request:
push:
branches:
- main
env:
CARGO_TERM_COLOR: always
ACTION_RUN_URL: ${{ github.server_url }}/${{ github.repository }}/actions/runs/${{ github.run_id }}
IS_PULL_REQUEST: ${{ github.event_name == 'pull_request' }}
CHECKOUT_TOKEN: ${{ secrets.REPO_CHECKOUT_TOKEN || secrets.GITHUB_TOKEN }}
SECRETS_AVAILABLE: ${{ secrets.JOB_SECRET != '' }}
concurrency:
group: ${{ github.workflow }}-${{ github.head_ref }}${{ github.ref == 'refs/heads/main' && github.sha || '' }}
group: ${{ github.workflow }}-${{ github.head_ref }}
cancel-in-progress: true
permissions:
contents: read
jobs:
should-run:
name: cargo_test_ntt/should-run
if: github.event_name != 'push' ||
(github.event_name == 'push' && github.repository == 'zama-ai/tfhe-rs')
runs-on: ubuntu-latest
permissions:
pull-requests: read # Needed to check for file change
outputs:
ntt_test: ${{ env.IS_PULL_REQUEST == 'false' || steps.changed-files.outputs.ntt_any_changed }}
steps:
- name: Checkout tfhe-rs
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd
with:
fetch-depth: 0
persist-credentials: "false"
token: ${{ env.CHECKOUT_TOKEN }}
- name: Check for file changes
id: changed-files
uses: tj-actions/changed-files@9426d40962ed5378910ee2e21d5f8c6fcbf2dd96 # v47.0.6
with:
files_yaml: |
ntt:
- tfhe/Cargo.toml
- Makefile
- tfhe-ntt/**
- '.github/workflows/cargo_test_ntt.yml'
setup-instance:
name: cargo_test_ntt/setup-instance
needs: should-run
if: needs.should-run.outputs.ntt_test == 'true'
runs-on: ubuntu-latest
outputs:
matrix_os: ${{ steps.set-os-matrix.outputs.matrix_os }}
runner-name: ${{ steps.start-remote-instance.outputs.label }}
steps:
- name: Start remote instance
id: start-remote-instance
if: env.SECRETS_AVAILABLE == 'true'
uses: zama-ai/slab-github-runner@5aee5d157f4a0201e5eaefc9cc648e5f9f5472a5 # v1.6.0
with:
mode: start
github-token: ${{ secrets.SLAB_ACTION_TOKEN }}
slab-url: ${{ secrets.SLAB_BASE_URL }}
job-secret: ${{ secrets.JOB_SECRET }}
backend: aws
profile: cpu-small
- name: Set os matrix
id: set-os-matrix
env:
SLAB_INSTANCE: ${{ steps.start-remote-instance.outputs.label }}
run: |
INSTANCE_TO_USE="${SLAB_INSTANCE:-ubuntu-latest}"
echo "matrix_os=[\"${INSTANCE_TO_USE}\", \"macos-latest\", \"windows-latest\"]" >> "$GITHUB_OUTPUT"
cargo-tests-ntt:
name: cargo_test_ntt/cargo-tests-ntt
needs: [should-run, setup-instance]
if: needs.should-run.outputs.ntt_test == 'true'
cargo-tests:
runs-on: ${{ matrix.os }}
strategy:
matrix:
os: ${{fromJson(needs.setup-instance.outputs.matrix_os)}}
os: [ubuntu-latest, macos-latest, windows-latest]
fail-fast: false
steps:
- uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd
with:
persist-credentials: "false"
token: ${{ env.CHECKOUT_TOKEN }}
- uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683
- name: Install Rust
uses: dtolnay/rust-toolchain@e97e2d8cc328f1b50210efc529dca0028893a2d9 # zizmor: ignore[stale-action-refs] this action doesn't create releases
uses: actions-rs/toolchain@16499b5e05bf2e26879000db0c1d13f7e13fa3af
with:
toolchain: stable
override: true
- name: Test avx2
- name: Test debug
run: make test_ntt
- name: Test no-std avx2
- name: Test no-std
run: make test_ntt_no_std
- name: Test avx512
run: make test_ntt_avx512
- name: Test no-std avx512
run: make test_ntt_no_std_avx512
cargo-tests-ntt-successful:
name: cargo_test_ntt/cargo-tests-ntt-successful (bpr)
needs: [should-run, cargo-tests-ntt]
if: ${{ always() }}
runs-on: ubuntu-latest
cargo-tests-nightly:
runs-on: ${{ matrix.os }}
strategy:
matrix:
os: [ubuntu-latest, macos-latest, windows-latest]
steps:
- name: Tests do not need to run
if: needs.should-run.outputs.ntt_test == 'false'
run: |
echo "tfhe-ntt files haven't changed tests don't need to run"
- uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683
- name: Check all tests success
if: needs.should-run.outputs.ntt_test == 'true' &&
needs.cargo-tests-ntt.result == 'success'
run: |
echo "All tfhe-ntt tests passed"
- name: Check tests failure
if: needs.should-run.outputs.ntt_test == 'true' &&
needs.cargo-tests-ntt.result != 'success'
run: |
echo "Some tfhe-ntt tests failed"
exit 1
teardown-instance:
name: cargo_test_ntt/teardown-instance
if: ${{ always() && needs.setup-instance.result == 'success' }}
needs: [setup-instance, cargo-tests-ntt-successful]
runs-on: ubuntu-latest
steps:
- name: Stop remote instance
id: stop-instance
if: env.SECRETS_AVAILABLE == 'true'
uses: zama-ai/slab-github-runner@5aee5d157f4a0201e5eaefc9cc648e5f9f5472a5 # v1.6.0
- name: Install Rust
uses: actions-rs/toolchain@16499b5e05bf2e26879000db0c1d13f7e13fa3af
with:
mode: stop
github-token: ${{ secrets.SLAB_ACTION_TOKEN }}
slab-url: ${{ secrets.SLAB_BASE_URL }}
job-secret: ${{ secrets.JOB_SECRET }}
label: ${{ needs.setup-instance.outputs.runner-name }}
toolchain: nightly
override: true
- name: Slack Notification
if: ${{ failure() }}
continue-on-error: true
uses: rtCamp/action-slack-notify@e31e87e03dd19038e411e38ae27cbad084a90661
env:
SLACK_COLOR: ${{ job.status }}
SLACK_MESSAGE: "Instance teardown (cargo-tests-ntt) finished with status: ${{ job.status }}. (${{ env.ACTION_RUN_URL }})"
- name: Test nightly
run: make test_ntt_nightly
- name: Test no-std nightly
run: make test_ntt_no_std_nightly

View File

@@ -1,24 +1,16 @@
# Check commit and PR compliance
name: check_commit
name: Check commit and PR compliance
on:
pull_request:
permissions: {}
# zizmor: ignore[concurrency-limits] only Zama organization members can trigger this workflow (via manual approval for PR from forks)
jobs:
check-commit-pr:
name: check_commit/check-commit-pr (bpr)
name: Check commit and PR
runs-on: ubuntu-latest
permissions:
contents: read
pull-requests: write # Permission needed to scan commits in a pull-request and write issue comment
steps:
- name: Check first line
uses: gsactions/commit-message-checker@16fa2d5de096ae0d35626443bcd24f1e756cafee
with:
pattern: '^((feat|fix|chore|refactor|style|test|docs|doc|perf)(\([\w\-_]+\))?\!?\:) .+$'
pattern: '^((feat|fix|chore|refactor|style|test|docs|doc)(\([\w\-_]+\))?\!?\:) .+$'
flags: "gs"
error: 'Your first line has to contain a commit type and scope like "feat(my_feature): msg".'
excludeDescription: "true" # optional: this excludes the description body of a pull request

View File

@@ -1,57 +1,33 @@
# Lint and check CI
name: ci_lint
name: CI Lint and Checks
on:
pull_request:
env:
ACTIONLINT_VERSION: 1.7.7
ACTIONLINT_CHECKSUM: "023070a287cd8cccd71515fedc843f1985bf96c436b7effaecce67290e7e0757"
CHECKOUT_TOKEN: ${{ secrets.REPO_CHECKOUT_TOKEN || secrets.GITHUB_TOKEN }}
permissions:
contents: read
# zizmor: ignore[concurrency-limits] only Zama organization members can trigger this workflow (via manual approval for PR from forks)
ACTIONLINT_VERSION: 1.6.27
jobs:
lint-check:
name: ci_lint/lint-check (bpr)
name: Lint and checks
runs-on: ubuntu-latest
steps:
- name: Checkout tfhe-rs
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd
with:
persist-credentials: 'false'
token: ${{ env.CHECKOUT_TOKEN }}
uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683
- name: Get actionlint
run: |
wget "https://github.com/rhysd/actionlint/releases/download/v${ACTIONLINT_VERSION}/actionlint_${ACTIONLINT_VERSION}_linux_amd64.tar.gz"
echo "${ACTIONLINT_CHECKSUM} actionlint_${ACTIONLINT_VERSION}_linux_amd64.tar.gz" > checksum
bash <(curl https://raw.githubusercontent.com/rhysd/actionlint/main/scripts/download-actionlint.bash) ${{ env.ACTIONLINT_VERSION }}
echo "f2ee6d561ce00fa93aab62a7791c1a0396ec7e8876b2a8f2057475816c550782 actionlint" > checksum
sha256sum -c checksum
tar -xf actionlint_"${ACTIONLINT_VERSION}"_linux_amd64.tar.gz actionlint
ln -s "$(pwd)/actionlint" /usr/local/bin/
- name: Lint workflows
run: |
make lint_workflow
- name: Get Zimzor version to use
id: get_zizmor
run: |
echo "version=$(make zizmor_version)" >> "${GITHUB_OUTPUT}"
- name: Check workflows security
uses: zizmorcore/zizmor-action@b1d7e1fb5de872772f31590499237e7cce841e8e # v0.5.3
with:
advanced-security: 'false' # Print results directly in logs
persona: pedantic
version: ${{ steps.get_zizmor.outputs.version }}
- name: Ensure SHA pinned actions
uses: zgosalvez/github-actions-ensure-sha-pinned-actions@ca46236c6ce584ae24bc6283ba8dcf4b3ec8a066 # v5.0.4
uses: zgosalvez/github-actions-ensure-sha-pinned-actions@5d6ac37a4cef8b8df67f482a8e384987766f0213 # v3.0.17
with:
allowlist: |
slsa-framework/slsa-github-generator
./

View File

@@ -1,4 +1,4 @@
name: code_coverage
name: Code Coverage
env:
CARGO_TERM_COLOR: always
@@ -10,41 +10,50 @@ env:
SLACK_ICON: https://pbs.twimg.com/profile_images/1274014582265298945/OjBKP9kn_400x400.png
SLACK_USERNAME: ${{ secrets.BOT_USERNAME }}
SLACK_WEBHOOK: ${{ secrets.SLACK_WEBHOOK }}
CHECKOUT_TOKEN: ${{ secrets.REPO_CHECKOUT_TOKEN || secrets.GITHUB_TOKEN }}
on:
# Allows you to run this workflow manually from the Actions tab as an alternative.
workflow_dispatch:
# Code coverage workflow is only run via workflow_dispatch event since execution duration is not stabilized yet.
permissions:
contents: read
# zizmor: ignore[concurrency-limits] only Zama organization members can trigger this workflow
jobs:
code-coverage-tests:
name: code_coverage/code-coverage-tests
setup-instance:
name: Setup instance (code-coverage)
runs-on: ubuntu-latest
outputs:
runner-name: ${{ steps.start-instance.outputs.label }}
steps:
- name: Start instance
id: start-instance
uses: zama-ai/slab-github-runner@98f0788261a7323d5d695a883e20df36591a92b7
with:
mode: start
github-token: ${{ secrets.SLAB_ACTION_TOKEN }}
slab-url: ${{ secrets.SLAB_BASE_URL }}
job-secret: ${{ secrets.JOB_SECRET }}
backend: aws
profile: cpu-small
code-coverage:
name: Code coverage tests
needs: setup-instance
concurrency:
group: ${{ github.workflow_ref }}_${{ github.event_name }}
group: ${{ github.workflow }}_${{ github.event_name }}_${{ github.ref }}
cancel-in-progress: true
runs-on: "runs-on=${{ github.run_id }}/runner=cpu-small"
runs-on: ${{ needs.setup-instance.outputs.runner-name }}
timeout-minutes: 5760 # 4 days
steps:
- name: Checkout tfhe-rs
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd
with:
persist-credentials: 'false'
token: ${{ env.CHECKOUT_TOKEN }}
uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683
- name: Install latest stable
uses: dtolnay/rust-toolchain@e97e2d8cc328f1b50210efc529dca0028893a2d9 # zizmor: ignore[stale-action-refs] this action doesn't create releases
uses: dtolnay/rust-toolchain@315e265cd78dad1e1dcf3a5074f6d6c47029d5aa
with:
toolchain: stable
- name: Check for file changes
id: changed-files
uses: tj-actions/changed-files@9426d40962ed5378910ee2e21d5f8c6fcbf2dd96 # v47.0.6
uses: tj-actions/changed-files@bab30c2299617f6615ec02a68b9a40d10bd21366
with:
files_yaml: |
tfhe:
@@ -74,7 +83,7 @@ jobs:
make test_shortint_cov
- name: Upload tfhe coverage to Codecov
uses: codecov/codecov-action@57e3a136b779b570ffcdbf80b3bdc90e7fab3de2
uses: codecov/codecov-action@7f8b4b4bde536c465e797be725718b88c5d95e0e
if: steps.changed-files.outputs.tfhe_any_changed == 'true'
with:
token: ${{ secrets.CODECOV_TOKEN }}
@@ -88,7 +97,7 @@ jobs:
make test_integer_cov
- name: Upload tfhe coverage to Codecov
uses: codecov/codecov-action@57e3a136b779b570ffcdbf80b3bdc90e7fab3de2
uses: codecov/codecov-action@7f8b4b4bde536c465e797be725718b88c5d95e0e
if: steps.changed-files.outputs.tfhe_any_changed == 'true'
with:
token: ${{ secrets.CODECOV_TOKEN }}
@@ -97,9 +106,33 @@ jobs:
files: integer/cobertura.xml
- name: Slack Notification
if: ${{ failure() || (cancelled() && github.event_name != 'pull_request') }}
if: ${{ failure() }}
continue-on-error: true
uses: rtCamp/action-slack-notify@e31e87e03dd19038e411e38ae27cbad084a90661
uses: rtCamp/action-slack-notify@c33737706dea87cd7784c687dadc9adf1be59990
env:
SLACK_COLOR: ${{ job.status }}
SLACK_MESSAGE: "Code coverage finished with status: ${{ job.status }}. (${{ env.ACTION_RUN_URL }})"
teardown-instance:
name: Teardown instance (code-coverage)
if: ${{ always() && needs.setup-instance.result != 'skipped' }}
needs: [ setup-instance, code-coverage ]
runs-on: ubuntu-latest
steps:
- name: Stop instance
id: stop-instance
uses: zama-ai/slab-github-runner@98f0788261a7323d5d695a883e20df36591a92b7
with:
mode: stop
github-token: ${{ secrets.SLAB_ACTION_TOKEN }}
slab-url: ${{ secrets.SLAB_BASE_URL }}
job-secret: ${{ secrets.JOB_SECRET }}
label: ${{ needs.setup-instance.outputs.runner-name }}
- name: Slack Notification
if: ${{ failure() }}
continue-on-error: true
uses: rtCamp/action-slack-notify@c33737706dea87cd7784c687dadc9adf1be59990
env:
SLACK_COLOR: ${{ job.status }}
SLACK_MESSAGE: "Instance teardown (code-coverage) finished with status: ${{ job.status }}. (${{ env.ACTION_RUN_URL }})"

View File

@@ -1,4 +1,4 @@
name: csprng_randomness_tests
name: CSPRNG randomness testing Workflow
env:
CARGO_TERM_COLOR: always
@@ -10,10 +10,6 @@ env:
SLACK_ICON: https://pbs.twimg.com/profile_images/1274014582265298945/OjBKP9kn_400x400.png
SLACK_USERNAME: ${{ secrets.BOT_USERNAME }}
SLACK_WEBHOOK: ${{ secrets.SLACK_WEBHOOK }}
SLACKIFY_MARKDOWN: true
PULL_REQUEST_MD_LINK: ""
CHECKOUT_TOKEN: ${{ secrets.REPO_CHECKOUT_TOKEN || secrets.GITHUB_TOKEN }}
on:
# Allows you to run this workflow manually from the Actions tab as an alternative.
@@ -21,59 +17,41 @@ on:
pull_request:
types: [ labeled ]
permissions:
contents: read
# zizmor: ignore[concurrency-limits] concurrency is managed after instance setup to ensure safe provisioning
jobs:
should-run:
name: csprng_randomness_tests/should-run
if: github.event_name == 'workflow_dispatch' || contains(github.event.label.name, 'approved')
setup-instance:
name: Setup instance (csprng-randomness-tests)
if: ${{ github.event_name == 'workflow_dispatch' || contains(github.event.label.name, 'approved') }}
runs-on: ubuntu-latest
permissions:
pull-requests: read # Needed to check for file change
outputs:
csprng_test: ${{ github.event_name == 'workflow_dispatch' ||
steps.changed-files.outputs.csprng_any_changed }}
runner-name: ${{ steps.start-instance.outputs.label }}
steps:
- name: Checkout tfhe-rs
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd
- name: Start instance
id: start-instance
uses: zama-ai/slab-github-runner@98f0788261a7323d5d695a883e20df36591a92b7
with:
fetch-depth: 0
persist-credentials: 'false'
token: ${{ env.CHECKOUT_TOKEN }}
- name: Check for file changes
id: changed-files
uses: tj-actions/changed-files@9426d40962ed5378910ee2e21d5f8c6fcbf2dd96 # v47.0.6
with:
files_yaml: |
csprng:
- Cargo.toml
- tfhe/Cargo.toml
- tfhe-csprng/**
- utils/tfhe-versionable/**
- .github/workflows/csprng_randomness_tests.yml
mode: start
github-token: ${{ secrets.SLAB_ACTION_TOKEN }}
slab-url: ${{ secrets.SLAB_BASE_URL }}
job-secret: ${{ secrets.JOB_SECRET }}
backend: aws
profile: cpu-small
csprng-randomness-tests:
name: csprng_randomness_tests/csprng-randomness-tests
needs: should-run
if: github.event_name == 'workflow_dispatch' ||
(contains(github.event.label.name, 'approved') && needs.should-run.outputs.csprng_test == 'true')
name: CSPRNG randomness tests
needs: setup-instance
concurrency:
group: ${{ github.workflow_ref }}_${{ github.sha }}_${{ github.event_name }}
group: ${{ github.workflow }}_${{ github.ref }}
cancel-in-progress: true
runs-on: "runs-on=${{ github.run_id }}/runner=cpu-small"
runs-on: ${{ needs.setup-instance.outputs.runner-name }}
steps:
- name: Checkout tfhe-rs
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd
uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683
with:
persist-credentials: 'false'
token: ${{ env.CHECKOUT_TOKEN }}
token: ${{ secrets.FHE_ACTIONS_TOKEN }}
- name: Install latest stable
uses: dtolnay/rust-toolchain@e97e2d8cc328f1b50210efc529dca0028893a2d9 # zizmor: ignore[stale-action-refs] this action doesn't create releases
uses: dtolnay/rust-toolchain@315e265cd78dad1e1dcf3a5074f6d6c47029d5aa
with:
toolchain: stable
@@ -81,18 +59,34 @@ jobs:
run: |
make dieharder_csprng
- name: Set pull-request URL
if: ${{ failure() && github.event_name == 'pull_request' }}
run: |
echo "PULL_REQUEST_MD_LINK=[pull-request](${PR_BASE_URL}${PR_NUMBER}), " >> "${GITHUB_ENV}"
env:
PR_BASE_URL: ${{ vars.PR_BASE_URL }}
PR_NUMBER: ${{ github.event.pull_request.number }}
- name: Slack Notification
if: ${{ failure() || (cancelled() && github.event_name != 'pull_request') }}
if: ${{ failure() }}
continue-on-error: true
uses: rtCamp/action-slack-notify@e31e87e03dd19038e411e38ae27cbad084a90661
uses: rtCamp/action-slack-notify@c33737706dea87cd7784c687dadc9adf1be59990
env:
SLACK_COLOR: ${{ job.status }}
SLACK_MESSAGE: "tfhe-csprng randomness check finished with status: ${{ job.status }}. (${{ env.PULL_REQUEST_MD_LINK }}[action run](${{ env.ACTION_RUN_URL }}))"
SLACK_MESSAGE: "tfhe-csprng randomness check finished with status: ${{ job.status }}. (${{ env.ACTION_RUN_URL }})"
teardown-instance:
name: Teardown instance (csprng-randomness-tests)
if: ${{ always() && needs.setup-instance.result != 'skipped' }}
needs: [ setup-instance, csprng-randomness-tests ]
runs-on: ubuntu-latest
steps:
- name: Stop instance
id: stop-instance
uses: zama-ai/slab-github-runner@98f0788261a7323d5d695a883e20df36591a92b7
with:
mode: stop
github-token: ${{ secrets.SLAB_ACTION_TOKEN }}
slab-url: ${{ secrets.SLAB_BASE_URL }}
job-secret: ${{ secrets.JOB_SECRET }}
label: ${{ needs.setup-instance.outputs.runner-name }}
- name: Slack Notification
if: ${{ failure() }}
continue-on-error: true
uses: rtCamp/action-slack-notify@c33737706dea87cd7784c687dadc9adf1be59990
env:
SLACK_COLOR: ${{ job.status }}
SLACK_MESSAGE: "Instance teardown (csprng-randomness-tests) finished with status: ${{ job.status }}. (${{ env.ACTION_RUN_URL }})"

123
.github/workflows/data_pr_close.yml vendored Normal file
View File

@@ -0,0 +1,123 @@
name: Close or Merge corresponding PR on the data repo
# When a PR with the data_PR tag is closed or merged, this will close the corresponding PR in the data repo.
env:
TARGET_REPO_API_URL: ${{ github.api_url }}/repos/zama-ai/tfhe-backward-compat-data
SLACK_CHANNEL: ${{ secrets.SLACK_CHANNEL }}
SLACK_ICON: https://pbs.twimg.com/profile_images/1274014582265298945/OjBKP9kn_400x400.png
SLACK_USERNAME: ${{ secrets.BOT_USERNAME }}
SLACK_WEBHOOK: ${{ secrets.SLACK_WEBHOOK }}
PR_BRANCH: ${{ github.head_ref || github.ref_name }}
CLOSE_TYPE: ${{ github.event.pull_request.merged && 'merge' || 'close' }}
# only trigger on pull request closed events
on:
pull_request:
types: [ closed ]
# The same pattern is used for jobs that use the github api:
# - save the result of the API call in the env var "GH_API_RES". Since the var is multiline
# we use this trick: https://docs.github.com/en/actions/using-workflows/workflow-commands-for-github-actions#example-of-a-multiline-string
# - "set +e" will make sure we reach the last "echo EOF" even in case of error
# - "set -o" pipefail makes one line piped command return the error of the first failure
# - 'RES="$?"' and 'exit $RES' are used to return the error code if a command failed. Without it, with "set +e"
# the script will always return 0 because of the "echo EOF".
jobs:
auto_close_job:
if: ${{ contains(github.event.pull_request.labels.*.name, 'data_PR') }}
runs-on: ubuntu-latest
steps:
- name: Find corresponding Pull Request in the data repo
run: |
{
set +e
set -o pipefail
echo 'TARGET_REPO_PR<<EOF'
curl --fail-with-body --no-progress-meter -L -X GET \
-H "Accept: application/vnd.github+json" \
-H "X-GitHub-Api-Version: 2022-11-28" \
${{ env.TARGET_REPO_API_URL }}/pulls\?head=${{ github.repository_owner }}:${{ env.PR_BRANCH }} | jq -e '.[0]' | sed 's/null/{ "message": "corresponding PR not found" }/'
RES="$?"
echo EOF
} >> "${GITHUB_ENV}"
exit $RES
- name: Comment on the PR to indicate the reason of the close
run: |
{
set +e
set -o pipefail
echo 'GH_API_RES<<EOF'
curl --fail-with-body --no-progress-meter -L -X POST \
-H "Accept: application/vnd.github+json" \
-H "Authorization: Bearer ${{ secrets.FHE_ACTIONS_TOKEN }}" \
-H "X-GitHub-Api-Version: 2022-11-28" \
${{ fromJson(env.TARGET_REPO_PR).comments_url }} \
-d '{ "body": "PR ${{ env.CLOSE_TYPE }}d because the corresponding PR in main repo was ${{ env.CLOSE_TYPE }}d: ${{ github.repository }}#${{ github.event.number }}" }'
RES="$?"
echo EOF
} >> "${GITHUB_ENV}"
exit $RES
- name: Merge the Pull Request in the data repo
if: ${{ github.event.pull_request.merged }}
run: |
{
set +e
set -o pipefail
echo 'GH_API_RES<<EOF'
curl --fail-with-body --no-progress-meter -L -X PUT \
-H "Accept: application/vnd.github+json" \
-H "Authorization: Bearer ${{ secrets.FHE_ACTIONS_TOKEN }}" \
-H "X-GitHub-Api-Version: 2022-11-28" \
${{ fromJson(env.TARGET_REPO_PR).url }}/merge \
-d '{ "merge_method": "rebase" }'
RES="$?"
echo EOF
} >> "${GITHUB_ENV}"
exit $RES
- name: Close the Pull Request in the data repo
if: ${{ !github.event.pull_request.merged }}
run: |
{
set +e
set -o pipefail
echo 'GH_API_RES<<EOF'
curl --fail-with-body --no-progress-meter -L -X PATCH \
-H "Accept: application/vnd.github+json" \
-H "Authorization: Bearer ${{ secrets.FHE_ACTIONS_TOKEN }}" \
-H "X-GitHub-Api-Version: 2022-11-28" \
${{ fromJson(env.TARGET_REPO_PR).url }} \
-d '{ "state": "closed" }'
RES="$?"
echo EOF
} >> "${GITHUB_ENV}"
exit $RES
- name: Delete the associated branch in the data repo
run: |
{
set +e
set -o pipefail
echo 'GH_API_RES<<EOF'
curl --fail-with-body --no-progress-meter -L -X DELETE \
-H "Accept: application/vnd.github+json" \
-H "Authorization: Bearer ${{ secrets.FHE_ACTIONS_TOKEN }}" \
-H "X-GitHub-Api-Version: 2022-11-28" \
${{ env.TARGET_REPO_API_URL }}/git/refs/heads/${{ env.PR_BRANCH }}
RES="$?"
echo EOF
} >> "${GITHUB_ENV}"
exit $RES
- name: Slack Notification
if: ${{ always() && job.status == 'failure' }}
continue-on-error: true
uses: rtCamp/action-slack-notify@c33737706dea87cd7784c687dadc9adf1be59990
env:
SLACK_COLOR: ${{ job.status }}
SLACK_MESSAGE: "Failed to auto-${{ env.CLOSE_TYPE }} PR on data repo: ${{ fromJson(env.GH_API_RES || env.TARGET_REPO_PR).message }}"

View File

@@ -1,119 +0,0 @@
name: generate_svg_common
on:
workflow_call:
inputs:
backend:
type: string
hardware_name:
type: string
layer:
type: string
bench_subset:
type: string
default: all
pbs_kind: # Valid values are 'classical', 'multi_bit' or 'any'
type: string
grouping_factor: # Valid values are 2, 3, or 4
type: string
default: 4
bench_type: # Valid values are 'latency', 'throughput'
type: string
name_suffix:
type: string
default: _mean_avx512
backend_comparison:
type: boolean
default: false
time_span_days:
type: string
default: 60
output_filename:
type: string
required: true
secrets:
DATA_EXTRACTOR_DATABASE_USER:
required: true
DATA_EXTRACTOR_DATABASE_HOST:
required: true
DATA_EXTRACTOR_DATABASE_PASSWORD:
required: true
permissions: {}
# zizmor: ignore[concurrency-limits] caller workflow is responsible for the concurrency
jobs:
generate-table:
name: generate_svg_common/generate-table
runs-on: ubuntu-latest
steps:
- name: Checkout tfhe-rs
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd
with:
persist-credentials: 'false'
- name: Produce table from database
if: inputs.backend_comparison == false
run: |
python3 -m pip install -r ci/data_extractor/requirements.txt
python3 ci/data_extractor/src/data_extractor.py "${OUTPUT_FILENAME}" \
--generate-svg \
--branch "${REF_NAME}" \
--backend "${BACKEND}" \
--hardware "${HARDWARE_NAME}" \
--tfhe-rs-layer "${LAYER}" \
--pbs-kind "${PBS_KIND}" \
--grouping-factor "${GROUPING_FACTOR}" \
--bench-type "${BENCH_TYPE}" \
--bench-subset "${BENCH_SUBSET}" \
--name-suffix "${NAME_SUFFIX}" \
--time-span-days "${TIME_SPAN}"
env:
OUTPUT_FILENAME: ${{ inputs.output_filename }}
REF_NAME: ${{ github.ref_name }}
BACKEND: ${{ inputs.backend }}
HARDWARE_NAME: ${{ inputs.hardware_name }}
LAYER: ${{ inputs.layer }}
PBS_KIND: ${{ inputs.pbs_kind }}
GROUPING_FACTOR: ${{ inputs.grouping_factor }}
BENCH_TYPE: ${{ inputs.bench_type }}
BENCH_SUBSET: ${{ inputs.bench_subset }}
NAME_SUFFIX: ${{ inputs.name_suffix }}
TIME_SPAN: ${{ inputs.time_span_days }}
DATA_EXTRACTOR_DATABASE_USER: ${{ secrets.DATA_EXTRACTOR_DATABASE_USER }}
DATA_EXTRACTOR_DATABASE_HOST: ${{ secrets.DATA_EXTRACTOR_DATABASE_HOST }}
DATA_EXTRACTOR_DATABASE_PASSWORD: ${{ secrets.DATA_EXTRACTOR_DATABASE_PASSWORD }}
- name: Upload tables
if: inputs.backend_comparison == false
uses: actions/upload-artifact@043fb46d1a93c77aae656e7c1c64a875d1fc6a0a
with:
name: ${{ github.sha }}_${{ inputs.backend }}_${{ inputs.layer }}_subset_${{inputs.bench_subset}}_${{ inputs.pbs_kind }}_${{ inputs.bench_type }}_tables
# This will upload all the file generated
path: ${{ inputs.output_filename }}*.svg
retention-days: 60
- name: Produce backends comparison table from database
if: inputs.backend_comparison == true
run: |
python3 -m pip install -r ci/data_extractor/requirements.txt
python3 ci/data_extractor/src/data_extractor.py "${OUTPUT_FILENAME}" \
--generate-svg \
--backends-comparison \
--time-span-days "${TIME_SPAN}"
env:
OUTPUT_FILENAME: ${{ inputs.output_filename }}
TIME_SPAN: ${{ inputs.time_span_days }}
DATA_EXTRACTOR_DATABASE_USER: ${{ secrets.DATA_EXTRACTOR_DATABASE_USER }}
DATA_EXTRACTOR_DATABASE_HOST: ${{ secrets.DATA_EXTRACTOR_DATABASE_HOST }}
DATA_EXTRACTOR_DATABASE_PASSWORD: ${{ secrets.DATA_EXTRACTOR_DATABASE_PASSWORD }}
- name: Upload comparison tables
if: inputs.backend_comparison == true
uses: actions/upload-artifact@043fb46d1a93c77aae656e7c1c64a875d1fc6a0a
with:
name: ${{ github.sha }}_backends_comparison_tables
# This will upload all the file generated
path: ${{ inputs.output_filename }}*.svg
retention-days: 60

View File

@@ -1,347 +0,0 @@
# Generate benchmark SVGs for public documentation
name: generate_documentation_svgs
on:
workflow_call:
inputs:
time_span_days:
type: string
required: true
generate-cpu-svgs:
type: boolean
default: true
generate-gpu-svgs:
type: boolean
default: true
generate-hpu-svgs:
type: boolean
default: true
secrets:
DATA_EXTRACTOR_DATABASE_USER:
required: true
DATA_EXTRACTOR_DATABASE_HOST:
required: true
DATA_EXTRACTOR_DATABASE_PASSWORD:
required: true
permissions: {}
# zizmor: ignore[concurrency-limits] caller workflow is responsible for the concurrency
jobs:
# -----------------------------------------------------------
# Integer benchmarks tables
# -----------------------------------------------------------
cpu-integer-latency-table:
name: generate_documentation_svgs/cpu-integer-latency-table
uses: ./.github/workflows/generate_svg_common.yml
if: inputs.generate-cpu-svgs
with:
backend: cpu
hardware_name: hpc7a.96xlarge
layer: integer
pbs_kind: classical
bench_type: latency
time_span_days: ${{ inputs.time_span_days }}
output_filename: cpu-integer-benchmark-tuniform-2m128-latency
secrets:
DATA_EXTRACTOR_DATABASE_USER: ${{ secrets.DATA_EXTRACTOR_DATABASE_USER }}
DATA_EXTRACTOR_DATABASE_HOST: ${{ secrets.DATA_EXTRACTOR_DATABASE_HOST }}
DATA_EXTRACTOR_DATABASE_PASSWORD: ${{ secrets.DATA_EXTRACTOR_DATABASE_PASSWORD }}
cpu-integer-throughput-table:
name: generate_documentation_svgs/cpu-integer-throughput-table
uses: ./.github/workflows/generate_svg_common.yml
if: inputs.generate-cpu-svgs
with:
backend: cpu
hardware_name: hpc7a.96xlarge
layer: integer
pbs_kind: classical
bench_type: throughput
time_span_days: ${{ inputs.time_span_days }}
output_filename: cpu-integer-benchmark-tuniform-2m128-throughput
secrets:
DATA_EXTRACTOR_DATABASE_USER: ${{ secrets.DATA_EXTRACTOR_DATABASE_USER }}
DATA_EXTRACTOR_DATABASE_HOST: ${{ secrets.DATA_EXTRACTOR_DATABASE_HOST }}
DATA_EXTRACTOR_DATABASE_PASSWORD: ${{ secrets.DATA_EXTRACTOR_DATABASE_PASSWORD }}
gpu-integer-latency-table:
name: generate_documentation_svgs/gpu-integer-latency-table
uses: ./.github/workflows/generate_svg_common.yml
if: inputs.generate-gpu-svgs
with:
backend: gpu
hardware_name: n3-H100-SXM5x8
layer: integer
pbs_kind: multi_bit
grouping_factor: 4
bench_type: latency
time_span_days: ${{ inputs.time_span_days }}
output_filename: gpu-integer-benchmark-h100x8-sxm5-multi-bit-tuniform-2m128-latency
secrets:
DATA_EXTRACTOR_DATABASE_USER: ${{ secrets.DATA_EXTRACTOR_DATABASE_USER }}
DATA_EXTRACTOR_DATABASE_HOST: ${{ secrets.DATA_EXTRACTOR_DATABASE_HOST }}
DATA_EXTRACTOR_DATABASE_PASSWORD: ${{ secrets.DATA_EXTRACTOR_DATABASE_PASSWORD }}
gpu-integer-throughput-table:
name: generate_documentation_svgs/gpu-integer-throughput-table
uses: ./.github/workflows/generate_svg_common.yml
if: inputs.generate-gpu-svgs
with:
backend: gpu
hardware_name: n3-H100-SXM5x8
layer: integer
pbs_kind: multi_bit
grouping_factor: 4
bench_type: throughput
time_span_days: ${{ inputs.time_span_days }}
output_filename: gpu-integer-benchmark-h100x8-sxm5-multi-bit-tuniform-2m128-throughput
secrets:
DATA_EXTRACTOR_DATABASE_USER: ${{ secrets.DATA_EXTRACTOR_DATABASE_USER }}
DATA_EXTRACTOR_DATABASE_HOST: ${{ secrets.DATA_EXTRACTOR_DATABASE_HOST }}
DATA_EXTRACTOR_DATABASE_PASSWORD: ${{ secrets.DATA_EXTRACTOR_DATABASE_PASSWORD }}
hpu-integer-latency-table:
name: generate_documentation_svgs/hpu-integer-latency-table
uses: ./.github/workflows/generate_svg_common.yml
if: inputs.generate-hpu-svgs
with:
backend: hpu
hardware_name: hpu_x1
layer: integer
pbs_kind: classical
bench_type: latency
time_span_days: ${{ inputs.time_span_days }}
output_filename: hpu-integer-benchmark-hpux1-tuniform-2m128-latency
secrets:
DATA_EXTRACTOR_DATABASE_USER: ${{ secrets.DATA_EXTRACTOR_DATABASE_USER }}
DATA_EXTRACTOR_DATABASE_HOST: ${{ secrets.DATA_EXTRACTOR_DATABASE_HOST }}
DATA_EXTRACTOR_DATABASE_PASSWORD: ${{ secrets.DATA_EXTRACTOR_DATABASE_PASSWORD }}
hpu-integer-throughput-table:
name: generate_documentation_svgs/hpu-integer-throughput-table
uses: ./.github/workflows/generate_svg_common.yml
if: inputs.generate-hpu-svgs
with:
backend: hpu
hardware_name: hpu_x1
layer: integer
pbs_kind: classical
bench_type: throughput
time_span_days: ${{ inputs.time_span_days }}
output_filename: hpu-integer-benchmark-hpux1-tuniform-2m128-throughput
secrets:
DATA_EXTRACTOR_DATABASE_USER: ${{ secrets.DATA_EXTRACTOR_DATABASE_USER }}
DATA_EXTRACTOR_DATABASE_HOST: ${{ secrets.DATA_EXTRACTOR_DATABASE_HOST }}
DATA_EXTRACTOR_DATABASE_PASSWORD: ${{ secrets.DATA_EXTRACTOR_DATABASE_PASSWORD }}
backend-comparison-latency-table:
name: generate_documentation_svgs/backend-comparison-latency-table
uses: ./.github/workflows/generate_svg_common.yml
if: inputs.generate-cpu-svgs && inputs.generate-gpu-svgs && inputs.generate-hpu-svgs
with:
backend_comparison: true
time_span_days: ${{ inputs.time_span_days }}
output_filename: cpu-gpu-hpu-integer-benchmark-fheuint64-tuniform-2m128-ciphertext
secrets:
DATA_EXTRACTOR_DATABASE_USER: ${{ secrets.DATA_EXTRACTOR_DATABASE_USER }}
DATA_EXTRACTOR_DATABASE_HOST: ${{ secrets.DATA_EXTRACTOR_DATABASE_HOST }}
DATA_EXTRACTOR_DATABASE_PASSWORD: ${{ secrets.DATA_EXTRACTOR_DATABASE_PASSWORD }}
# -----------------------------------------------------------
# ZK benchmarks tables
# -----------------------------------------------------------
cpu-zk-server-latency-table:
name: generate_documentation_svgs/cpu-zk-server-latency-table
uses: ./.github/workflows/generate_svg_common.yml
if: inputs.generate-cpu-svgs
with:
backend: cpu
hardware_name: hpc7a.96xlarge
layer: integer
bench_subset: zk
pbs_kind: classical
bench_type: latency
time_span_days: ${{ inputs.time_span_days }}
output_filename: cpu-zk-benchmark-latency
secrets:
DATA_EXTRACTOR_DATABASE_USER: ${{ secrets.DATA_EXTRACTOR_DATABASE_USER }}
DATA_EXTRACTOR_DATABASE_HOST: ${{ secrets.DATA_EXTRACTOR_DATABASE_HOST }}
DATA_EXTRACTOR_DATABASE_PASSWORD: ${{ secrets.DATA_EXTRACTOR_DATABASE_PASSWORD }}
cpu-zk-server-throughput-table:
name: generate_documentation_svgs/cpu-zk-server-throughput-table
uses: ./.github/workflows/generate_svg_common.yml
if: inputs.generate-cpu-svgs
with:
backend: cpu
hardware_name: hpc7a.96xlarge
layer: integer
bench_subset: zk
pbs_kind: classical
bench_type: throughput
time_span_days: ${{ inputs.time_span_days }}
output_filename: cpu-zk-benchmark-throughput
secrets:
DATA_EXTRACTOR_DATABASE_USER: ${{ secrets.DATA_EXTRACTOR_DATABASE_USER }}
DATA_EXTRACTOR_DATABASE_HOST: ${{ secrets.DATA_EXTRACTOR_DATABASE_HOST }}
DATA_EXTRACTOR_DATABASE_PASSWORD: ${{ secrets.DATA_EXTRACTOR_DATABASE_PASSWORD }}
cpu-zk-client-latency-table:
name: generate_documentation_svgs/cpu-zk-client-latency-table
uses: ./.github/workflows/generate_svg_common.yml
if: inputs.generate-cpu-svgs
with:
backend: cpu
hardware_name: m6i.4xlarge
layer: wasm
bench_subset: zk
pbs_kind: classical
bench_type: latency
name_suffix: _chrome_mean
time_span_days: ${{ inputs.time_span_days }}
output_filename: cpu-zk-wasm-benchmark-latency
secrets:
DATA_EXTRACTOR_DATABASE_USER: ${{ secrets.DATA_EXTRACTOR_DATABASE_USER }}
DATA_EXTRACTOR_DATABASE_HOST: ${{ secrets.DATA_EXTRACTOR_DATABASE_HOST }}
DATA_EXTRACTOR_DATABASE_PASSWORD: ${{ secrets.DATA_EXTRACTOR_DATABASE_PASSWORD }}
gpu-zk-server-latency-table:
name: generate_documentation_svgs/gpu-zk-server-latency-table
uses: ./.github/workflows/generate_svg_common.yml
if: inputs.generate-gpu-svgs
with:
backend: gpu
hardware_name: n3-H100-SXM5x8
layer: integer
bench_subset: zk
pbs_kind: multi_bit
grouping_factor: 4
bench_type: latency
time_span_days: ${{ inputs.time_span_days }}
output_filename: gpu-zk-benchmark-latency
secrets:
DATA_EXTRACTOR_DATABASE_USER: ${{ secrets.DATA_EXTRACTOR_DATABASE_USER }}
DATA_EXTRACTOR_DATABASE_HOST: ${{ secrets.DATA_EXTRACTOR_DATABASE_HOST }}
DATA_EXTRACTOR_DATABASE_PASSWORD: ${{ secrets.DATA_EXTRACTOR_DATABASE_PASSWORD }}
gpu-zk-server-throughput-table:
name: generate_documentation_svgs/gpu-zk-server-throughput-table
uses: ./.github/workflows/generate_svg_common.yml
if: inputs.generate-gpu-svgs
with:
backend: gpu
hardware_name: n3-H100-SXM5x8
layer: integer
bench_subset: zk
pbs_kind: multi_bit
grouping_factor: 4
bench_type: throughput
time_span_days: ${{ inputs.time_span_days }}
output_filename: gpu-zk-benchmark-throughput
secrets:
DATA_EXTRACTOR_DATABASE_USER: ${{ secrets.DATA_EXTRACTOR_DATABASE_USER }}
DATA_EXTRACTOR_DATABASE_HOST: ${{ secrets.DATA_EXTRACTOR_DATABASE_HOST }}
DATA_EXTRACTOR_DATABASE_PASSWORD: ${{ secrets.DATA_EXTRACTOR_DATABASE_PASSWORD }}
# -----------------------------------------------------------
# ERC7984 benchmarks tables
# -----------------------------------------------------------
cpu-erc7984-latency-throughput-table:
name: generate_documentation_svgs/cpu-erc7984-latency-throughput-table
uses: ./.github/workflows/generate_svg_common.yml
if: inputs.generate-cpu-svgs
with:
backend: cpu
hardware_name: hpc7a.96xlarge
layer: hlapi
bench_subset: erc7984
pbs_kind: classical
bench_type: both
time_span_days: ${{ inputs.time_span_days }}
output_filename: cpu-hlapi-erc7984-benchmark-latency-throughput
secrets:
DATA_EXTRACTOR_DATABASE_USER: ${{ secrets.DATA_EXTRACTOR_DATABASE_USER }}
DATA_EXTRACTOR_DATABASE_HOST: ${{ secrets.DATA_EXTRACTOR_DATABASE_HOST }}
DATA_EXTRACTOR_DATABASE_PASSWORD: ${{ secrets.DATA_EXTRACTOR_DATABASE_PASSWORD }}
gpu-erc7984-latency-throughput-table:
name: generate_documentation_svgs/gpu-erc7984-latency-throughput-table
uses: ./.github/workflows/generate_svg_common.yml
if: inputs.generate-gpu-svgs
with:
backend: gpu
hardware_name: n3-H100-SXM5x8
layer: hlapi
bench_subset: erc7984
pbs_kind: multi_bit
grouping_factor: 4
bench_type: both
time_span_days: ${{ inputs.time_span_days }}
output_filename: gpu-hlapi-erc7984-benchmark-h100x8-sxm5-latency-throughput
secrets:
DATA_EXTRACTOR_DATABASE_USER: ${{ secrets.DATA_EXTRACTOR_DATABASE_USER }}
DATA_EXTRACTOR_DATABASE_HOST: ${{ secrets.DATA_EXTRACTOR_DATABASE_HOST }}
DATA_EXTRACTOR_DATABASE_PASSWORD: ${{ secrets.DATA_EXTRACTOR_DATABASE_PASSWORD }}
hpu-erc7984-latency-throughput-table:
name: generate_documentation_svgs/hpu-erc7984-latency-throughput-table
uses: ./.github/workflows/generate_svg_common.yml
if: inputs.generate-hpu-svgs
with:
backend: hpu
hardware_name: hpu_x1
layer: hlapi
bench_subset: erc7984
pbs_kind: classical
bench_type: both
time_span_days: ${{ inputs.time_span_days }}
output_filename: hpu-hlapi-erc7984-benchmark-hpux1-latency-throughput.svg
secrets:
DATA_EXTRACTOR_DATABASE_USER: ${{ secrets.DATA_EXTRACTOR_DATABASE_USER }}
DATA_EXTRACTOR_DATABASE_HOST: ${{ secrets.DATA_EXTRACTOR_DATABASE_HOST }}
DATA_EXTRACTOR_DATABASE_PASSWORD: ${{ secrets.DATA_EXTRACTOR_DATABASE_PASSWORD }}
# -----------------------------------------------------------
# PBS benchmarks tables
# -----------------------------------------------------------
cpu-pbs-tables:
name: generate_documentation_svgs/cpu-pbs-tables
uses: ./.github/workflows/generate_svg_common.yml
if: inputs.generate-cpu-svgs
with:
backend: cpu
hardware_name: hpc7a.96xlarge
layer: core_crypto
pbs_kind: any
grouping_factor: 4
bench_type: latency
time_span_days: ${{ inputs.time_span_days }}
output_filename: cpu-pbs-benchmark
secrets:
DATA_EXTRACTOR_DATABASE_USER: ${{ secrets.DATA_EXTRACTOR_DATABASE_USER }}
DATA_EXTRACTOR_DATABASE_HOST: ${{ secrets.DATA_EXTRACTOR_DATABASE_HOST }}
DATA_EXTRACTOR_DATABASE_PASSWORD: ${{ secrets.DATA_EXTRACTOR_DATABASE_PASSWORD }}
gpu-pbs-tables:
name: generate_documentation_svgs/gpu-pbs-tables
uses: ./.github/workflows/generate_svg_common.yml
if: inputs.generate-gpu-svgs
with:
backend: gpu
hardware_name: n3-H100-SXM5x8
layer: core_crypto
pbs_kind: any
grouping_factor: 4
bench_type: latency
time_span_days: ${{ inputs.time_span_days }}
output_filename: gpu-pbs-benchmark
secrets:
DATA_EXTRACTOR_DATABASE_USER: ${{ secrets.DATA_EXTRACTOR_DATABASE_USER }}
DATA_EXTRACTOR_DATABASE_HOST: ${{ secrets.DATA_EXTRACTOR_DATABASE_HOST }}
DATA_EXTRACTOR_DATABASE_PASSWORD: ${{ secrets.DATA_EXTRACTOR_DATABASE_PASSWORD }}

View File

@@ -1,5 +1,5 @@
# Compile and test tfhe-cuda-backend on an RTX 4090 machine
name: gpu_4090_tests
name: TFHE Cuda Backend - 4090 full tests
env:
CARGO_TERM_COLOR: always
@@ -11,7 +11,6 @@ env:
SLACK_ICON: https://pbs.twimg.com/profile_images/1274014582265298945/OjBKP9kn_400x400.png
SLACK_USERNAME: ${{ secrets.BOT_USERNAME }}
SLACK_WEBHOOK: ${{ secrets.SLACK_WEBHOOK }}
CHECKOUT_TOKEN: ${{ secrets.REPO_CHECKOUT_TOKEN || secrets.GITHUB_TOKEN }}
on:
# Allows you to run this workflow manually from the Actions tab as an alternative.
@@ -19,35 +18,29 @@ on:
pull_request:
types: [ labeled ]
schedule:
# Every other day at 1AM
- cron: "0 1 */2 * *"
permissions:
contents: read
# zizmor: ignore[concurrency-limits] only Zama organization members and GitHub can trigger this workflow
# Nightly tests @ 1AM after each work day
- cron: "0 1 * * MON-FRI"
jobs:
cuda-tests-linux:
name: gpu_4090_tests/cuda-tests-linux
name: CUDA tests (RTX 4090)
if: github.event_name == 'workflow_dispatch' ||
contains(github.event.label.name, '4090_test') ||
(github.event_name == 'schedule' && github.repository == 'zama-ai/tfhe-rs')
concurrency:
group: ${{ github.workflow_ref }}
group: ${{ github.workflow }}_${{ github.ref }}
cancel-in-progress: true
runs-on: ["self-hosted", "4090-desktop"]
timeout-minutes: 2880 # 48 hours
steps:
- name: Checkout tfhe-rs
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd
uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683
with:
persist-credentials: 'false'
token: ${{ env.CHECKOUT_TOKEN }}
token: ${{ secrets.FHE_ACTIONS_TOKEN }}
- name: Install latest stable
uses: dtolnay/rust-toolchain@e97e2d8cc328f1b50210efc529dca0028893a2d9 # zizmor: ignore[stale-action-refs] this action doesn't create releases
uses: dtolnay/rust-toolchain@315e265cd78dad1e1dcf3a5074f6d6c47029d5aa
with:
toolchain: stable
@@ -82,9 +75,9 @@ jobs:
github_token: ${{ secrets.GITHUB_TOKEN }}
- name: Slack Notification
if: ${{ failure() || (cancelled() && github.event_name != 'pull_request') }}
if: ${{ failure() }}
continue-on-error: true
uses: rtCamp/action-slack-notify@e31e87e03dd19038e411e38ae27cbad084a90661
uses: rtCamp/action-slack-notify@c33737706dea87cd7784c687dadc9adf1be59990
env:
SLACK_COLOR: ${{ job.status }}
SLACK_MESSAGE: "CUDA RTX 4090 tests finished with status: ${{ job.status }}. (${{ env.ACTION_RUN_URL }})"

View File

@@ -1,165 +0,0 @@
# Compile and test tfhe-cuda-backend on an AWS instance
name: gpu_code_validation_tests
env:
CARGO_TERM_COLOR: always
ACTION_RUN_URL: ${{ github.server_url }}/${{ github.repository }}/actions/runs/${{ github.run_id }}
RUSTFLAGS: "-C target-cpu=native"
RUST_BACKTRACE: "full"
RUST_MIN_STACK: "8388608"
SLACK_CHANNEL: ${{ secrets.SLACK_CHANNEL }}
SLACK_ICON: https://pbs.twimg.com/profile_images/1274014582265298945/OjBKP9kn_400x400.png
SLACK_USERNAME: ${{ secrets.BOT_USERNAME }}
SLACK_WEBHOOK: ${{ secrets.SLACK_WEBHOOK }}
SLACKIFY_MARKDOWN: true
IS_PULL_REQUEST: ${{ github.event_name == 'pull_request' }}
PULL_REQUEST_MD_LINK: ""
CHECKOUT_TOKEN: ${{ secrets.REPO_CHECKOUT_TOKEN || secrets.GITHUB_TOKEN }}
# Secrets will be available only to zama-ai organization members
SECRETS_AVAILABLE: ${{ secrets.JOB_SECRET != '' }}
EXTERNAL_CONTRIBUTION_RUNNER: "gpu_ubuntu-22.04"
on:
# Allows you to run this workflow manually from the Actions tab as an alternative.
workflow_dispatch:
schedule:
# every friday noon
- cron: "0 12 * * 5"
permissions:
contents: read
# zizmor: ignore[concurrency-limits] concurrency is managed after instance setup to ensure safe provisioning
jobs:
setup-instance:
name: gpu_code_validation_tests/setup-instance
runs-on: ubuntu-latest
if: github.event_name == 'workflow_dispatch' ||
(github.event_name == 'schedule' && github.repository == 'zama-ai/tfhe-rs')
outputs:
runner-name: ${{ steps.start-remote-instance.outputs.label || steps.start-github-instance.outputs.runner_group }}
steps:
- name: Start remote instance
id: start-remote-instance
if: env.SECRETS_AVAILABLE == 'true'
uses: zama-ai/slab-github-runner@5aee5d157f4a0201e5eaefc9cc648e5f9f5472a5 # v1.6.0
with:
mode: start
github-token: ${{ secrets.SLAB_ACTION_TOKEN }}
slab-url: ${{ secrets.SLAB_BASE_URL }}
job-secret: ${{ secrets.JOB_SECRET }}
backend: hyperstack
profile: single-h100
# This instance will be spawned especially for pull-request from forked repository
- name: Start GitHub instance
id: start-github-instance
if: env.SECRETS_AVAILABLE == 'false'
run: |
echo "runner_group=${EXTERNAL_CONTRIBUTION_RUNNER}" >> "$GITHUB_OUTPUT"
cuda-tests-linux:
name: gpu_code_validation_tests/cuda-tests-linux
needs: [ setup-instance ]
if: github.event_name != 'pull_request' ||
(github.event_name == 'pull_request' && needs.setup-instance.result != 'skipped')
concurrency:
group: ${{ github.workflow_ref }}
cancel-in-progress: ${{ github.ref != 'refs/heads/main' }}
runs-on: ${{ needs.setup-instance.outputs.runner-name }}
timeout-minutes: 14400
strategy:
fail-fast: false
# explicit include-based build matrix, of known valid options
matrix:
include:
- os: ubuntu-22.04
cuda: "12.8"
gcc: 11
steps:
- name: Checkout tfhe-rs
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd
with:
persist-credentials: 'false'
token: ${{ env.CHECKOUT_TOKEN }}
- name: Setup Hyperstack dependencies
uses: ./.github/actions/gpu_setup
with:
cuda-version: ${{ matrix.cuda }}
gcc-version: ${{ matrix.gcc }}
github-instance: ${{ env.SECRETS_AVAILABLE == 'false' }}
- name: Find tools
run: |
# Disable unattended-upgrades to avoid lock issues
sudo systemctl mask --now unattended-upgrades
sudo systemctl stop --now unattended-upgrades
sudo apt-get clean
sudo rm -rf /var/lib/apt/lists/*
sudo apt purge -y unattended-upgrades
sudo apt update && sudo apt install -y valgrind
find /usr -executable -name "compute-sanitizer"
which valgrind
- name: Install latest stable
uses: dtolnay/rust-toolchain@e97e2d8cc328f1b50210efc529dca0028893a2d9 # zizmor: ignore[stale-action-refs] this action doesn't create releases
with:
toolchain: stable
- name: Run memory sanitizer
run: |
make test_high_level_api_gpu_valgrind
- name: Run CUDA backend racecheck tests
run: |
make test_cuda_backend_race_check
slack-notify:
name: gpu_code_validation_tests/slack-notify
needs: [ setup-instance, cuda-tests-linux ]
runs-on: ubuntu-latest
if: ${{ always() && needs.cuda-tests-linux.result != 'skipped' && failure() }}
continue-on-error: true
steps:
- name: Set pull-request URL
if: env.SECRETS_AVAILABLE == 'true' && github.event_name == 'pull_request'
run: |
echo "PULL_REQUEST_MD_LINK=[pull-request](${PR_BASE_URL}${PR_NUMBER}), " >> "${GITHUB_ENV}"
env:
PR_BASE_URL: ${{ vars.PR_BASE_URL }}
PR_NUMBER: ${{ github.event.pull_request.number }}
- name: Send message
if: env.SECRETS_AVAILABLE == 'true'
uses: rtCamp/action-slack-notify@e31e87e03dd19038e411e38ae27cbad084a90661
env:
SLACK_COLOR: ${{ needs.cuda-tests-linux.result }}
SLACK_MESSAGE: "GPU Memory Checks tests finished with status: ${{ needs.cuda-tests-linux.result }}. (${{ env.PULL_REQUEST_MD_LINK }}[action run](${{ env.ACTION_RUN_URL }}))"
teardown-instance:
name: gpu_code_validation_tests/teardown-instance
if: ${{ always() && needs.setup-instance.result == 'success' }}
needs: [ setup-instance, cuda-tests-linux ]
runs-on: ubuntu-latest
steps:
- name: Stop remote instance
id: stop-instance
if: env.SECRETS_AVAILABLE == 'true'
uses: zama-ai/slab-github-runner@5aee5d157f4a0201e5eaefc9cc648e5f9f5472a5 # v1.6.0
with:
mode: stop
github-token: ${{ secrets.SLAB_ACTION_TOKEN }}
slab-url: ${{ secrets.SLAB_BASE_URL }}
job-secret: ${{ secrets.JOB_SECRET }}
label: ${{ needs.setup-instance.outputs.runner-name }}
- name: Slack Notification
if: ${{ failure() }}
uses: rtCamp/action-slack-notify@e31e87e03dd19038e411e38ae27cbad084a90661
env:
SLACK_COLOR: ${{ job.status }}
SLACK_MESSAGE: "Instance teardown (cuda-tests) finished with status: ${{ job.status }}. (${{ env.ACTION_RUN_URL }})"

View File

@@ -1,5 +1,5 @@
# Compile and test tfhe-cuda-backend on an H100 VM on hyperstack
name: gpu_core_h100_tests
name: TFHE Cuda Backend - Fast tests on H100
env:
CARGO_TERM_COLOR: always
@@ -11,46 +11,33 @@ env:
SLACK_ICON: https://pbs.twimg.com/profile_images/1274014582265298945/OjBKP9kn_400x400.png
SLACK_USERNAME: ${{ secrets.BOT_USERNAME }}
SLACK_WEBHOOK: ${{ secrets.SLACK_WEBHOOK }}
SLACKIFY_MARKDOWN: true
IS_PULL_REQUEST: ${{ github.event_name == 'pull_request' }}
PULL_REQUEST_MD_LINK: ""
CHECKOUT_TOKEN: ${{ secrets.REPO_CHECKOUT_TOKEN || secrets.GITHUB_TOKEN }}
# Secrets will be available only to zama-ai organization members
SECRETS_AVAILABLE: ${{ secrets.JOB_SECRET != '' }}
EXTERNAL_CONTRIBUTION_RUNNER: "gpu_ubuntu-22.04"
on:
# Allows you to run this workflow manually from the Actions tab as an alternative.
workflow_dispatch:
pull_request:
types: [ labeled, opened, synchronize ]
permissions:
contents: read
# zizmor: ignore[concurrency-limits] concurrency is managed after instance setup to ensure safe provisioning
types: [ labeled ]
jobs:
should-run:
name: gpu_core_h100_tests/should-run
runs-on: ubuntu-latest
permissions:
pull-requests: read # Needed to check for file change
pull-requests: write
outputs:
gpu_test: ${{ env.IS_PULL_REQUEST == 'false' || steps.changed-files.outputs.gpu_any_changed }}
core_crypto_changed: ${{ steps.changed-files.outputs.core_crypto_any_changed }}
steps:
- name: Checkout tfhe-rs
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd
uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683
with:
fetch-depth: 0
persist-credentials: 'false'
token: ${{ env.CHECKOUT_TOKEN }}
token: ${{ secrets.FHE_ACTIONS_TOKEN }}
- name: Check for file changes
id: changed-files
uses: tj-actions/changed-files@9426d40962ed5378910ee2e21d5f8c6fcbf2dd96 # v47.0.6
uses: tj-actions/changed-files@bab30c2299617f6615ec02a68b9a40d10bd21366
with:
since_last_remote_commit: true
files_yaml: |
gpu:
- tfhe/Cargo.toml
@@ -62,25 +49,26 @@ jobs:
- tfhe/src/integer/server_key/radix_parallel/tests_signed/**
- tfhe/src/integer/server_key/radix_parallel/tests_cases_unsigned.rs
- tfhe/src/shortint/parameters/**
- tfhe/src/high_level_api/**
- tfhe/src/c_api/**
- '.github/workflows/gpu_core_h100_tests.yml'
core_crypto:
- tfhe/src/core_crypto/gpu/**
- 'tfhe/docs/**.md'
- '.github/workflows/gpu_fast_h100_tests.yml'
- scripts/integer-tests.sh
- ci/slab.toml
setup-instance:
name: gpu_core_h100_tests/setup-instance
name: Setup instance (cuda-h100-tests)
needs: should-run
if: github.event_name != 'pull_request' ||
(github.event.action == 'labeled' && github.event.label.name == 'approved' && needs.should-run.outputs.gpu_test == 'true') ||
(github.event.action != 'labeled' && needs.should-run.outputs.core_crypto_changed == 'true')
(github.event.action != 'labeled' && needs.should-run.outputs.gpu_test == 'true') ||
(github.event.action == 'labeled' && github.event.label.name == 'approved' && needs.should-run.outputs.gpu_test == 'true')
runs-on: ubuntu-latest
outputs:
runner-name: ${{ steps.start-remote-instance.outputs.label || steps.start-github-instance.outputs.runner_group }}
runner-name: ${{ steps.start-instance.outputs.label }}
steps:
- name: Start remote instance
id: start-remote-instance
if: env.SECRETS_AVAILABLE == 'true'
uses: zama-ai/slab-github-runner@5aee5d157f4a0201e5eaefc9cc648e5f9f5472a5 # v1.6.0
- name: Start instance
id: start-instance
uses: zama-ai/slab-github-runner@98f0788261a7323d5d695a883e20df36591a92b7
with:
mode: start
github-token: ${{ secrets.SLAB_ACTION_TOKEN }}
@@ -89,20 +77,13 @@ jobs:
backend: hyperstack
profile: single-h100
# This instance will be spawned especially for pull-request from forked repository
- name: Start GitHub instance
id: start-github-instance
if: env.SECRETS_AVAILABLE == 'false'
run: |
echo "runner_group=${EXTERNAL_CONTRIBUTION_RUNNER}" >> "$GITHUB_OUTPUT"
cuda-tests-linux:
name: gpu_core_h100_tests/cuda-tests-linux
name: CUDA H100 tests
needs: [ should-run, setup-instance ]
if: github.event_name != 'pull_request' ||
(github.event_name == 'pull_request' && needs.setup-instance.result != 'skipped')
concurrency:
group: ${{ github.workflow_ref }}
group: ${{ github.workflow }}_${{ github.ref }}
cancel-in-progress: ${{ github.ref != 'refs/heads/main' }}
runs-on: ${{ needs.setup-instance.outputs.runner-name }}
strategy:
@@ -111,67 +92,70 @@ jobs:
matrix:
include:
- os: ubuntu-22.04
cuda: "12.8"
cuda: "12.2"
gcc: 11
steps:
- name: Checkout tfhe-rs
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd
uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683
with:
persist-credentials: 'false'
token: ${{ env.CHECKOUT_TOKEN }}
token: ${{ secrets.FHE_ACTIONS_TOKEN }}
- name: Setup Hyperstack dependencies
uses: ./.github/actions/gpu_setup
uses: ./.github/actions/hyperstack_setup
with:
cuda-version: ${{ matrix.cuda }}
gcc-version: ${{ matrix.gcc }}
github-instance: ${{ env.SECRETS_AVAILABLE == 'false' }}
- name: Set up home
run: |
echo "HOME=/home/ubuntu" >> "${GITHUB_ENV}"
- name: Install latest stable
uses: dtolnay/rust-toolchain@e97e2d8cc328f1b50210efc529dca0028893a2d9 # zizmor: ignore[stale-action-refs] this action doesn't create releases
uses: dtolnay/rust-toolchain@315e265cd78dad1e1dcf3a5074f6d6c47029d5aa
with:
toolchain: stable
- name: Enable nvidia multi-process service
run: |
nvidia-cuda-mps-control -d
- name: Run core crypto and internal CUDA backend tests
run: |
BIG_TESTS_INSTANCE=TRUE make test_core_crypto_gpu
BIG_TESTS_INSTANCE=TRUE make test_integer_compression_gpu
BIG_TESTS_INSTANCE=TRUE make test_cuda_backend
- name: Run user docs tests
run: |
BIG_TESTS_INSTANCE=TRUE make test_user_doc_gpu
- name: Test C API
run: |
BIG_TESTS_INSTANCE=TRUE make test_c_api_gpu
- name: Run High Level API Tests
run: |
BIG_TESTS_INSTANCE=TRUE make test_high_level_api_gpu
slack-notify:
name: gpu_core_h100_tests/slack-notify
name: Slack Notification
needs: [ setup-instance, cuda-tests-linux ]
runs-on: ubuntu-latest
if: ${{ always() && needs.cuda-tests-linux.result != 'skipped' && failure() }}
continue-on-error: true
steps:
- name: Set pull-request URL
if: env.SECRETS_AVAILABLE == 'true' && github.event_name == 'pull_request'
run: |
echo "PULL_REQUEST_MD_LINK=[pull-request](${PR_BASE_URL}${PR_NUMBER}), " >> "${GITHUB_ENV}"
env:
PR_BASE_URL: ${{ vars.PR_BASE_URL }}
PR_NUMBER: ${{ github.event.pull_request.number }}
- name: Send message
if: env.SECRETS_AVAILABLE == 'true'
uses: rtCamp/action-slack-notify@e31e87e03dd19038e411e38ae27cbad084a90661
uses: rtCamp/action-slack-notify@c33737706dea87cd7784c687dadc9adf1be59990
env:
SLACK_COLOR: ${{ needs.cuda-tests-linux.result }}
SLACK_MESSAGE: "Core H100 tests finished with status: ${{ needs.cuda-tests-linux.result }}. (${{ env.PULL_REQUEST_MD_LINK }}[action run](${{ env.ACTION_RUN_URL }}))"
SLACK_MESSAGE: "Fast H100 tests finished with status: ${{ needs.cuda-tests-linux.result }}. (${{ env.ACTION_RUN_URL }})"
teardown-instance:
name: gpu_core_h100_tests/teardown-instance
if: ${{ always() && needs.setup-instance.result == 'success' }}
name: Teardown instance (cuda-h100-tests)
if: ${{ always() && needs.setup-instance.result != 'skipped' }}
needs: [ setup-instance, cuda-tests-linux ]
runs-on: ubuntu-latest
steps:
- name: Stop remote instance
- name: Stop instance
id: stop-instance
if: env.SECRETS_AVAILABLE == 'true'
uses: zama-ai/slab-github-runner@5aee5d157f4a0201e5eaefc9cc648e5f9f5472a5 # v1.6.0
uses: zama-ai/slab-github-runner@98f0788261a7323d5d695a883e20df36591a92b7
with:
mode: stop
github-token: ${{ secrets.SLAB_ACTION_TOKEN }}
@@ -181,7 +165,8 @@ jobs:
- name: Slack Notification
if: ${{ failure() }}
uses: rtCamp/action-slack-notify@e31e87e03dd19038e411e38ae27cbad084a90661
continue-on-error: true
uses: rtCamp/action-slack-notify@c33737706dea87cd7784c687dadc9adf1be59990
env:
SLACK_COLOR: ${{ job.status }}
SLACK_MESSAGE: "Instance teardown (cuda-h100-tests) finished with status: ${{ job.status }}. (${{ env.ACTION_RUN_URL }})"

View File

@@ -1,5 +1,5 @@
# Compile and test tfhe-cuda-backend on an AWS instance
name: gpu_fast_tests
name: TFHE Cuda Backend - Fast tests
env:
CARGO_TERM_COLOR: always
@@ -11,44 +11,32 @@ env:
SLACK_ICON: https://pbs.twimg.com/profile_images/1274014582265298945/OjBKP9kn_400x400.png
SLACK_USERNAME: ${{ secrets.BOT_USERNAME }}
SLACK_WEBHOOK: ${{ secrets.SLACK_WEBHOOK }}
SLACKIFY_MARKDOWN: true
IS_PULL_REQUEST: ${{ github.event_name == 'pull_request' }}
PULL_REQUEST_MD_LINK: ""
CHECKOUT_TOKEN: ${{ secrets.REPO_CHECKOUT_TOKEN || secrets.GITHUB_TOKEN }}
# Secrets will be available only to zama-ai organization members
SECRETS_AVAILABLE: ${{ secrets.JOB_SECRET != '' }}
EXTERNAL_CONTRIBUTION_RUNNER: "gpu_ubuntu-22.04"
on:
# Allows you to run this workflow manually from the Actions tab as an alternative.
workflow_dispatch:
pull_request:
permissions:
contents: read
# zizmor: ignore[concurrency-limits] concurrency is managed after instance setup to ensure safe provisioning
jobs:
should-run:
name: gpu_fast_tests/should-run
runs-on: ubuntu-latest
permissions:
pull-requests: read # Needed to check for file change
pull-requests: write
outputs:
gpu_test: ${{ env.IS_PULL_REQUEST == 'false' || steps.changed-files.outputs.gpu_any_changed }}
steps:
- name: Checkout tfhe-rs
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd
uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683
with:
fetch-depth: 0
persist-credentials: 'false'
token: ${{ env.CHECKOUT_TOKEN }}
token: ${{ secrets.FHE_ACTIONS_TOKEN }}
- name: Check for file changes
id: changed-files
uses: tj-actions/changed-files@9426d40962ed5378910ee2e21d5f8c6fcbf2dd96 # v47.0.6
uses: tj-actions/changed-files@bab30c2299617f6615ec02a68b9a40d10bd21366
with:
since_last_remote_commit: true
files_yaml: |
gpu:
- tfhe/Cargo.toml
@@ -62,22 +50,23 @@ jobs:
- tfhe/src/shortint/parameters/**
- tfhe/src/high_level_api/**
- tfhe/src/c_api/**
- 'tfhe/docs/**/**.md'
- 'tfhe/docs/**.md'
- '.github/workflows/gpu_fast_tests.yml'
- scripts/integer-tests.sh
- ci/slab.toml
setup-instance:
name: gpu_fast_tests/setup-instance
name: Setup instance (cuda-tests)
needs: should-run
if: github.event_name == 'workflow_dispatch' ||
if: github.event_name != 'pull_request' ||
needs.should-run.outputs.gpu_test == 'true'
runs-on: ubuntu-latest
outputs:
runner-name: ${{ steps.start-remote-instance.outputs.label || steps.start-github-instance.outputs.runner_group }}
runner-name: ${{ steps.start-instance.outputs.label }}
steps:
- name: Start remote instance
id: start-remote-instance
if: env.SECRETS_AVAILABLE == 'true'
uses: zama-ai/slab-github-runner@5aee5d157f4a0201e5eaefc9cc648e5f9f5472a5 # v1.6.0
- name: Start instance
id: start-instance
uses: zama-ai/slab-github-runner@98f0788261a7323d5d695a883e20df36591a92b7
with:
mode: start
github-token: ${{ secrets.SLAB_ACTION_TOKEN }}
@@ -86,20 +75,13 @@ jobs:
backend: hyperstack
profile: gpu-test
# This instance will be spawned especially for pull-request from forked repository
- name: Start GitHub instance
id: start-github-instance
if: env.SECRETS_AVAILABLE == 'false'
run: |
echo "runner_group=${EXTERNAL_CONTRIBUTION_RUNNER}" >> "$GITHUB_OUTPUT"
cuda-tests-linux:
name: gpu_fast_tests/cuda-tests-linux
name: CUDA tests
needs: [ should-run, setup-instance ]
if: github.event_name != 'pull_request' ||
(github.event_name == 'pull_request' && needs.setup-instance.result != 'skipped')
concurrency:
group: ${{ github.workflow_ref }}
group: ${{ github.workflow }}_${{ github.ref }}
cancel-in-progress: ${{ github.ref != 'refs/heads/main' }}
runs-on: ${{ needs.setup-instance.outputs.runner-name }}
strategy:
@@ -108,31 +90,30 @@ jobs:
matrix:
include:
- os: ubuntu-22.04
cuda: "12.8"
cuda: "12.2"
gcc: 11
steps:
- name: Checkout tfhe-rs
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd
uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683
with:
persist-credentials: 'false'
token: ${{ env.CHECKOUT_TOKEN }}
token: ${{ secrets.FHE_ACTIONS_TOKEN }}
- name: Setup Hyperstack dependencies
uses: ./.github/actions/gpu_setup
uses: ./.github/actions/hyperstack_setup
with:
cuda-version: ${{ matrix.cuda }}
gcc-version: ${{ matrix.gcc }}
github-instance: ${{ env.SECRETS_AVAILABLE == 'false' }}
- name: Set up home
run: |
echo "HOME=/home/ubuntu" >> "${GITHUB_ENV}"
- name: Install latest stable
uses: dtolnay/rust-toolchain@e97e2d8cc328f1b50210efc529dca0028893a2d9 # zizmor: ignore[stale-action-refs] this action doesn't create releases
uses: dtolnay/rust-toolchain@315e265cd78dad1e1dcf3a5074f6d6c47029d5aa
with:
toolchain: stable
- name: Enable nvidia multi-process service
run: |
nvidia-cuda-mps-control -d
- name: Run core crypto and internal CUDA backend tests
run: |
make test_core_crypto_gpu
@@ -149,40 +130,30 @@ jobs:
- name: Run High Level API Tests
run: |
make test_high_level_api_gpu_fast
make test_high_level_api_gpu
slack-notify:
name: gpu_fast_tests/slack-notify
name: Slack Notification
needs: [ setup-instance, cuda-tests-linux ]
runs-on: ubuntu-latest
if: ${{ always() && needs.cuda-tests-linux.result != 'skipped' && failure() }}
continue-on-error: true
steps:
- name: Set pull-request URL
if: env.SECRETS_AVAILABLE == 'true' && github.event_name == 'pull_request'
run: |
echo "PULL_REQUEST_MD_LINK=[pull-request](${PR_BASE_URL}${PR_NUMBER}), " >> "${GITHUB_ENV}"
env:
PR_BASE_URL: ${{ vars.PR_BASE_URL }}
PR_NUMBER: ${{ github.event.pull_request.number }}
- name: Send message
if: env.SECRETS_AVAILABLE == 'true'
uses: rtCamp/action-slack-notify@e31e87e03dd19038e411e38ae27cbad084a90661
uses: rtCamp/action-slack-notify@c33737706dea87cd7784c687dadc9adf1be59990
env:
SLACK_COLOR: ${{ needs.cuda-tests-linux.result }}
SLACK_MESSAGE: "Base GPU tests finished with status: ${{ needs.cuda-tests-linux.result }}. (${{ env.PULL_REQUEST_MD_LINK }}[action run](${{ env.ACTION_RUN_URL }}))"
SLACK_MESSAGE: "Base GPU tests finished with status: ${{ needs.cuda-tests-linux.result }}. (${{ env.ACTION_RUN_URL }})"
teardown-instance:
name: gpu_fast_tests/teardown-instance
if: ${{ always() && needs.setup-instance.result == 'success' }}
name: Teardown instance (cuda-tests)
if: ${{ always() && needs.setup-instance.result != 'skipped' }}
needs: [ setup-instance, cuda-tests-linux ]
runs-on: ubuntu-latest
steps:
- name: Stop remote instance
- name: Stop instance
id: stop-instance
if: env.SECRETS_AVAILABLE == 'true'
uses: zama-ai/slab-github-runner@5aee5d157f4a0201e5eaefc9cc648e5f9f5472a5 # v1.6.0
uses: zama-ai/slab-github-runner@98f0788261a7323d5d695a883e20df36591a92b7
with:
mode: stop
github-token: ${{ secrets.SLAB_ACTION_TOKEN }}
@@ -192,7 +163,8 @@ jobs:
- name: Slack Notification
if: ${{ failure() }}
uses: rtCamp/action-slack-notify@e31e87e03dd19038e411e38ae27cbad084a90661
continue-on-error: true
uses: rtCamp/action-slack-notify@c33737706dea87cd7784c687dadc9adf1be59990
env:
SLACK_COLOR: ${{ job.status }}
SLACK_MESSAGE: "Instance teardown (cuda-tests) finished with status: ${{ job.status }}. (${{ env.ACTION_RUN_URL }})"

View File

@@ -1,5 +1,5 @@
# Compile and test tfhe-cuda-backend on an H100 VM on hyperstack
name: gpu_full_h100_tests
name: TFHE Cuda Backend - Full tests on H100
env:
CARGO_TERM_COLOR: always
@@ -11,25 +11,21 @@ env:
SLACK_ICON: https://pbs.twimg.com/profile_images/1274014582265298945/OjBKP9kn_400x400.png
SLACK_USERNAME: ${{ secrets.BOT_USERNAME }}
SLACK_WEBHOOK: ${{ secrets.SLACK_WEBHOOK }}
IS_PULL_REQUEST: ${{ github.event_name == 'pull_request' }}
on:
workflow_dispatch:
permissions: {}
# zizmor: ignore[concurrency-limits] concurrency is managed after instance setup to ensure safe provisioning
jobs:
setup-instance:
name: gpu_full_h100_tests/setup-instance
name: Setup instance (cuda-h100-tests)
runs-on: ubuntu-latest
outputs:
runner-name: ${{ steps.start-instance.outputs.label }}
steps:
- name: Start instance
id: start-instance
uses: zama-ai/slab-github-runner@5aee5d157f4a0201e5eaefc9cc648e5f9f5472a5 # v1.6.0
uses: zama-ai/slab-github-runner@98f0788261a7323d5d695a883e20df36591a92b7
with:
mode: start
github-token: ${{ secrets.SLAB_ACTION_TOKEN }}
@@ -39,10 +35,10 @@ jobs:
profile: single-h100
cuda-tests-linux:
name: gpu_full_h100_tests/cuda-tests-linux
name: CUDA H100 tests
needs: [ setup-instance ]
concurrency:
group: ${{ github.workflow_ref }}
group: ${{ github.workflow }}_${{ github.ref }}
cancel-in-progress: ${{ github.ref != 'refs/heads/main' }}
runs-on: ${{ needs.setup-instance.outputs.runner-name }}
strategy:
@@ -51,28 +47,42 @@ jobs:
matrix:
include:
- os: ubuntu-22.04
cuda: "12.8"
cuda: "12.2"
gcc: 11
steps:
# Mandatory on hyperstack since a bootable volume is not re-usable yet.
- name: Install dependencies
run: |
sudo apt update
sudo apt install -y checkinstall zlib1g-dev libssl-dev libclang-dev
wget https://github.com/Kitware/CMake/releases/download/v${{ env.CMAKE_VERSION }}/cmake-${{ env.CMAKE_VERSION }}.tar.gz
tar -zxvf cmake-${{ env.CMAKE_VERSION }}.tar.gz
cd cmake-${{ env.CMAKE_VERSION }}
./bootstrap
make -j"$(nproc)"
sudo make install
- name: Checkout tfhe-rs
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd
uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683
with:
persist-credentials: 'false'
token: ${{ secrets.REPO_CHECKOUT_TOKEN }}
token: ${{ secrets.FHE_ACTIONS_TOKEN }}
- name: Setup Hyperstack dependencies
uses: ./.github/actions/gpu_setup
uses: ./.github/actions/hyperstack_setup
with:
cuda-version: ${{ matrix.cuda }}
gcc-version: ${{ matrix.gcc }}
- name: Set up home
run: |
echo "HOME=/home/ubuntu" >> "${GITHUB_ENV}"
- name: Install latest stable
uses: dtolnay/rust-toolchain@e97e2d8cc328f1b50210efc529dca0028893a2d9 # zizmor: ignore[stale-action-refs] this action doesn't create releases
uses: dtolnay/rust-toolchain@315e265cd78dad1e1dcf3a5074f6d6c47029d5aa
with:
toolchain: stable
- name: Enable nvidia multi-process service
run: |
nvidia-cuda-mps-control -d
- name: Run core crypto, integer and internal CUDA backend tests
run: |
make test_gpu
@@ -90,27 +100,26 @@ jobs:
make test_high_level_api_gpu
slack-notify:
name: gpu_full_h100_tests/slack-notify
name: Slack Notification
needs: [ setup-instance, cuda-tests-linux ]
runs-on: ubuntu-latest
if: ${{ failure() }}
continue-on-error: true
steps:
- name: Send message
uses: rtCamp/action-slack-notify@e31e87e03dd19038e411e38ae27cbad084a90661
uses: rtCamp/action-slack-notify@c33737706dea87cd7784c687dadc9adf1be59990
env:
SLACK_COLOR: ${{ needs.cuda-tests-linux.result }}
SLACK_MESSAGE: "Full H100 tests finished with status: ${{ needs.cuda-tests-linux.result }}. (${{ env.ACTION_RUN_URL }})"
teardown-instance:
name: gpu_full_h100_tests/teardown-instance
if: ${{ always() && needs.setup-instance.result == 'success' }}
name: Teardown instance (cuda-h100-tests)
needs: [ setup-instance, cuda-tests-linux ]
runs-on: ubuntu-latest
steps:
- name: Stop instance
id: stop-instance
uses: zama-ai/slab-github-runner@5aee5d157f4a0201e5eaefc9cc648e5f9f5472a5 # v1.6.0
uses: zama-ai/slab-github-runner@98f0788261a7323d5d695a883e20df36591a92b7
with:
mode: stop
github-token: ${{ secrets.SLAB_ACTION_TOKEN }}
@@ -120,7 +129,8 @@ jobs:
- name: Slack Notification
if: ${{ failure() }}
uses: rtCamp/action-slack-notify@e31e87e03dd19038e411e38ae27cbad084a90661
continue-on-error: true
uses: rtCamp/action-slack-notify@c33737706dea87cd7784c687dadc9adf1be59990
env:
SLACK_COLOR: ${{ job.status }}
SLACK_MESSAGE: "Instance teardown (cuda-h100-tests) finished with status: ${{ job.status }}. (${{ env.ACTION_RUN_URL }})"

View File

@@ -1,5 +1,5 @@
# Compile and test tfhe-cuda-backend on an AWS instance
name: gpu_full_multi_gpu_tests
name: TFHE Cuda Backend - Full tests multi-GPU
env:
CARGO_TERM_COLOR: always
@@ -11,13 +11,7 @@ env:
SLACK_ICON: https://pbs.twimg.com/profile_images/1274014582265298945/OjBKP9kn_400x400.png
SLACK_USERNAME: ${{ secrets.BOT_USERNAME }}
SLACK_WEBHOOK: ${{ secrets.SLACK_WEBHOOK }}
SLACKIFY_MARKDOWN: true
IS_PULL_REQUEST: ${{ github.event_name == 'pull_request' }}
PULL_REQUEST_MD_LINK: ""
CHECKOUT_TOKEN: ${{ secrets.REPO_CHECKOUT_TOKEN || secrets.GITHUB_TOKEN }}
# Secrets will be available only to zama-ai organization members
SECRETS_AVAILABLE: ${{ secrets.JOB_SECRET != '' }}
EXTERNAL_CONTRIBUTION_RUNNER: "gpu_ubuntu-22.04"
on:
# Allows you to run this workflow manually from the Actions tab as an alternative.
@@ -25,31 +19,25 @@ on:
pull_request:
types: [ labeled ]
permissions:
contents: read
# zizmor: ignore[concurrency-limits] concurrency is managed after instance setup to ensure safe provisioning
jobs:
should-run:
name: gpu_full_multi_gpu_tests/should-run
runs-on: ubuntu-latest
permissions:
pull-requests: read # Needed to check for file change
pull-requests: write
outputs:
gpu_test: ${{ env.IS_PULL_REQUEST == 'false' || steps.changed-files.outputs.gpu_any_changed }}
steps:
- name: Checkout tfhe-rs
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd
uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683
with:
fetch-depth: 0
persist-credentials: 'false'
token: ${{ env.CHECKOUT_TOKEN }}
token: ${{ secrets.FHE_ACTIONS_TOKEN }}
- name: Check for file changes
id: changed-files
uses: tj-actions/changed-files@9426d40962ed5378910ee2e21d5f8c6fcbf2dd96 # v47.0.6
uses: tj-actions/changed-files@bab30c2299617f6615ec02a68b9a40d10bd21366
with:
since_last_remote_commit: true
files_yaml: |
gpu:
- tfhe/Cargo.toml
@@ -63,46 +51,39 @@ jobs:
- tfhe/src/shortint/parameters/**
- tfhe/src/high_level_api/**
- tfhe/src/c_api/**
- 'tfhe/docs/**/**.md'
- 'tfhe/docs/**.md'
- '.github/workflows/**_multi_gpu_tests.yml'
- scripts/integer-tests.sh
- ci/slab.toml
setup-instance:
name: gpu_full_multi_gpu_tests/setup-instance
name: Setup instance (cuda-tests-multi-gpu)
needs: should-run
if: github.event_name != 'pull_request' ||
(github.event.action != 'labeled' && needs.should-run.outputs.gpu_test == 'true') ||
(github.event.action == 'labeled' && github.event.label.name == 'approved' && needs.should-run.outputs.gpu_test == 'true')
runs-on: ubuntu-latest
outputs:
runner-name: ${{ steps.start-remote-instance.outputs.label || steps.start-github-instance.outputs.runner_group }}
runner-name: ${{ steps.start-instance.outputs.label }}
steps:
- name: Start remote instance
id: start-remote-instance
if: env.SECRETS_AVAILABLE == 'true'
uses: zama-ai/slab-github-runner@5aee5d157f4a0201e5eaefc9cc648e5f9f5472a5 # v1.6.0
- name: Start instance
id: start-instance
uses: zama-ai/slab-github-runner@98f0788261a7323d5d695a883e20df36591a92b7
with:
mode: start
github-token: ${{ secrets.SLAB_ACTION_TOKEN }}
slab-url: ${{ secrets.SLAB_BASE_URL }}
job-secret: ${{ secrets.JOB_SECRET }}
backend: hyperstack
profile: 4-l40
# This instance will be spawned especially for pull-request from forked repository
- name: Start GitHub instance
id: start-github-instance
if: env.SECRETS_AVAILABLE == 'false'
run: |
echo "runner_group=${EXTERNAL_CONTRIBUTION_RUNNER}" >> "$GITHUB_OUTPUT"
profile: multi-gpu-test
cuda-tests-linux:
name: gpu_full_multi_gpu_tests/cuda-tests-linux
name: CUDA multi-GPU tests
needs: [ should-run, setup-instance ]
if: github.event_name != 'pull_request' ||
(github.event_name == 'pull_request' && needs.setup-instance.result != 'skipped')
concurrency:
group: ${{ github.workflow_ref }}
group: ${{ github.workflow }}_${{ github.ref }}
cancel-in-progress: ${{ github.ref != 'refs/heads/main' }}
runs-on: ${{ needs.setup-instance.outputs.runner-name }}
strategy:
@@ -111,29 +92,30 @@ jobs:
matrix:
include:
- os: ubuntu-22.04
cuda: "12.8"
cuda: "12.2"
gcc: 11
steps:
- name: Checkout tfhe-rs
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd
uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683
with:
persist-credentials: 'false'
token: ${{ env.CHECKOUT_TOKEN }}
token: ${{ secrets.FHE_ACTIONS_TOKEN }}
- name: Setup Hyperstack dependencies
uses: ./.github/actions/gpu_setup
uses: ./.github/actions/hyperstack_setup
with:
cuda-version: ${{ matrix.cuda }}
gcc-version: ${{ matrix.gcc }}
github-instance: ${{ env.SECRETS_AVAILABLE == 'false' }}
- name: Set up home
run: |
echo "HOME=/home/ubuntu" >> "${GITHUB_ENV}"
- name: Install latest stable
uses: dtolnay/rust-toolchain@e97e2d8cc328f1b50210efc529dca0028893a2d9 # zizmor: ignore[stale-action-refs] this action doesn't create releases
uses: dtolnay/rust-toolchain@315e265cd78dad1e1dcf3a5074f6d6c47029d5aa
with:
toolchain: stable
- name: Enable nvidia multi-process service
run: |
nvidia-cuda-mps-control -d
- name: Run multi-bit CUDA integer compression tests
run: |
BIG_TESTS_INSTANCE=TRUE make test_integer_compression_gpu
@@ -141,7 +123,7 @@ jobs:
# No need to test core_crypto and classic PBS in integer since it's already tested on single GPU.
- name: Run multi-bit CUDA integer tests
run: |
BIG_TESTS_INSTANCE=TRUE NO_BIG_PARAMS_GPU=TRUE make test_integer_multi_bit_gpu_ci
BIG_TESTS_INSTANCE=TRUE make test_integer_multi_bit_gpu_ci
- name: Run user docs tests
run: |
@@ -153,40 +135,30 @@ jobs:
- name: Run High Level API Tests
run: |
make test_high_level_api_gpu_fast
BIG_TESTS_INSTANCE=TRUE make test_high_level_api_gpu
slack-notify:
name: gpu_full_multi_gpu_tests/slack-notify
name: Slack Notification
needs: [ setup-instance, cuda-tests-linux ]
runs-on: ubuntu-latest
if: ${{ always() && needs.cuda-tests-linux.result != 'skipped' && failure() }}
continue-on-error: true
steps:
- name: Set pull-request URL
if: env.SECRETS_AVAILABLE == 'true' && github.event_name == 'pull_request'
run: |
echo "PULL_REQUEST_MD_LINK=[pull-request](${PR_BASE_URL}${PR_NUMBER}), " >> "${GITHUB_ENV}"
env:
PR_BASE_URL: ${{ vars.PR_BASE_URL }}
PR_NUMBER: ${{ github.event.pull_request.number }}
- name: Send message
if: env.SECRETS_AVAILABLE == 'true'
uses: rtCamp/action-slack-notify@e31e87e03dd19038e411e38ae27cbad084a90661
uses: rtCamp/action-slack-notify@c33737706dea87cd7784c687dadc9adf1be59990
env:
SLACK_COLOR: ${{ needs.cuda-tests-linux.result }}
SLACK_MESSAGE: "Multi-GPU tests finished with status: ${{ needs.cuda-tests-linux.result }}. (${{ env.PULL_REQUEST_MD_LINK }}[action run](${{ env.ACTION_RUN_URL }}))"
SLACK_MESSAGE: "Multi-GPU tests finished with status: ${{ needs.cuda-tests-linux.result }}. (${{ env.ACTION_RUN_URL }})"
teardown-instance:
name: gpu_full_multi_gpu_tests/teardown-instance
if: ${{ always() && needs.setup-instance.result == 'success' }}
name: Teardown instance (cuda-tests-multi-gpu)
if: ${{ always() && needs.setup-instance.result != 'skipped' }}
needs: [ setup-instance, cuda-tests-linux ]
runs-on: ubuntu-latest
steps:
- name: Stop remote instance
- name: Stop instance
id: stop-instance
if: env.SECRETS_AVAILABLE == 'true'
uses: zama-ai/slab-github-runner@5aee5d157f4a0201e5eaefc9cc648e5f9f5472a5 # v1.6.0
uses: zama-ai/slab-github-runner@98f0788261a7323d5d695a883e20df36591a92b7
with:
mode: stop
github-token: ${{ secrets.SLAB_ACTION_TOKEN }}
@@ -196,7 +168,8 @@ jobs:
- name: Slack Notification
if: ${{ failure() }}
uses: rtCamp/action-slack-notify@e31e87e03dd19038e411e38ae27cbad084a90661
continue-on-error: true
uses: rtCamp/action-slack-notify@c33737706dea87cd7784c687dadc9adf1be59990
env:
SLACK_COLOR: ${{ job.status }}
SLACK_MESSAGE: "Instance teardown (cuda-tests-multi-gpu) finished with status: ${{ job.status }}. (${{ env.ACTION_RUN_URL }})"

View File

@@ -1,196 +0,0 @@
# Compile and test tfhe-cuda-backend on an H100 VM on hyperstack
name: gpu_hlapi_h100_tests
env:
CARGO_TERM_COLOR: always
ACTION_RUN_URL: ${{ github.server_url }}/${{ github.repository }}/actions/runs/${{ github.run_id }}
RUSTFLAGS: "-C target-cpu=native"
RUST_BACKTRACE: "full"
RUST_MIN_STACK: "8388608"
SLACK_CHANNEL: ${{ secrets.SLACK_CHANNEL }}
SLACK_ICON: https://pbs.twimg.com/profile_images/1274014582265298945/OjBKP9kn_400x400.png
SLACK_USERNAME: ${{ secrets.BOT_USERNAME }}
SLACK_WEBHOOK: ${{ secrets.SLACK_WEBHOOK }}
SLACKIFY_MARKDOWN: true
IS_PULL_REQUEST: ${{ github.event_name == 'pull_request' }}
PULL_REQUEST_MD_LINK: ""
CHECKOUT_TOKEN: ${{ secrets.REPO_CHECKOUT_TOKEN || secrets.GITHUB_TOKEN }}
# Secrets will be available only to zama-ai organization members
SECRETS_AVAILABLE: ${{ secrets.JOB_SECRET != '' }}
EXTERNAL_CONTRIBUTION_RUNNER: "gpu_ubuntu-22.04"
on:
# Allows you to run this workflow manually from the Actions tab as an alternative.
workflow_dispatch:
pull_request:
types: [ labeled, opened, synchronize ]
permissions:
contents: read
# zizmor: ignore[concurrency-limits] concurrency is managed after instance setup to ensure safe provisioning
jobs:
should-run:
name: gpu_hlapi_h100_tests/should-run
runs-on: ubuntu-latest
permissions:
pull-requests: read # Needed to check for file change
outputs:
gpu_test: ${{ env.IS_PULL_REQUEST == 'false' || steps.changed-files.outputs.gpu_any_changed }}
core_crypto_changed: ${{ steps.changed-files.outputs.core_crypto_any_changed }}
steps:
- name: Checkout tfhe-rs
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd
with:
fetch-depth: 0
persist-credentials: 'false'
token: ${{ env.CHECKOUT_TOKEN }}
- name: Check for file changes
id: changed-files
uses: tj-actions/changed-files@9426d40962ed5378910ee2e21d5f8c6fcbf2dd96 # v47.0.6
with:
files_yaml: |
gpu:
- tfhe/Cargo.toml
- tfhe/build.rs
- backends/tfhe-cuda-backend/**
- tfhe/src/core_crypto/gpu/**
- tfhe/src/integer/gpu/**
- tfhe/src/integer/server_key/radix_parallel/tests_unsigned/**
- tfhe/src/integer/server_key/radix_parallel/tests_signed/**
- tfhe/src/integer/server_key/radix_parallel/tests_cases_unsigned.rs
- tfhe/src/shortint/parameters/**
- tfhe/src/high_level_api/**
- tfhe/src/c_api/**
- 'tfhe/docs/**/**.md'
- '.github/workflows/gpu_hlapi_h100_tests.yml'
core_crypto:
- tfhe/src/core_crypto/gpu/**
setup-instance:
name: gpu_hlapi_h100_tests/setup-instance
needs: should-run
if: github.event_name != 'pull_request' ||
(github.event.action == 'labeled' && github.event.label.name == 'approved' && needs.should-run.outputs.gpu_test == 'true') ||
(github.event.action != 'labeled' && needs.should-run.outputs.core_crypto_changed == 'true')
runs-on: ubuntu-latest
outputs:
runner-name: ${{ steps.start-remote-instance.outputs.label || steps.start-github-instance.outputs.runner_group }}
steps:
- name: Start remote instance
id: start-remote-instance
if: env.SECRETS_AVAILABLE == 'true'
uses: zama-ai/slab-github-runner@5aee5d157f4a0201e5eaefc9cc648e5f9f5472a5 # v1.6.0
with:
mode: start
github-token: ${{ secrets.SLAB_ACTION_TOKEN }}
slab-url: ${{ secrets.SLAB_BASE_URL }}
job-secret: ${{ secrets.JOB_SECRET }}
backend: hyperstack
profile: single-h100
# This instance will be spawned especially for pull-request from forked repository
- name: Start GitHub instance
id: start-github-instance
if: env.SECRETS_AVAILABLE == 'false'
run: |
echo "runner_group=${EXTERNAL_CONTRIBUTION_RUNNER}" >> "$GITHUB_OUTPUT"
cuda-tests-linux:
name: gpu_hlapi_h100_tests/cuda-tests-linux
needs: [ should-run, setup-instance ]
if: github.event_name != 'pull_request' ||
(github.event_name == 'pull_request' && needs.setup-instance.result != 'skipped')
concurrency:
group: ${{ github.workflow_ref }}
cancel-in-progress: ${{ github.ref != 'refs/heads/main' }}
runs-on: ${{ needs.setup-instance.outputs.runner-name }}
strategy:
fail-fast: false
# explicit include-based build matrix, of known valid options
matrix:
include:
- os: ubuntu-22.04
cuda: "12.8"
gcc: 11
steps:
- name: Checkout tfhe-rs
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd
with:
persist-credentials: 'false'
token: ${{ env.CHECKOUT_TOKEN }}
- name: Setup Hyperstack dependencies
uses: ./.github/actions/gpu_setup
with:
cuda-version: ${{ matrix.cuda }}
gcc-version: ${{ matrix.gcc }}
github-instance: ${{ env.SECRETS_AVAILABLE == 'false' }}
- name: Install latest stable
uses: dtolnay/rust-toolchain@e97e2d8cc328f1b50210efc529dca0028893a2d9 # zizmor: ignore[stale-action-refs] this action doesn't create releases
with:
toolchain: stable
- name: Enable nvidia multi-process service
run: |
nvidia-cuda-mps-control -d
- name: Run user docs tests
run: |
BIG_TESTS_INSTANCE=TRUE make test_user_doc_gpu
- name: Test C API
run: |
BIG_TESTS_INSTANCE=TRUE make test_c_api_gpu
- name: Run High Level API Tests
run: |
BIG_TESTS_INSTANCE=TRUE make test_high_level_api_gpu
slack-notify:
name: gpu_hlapi_h100_tests/slack-notify
needs: [ setup-instance, cuda-tests-linux ]
runs-on: ubuntu-latest
if: ${{ always() && needs.cuda-tests-linux.result != 'skipped' && failure() }}
continue-on-error: true
steps:
- name: Set pull-request URL
if: env.SECRETS_AVAILABLE == 'true' && github.event_name == 'pull_request'
run: |
echo "PULL_REQUEST_MD_LINK=[pull-request](${PR_BASE_URL}${PR_NUMBER}), " >> "${GITHUB_ENV}"
env:
PR_BASE_URL: ${{ vars.PR_BASE_URL }}
PR_NUMBER: ${{ github.event.pull_request.number }}
- name: Send message
if: env.SECRETS_AVAILABLE == 'true'
uses: rtCamp/action-slack-notify@e31e87e03dd19038e411e38ae27cbad084a90661
env:
SLACK_COLOR: ${{ needs.cuda-tests-linux.result }}
SLACK_MESSAGE: "HL API H100 tests finished with status: ${{ needs.cuda-tests-linux.result }}. (${{ env.PULL_REQUEST_MD_LINK }}[action run](${{ env.ACTION_RUN_URL }}))"
teardown-instance:
name: gpu_hlapi_h100_tests/teardown-instance
if: ${{ always() && needs.setup-instance.result == 'success' }}
needs: [ setup-instance, cuda-tests-linux ]
runs-on: ubuntu-latest
steps:
- name: Stop remote instance
id: stop-instance
if: env.SECRETS_AVAILABLE == 'true'
uses: zama-ai/slab-github-runner@5aee5d157f4a0201e5eaefc9cc648e5f9f5472a5 # v1.6.0
with:
mode: stop
github-token: ${{ secrets.SLAB_ACTION_TOKEN }}
slab-url: ${{ secrets.SLAB_BASE_URL }}
job-secret: ${{ secrets.JOB_SECRET }}
label: ${{ needs.setup-instance.outputs.runner-name }}
- name: Slack Notification
if: ${{ failure() }}
uses: rtCamp/action-slack-notify@e31e87e03dd19038e411e38ae27cbad084a90661
env:
SLACK_COLOR: ${{ job.status }}
SLACK_MESSAGE: "Instance teardown (cuda-h100-tests) finished with status: ${{ job.status }}. (${{ env.ACTION_RUN_URL }})"

View File

@@ -1,4 +1,4 @@
name: gpu_integer_long_run_tests
name: AWS Long Run Tests on GPU
env:
CARGO_TERM_COLOR: always
@@ -10,79 +10,39 @@ env:
SLACK_ICON: https://pbs.twimg.com/profile_images/1274014582265298945/OjBKP9kn_400x400.png
SLACK_USERNAME: ${{ secrets.BOT_USERNAME }}
SLACK_WEBHOOK: ${{ secrets.SLACK_WEBHOOK }}
CHECKOUT_TOKEN: ${{ secrets.REPO_CHECKOUT_TOKEN || secrets.GITHUB_TOKEN }}
IS_PR: ${{ github.event_name == 'pull_request' }}
on:
# Allows you to run this workflow manually from the Actions tab as an alternative.
workflow_dispatch:
schedule:
# Weekly tests will be triggered every Monday at 8p.m.
- cron: "0 20 * * 1"
pull_request:
permissions:
contents: read
# zizmor: ignore[concurrency-limits] concurrency is managed after instance setup to ensure safe provisioning
# Weekly tests will be triggered each Friday at 1a.m.
- cron: '0 1 * * FRI'
jobs:
should-run:
name: gpu_integer_long_run_tests/should-run
runs-on: ubuntu-latest
permissions:
pull-requests: read # Needed to check for file change
outputs:
is_needed_in_gpu_ci: ${{ env.IS_PR == 'false' || steps.changed-files.outputs.gpu_any_changed }}
steps:
- name: Checkout tfhe-rs
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd
with:
fetch-depth: 0
persist-credentials: 'false'
token: ${{ env.CHECKOUT_TOKEN }}
- name: Check for file changes
id: changed-files
uses: tj-actions/changed-files@9426d40962ed5378910ee2e21d5f8c6fcbf2dd96 # v47.0.6
with:
files_yaml: |
gpu:
- tfhe/Cargo.toml
- tfhe/build.rs
- backends/tfhe-cuda-backend/**
- tfhe/src/core_crypto/gpu/**
- tfhe/src/integer/gpu/**
- tfhe/src/shortint/parameters/**
- '.github/workflows/gpu_integer_long_run_tests.yml'
setup-instance:
name: gpu_integer_long_run_tests/setup-instance
needs: [should-run]
if: github.event_name == 'workflow_dispatch' ||
(github.event_name == 'schedule' && github.repository == 'zama-ai/tfhe-rs') ||
needs.should-run.outputs.is_needed_in_gpu_ci == 'true'
name: Setup instance (gpu-tests)
if: github.event_name != 'schedule' ||
(github.event_name == 'schedule' && github.repository == 'zama-ai/tfhe-rs')
runs-on: ubuntu-latest
outputs:
runner-name: ${{ steps.start-instance.outputs.label }}
steps:
- name: Start instance
id: start-instance
uses: zama-ai/slab-github-runner@5aee5d157f4a0201e5eaefc9cc648e5f9f5472a5 # v1.6.0
uses: zama-ai/slab-github-runner@98f0788261a7323d5d695a883e20df36591a92b7
with:
mode: start
github-token: ${{ secrets.SLAB_ACTION_TOKEN }}
slab-url: ${{ secrets.SLAB_BASE_URL }}
job-secret: ${{ secrets.JOB_SECRET }}
backend: hyperstack
profile: 4-l40
profile: 2-h100
cuda-tests:
name: gpu_integer_long_run_tests/cuda-tests
name: Long run GPU H100 tests
needs: [ setup-instance ]
concurrency:
group: ${{ github.workflow_ref }}_${{github.event_name}}
group: ${{ github.workflow }}_${{github.event_name}}_${{ github.ref }}
cancel-in-progress: true
runs-on: ${{ needs.setup-instance.outputs.runner-name }}
strategy:
@@ -91,59 +51,53 @@ jobs:
matrix:
include:
- os: ubuntu-22.04
cuda: "12.8"
cuda: "12.2"
gcc: 11
timeout-minutes: 4320 # 72 hours
steps:
- name: Checkout tfhe-rs
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd
with:
persist-credentials: 'false'
token: ${{ env.CHECKOUT_TOKEN }}
uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683
- name: Setup Hyperstack dependencies
uses: ./.github/actions/gpu_setup
uses: ./.github/actions/hyperstack_setup
with:
cuda-version: ${{ matrix.cuda }}
gcc-version: ${{ matrix.gcc }}
- name: Set up home
run: |
echo "HOME=/home/ubuntu" >> "${GITHUB_ENV}"
- name: Install latest stable
uses: dtolnay/rust-toolchain@e97e2d8cc328f1b50210efc529dca0028893a2d9 # zizmor: ignore[stale-action-refs] this action doesn't create releases
uses: dtolnay/rust-toolchain@315e265cd78dad1e1dcf3a5074f6d6c47029d5aa
with:
toolchain: stable
- name: Enable nvidia multi-process service
run: |
nvidia-cuda-mps-control -d
- name: Run tests
run: |
if [[ "${IS_PR}" == "true" ]]; then
make test_integer_short_run_gpu
else
make test_integer_long_run_gpu
fi
make test_integer_long_run_gpu
slack-notify:
name: gpu_integer_long_run_tests/slack-notify
name: Slack Notification
needs: [ setup-instance, cuda-tests ]
runs-on: ubuntu-latest
if: ${{ always() && needs.cuda-tests.result != 'skipped' && failure() }}
continue-on-error: true
steps:
- name: Send message
uses: rtCamp/action-slack-notify@e31e87e03dd19038e411e38ae27cbad084a90661
uses: rtCamp/action-slack-notify@c33737706dea87cd7784c687dadc9adf1be59990
env:
SLACK_COLOR: ${{ needs.cuda-tests.result }}
SLACK_MESSAGE: "Integer GPU long run tests finished with status: ${{ needs.cuda-tests.result }}. (${{ env.ACTION_RUN_URL }})"
SLACK_MESSAGE: "Integer GPU H100 long run tests finished with status: ${{ needs.cuda-tests.result }}. (${{ env.ACTION_RUN_URL }})"
teardown-instance:
name: gpu_integer_long_run_tests/teardown-instance
if: ${{ always() && needs.setup-instance.result == 'success' }}
name: Teardown instance (gpu-tests)
if: ${{ always() && needs.setup-instance.result != 'skipped' }}
needs: [ setup-instance, cuda-tests ]
runs-on: ubuntu-latest
steps:
- name: Stop instance
id: stop-instance
uses: zama-ai/slab-github-runner@5aee5d157f4a0201e5eaefc9cc648e5f9f5472a5 # v1.6.0
uses: zama-ai/slab-github-runner@98f0788261a7323d5d695a883e20df36591a92b7
with:
mode: stop
github-token: ${{ secrets.SLAB_ACTION_TOKEN }}
@@ -153,7 +107,8 @@ jobs:
- name: Slack Notification
if: ${{ failure() }}
uses: rtCamp/action-slack-notify@e31e87e03dd19038e411e38ae27cbad084a90661
continue-on-error: true
uses: rtCamp/action-slack-notify@c33737706dea87cd7784c687dadc9adf1be59990
env:
SLACK_COLOR: ${{ job.status }}
SLACK_MESSAGE: "Instance teardown (gpu-long-run-tests) finished with status: ${{ job.status }}. (${{ env.ACTION_RUN_URL }})"

View File

@@ -1,182 +0,0 @@
# Compile and test tfhe-cuda-backend on an AWS instance
name: gpu_memory_sanitizer
env:
CARGO_TERM_COLOR: always
ACTION_RUN_URL: ${{ github.server_url }}/${{ github.repository }}/actions/runs/${{ github.run_id }}
RUSTFLAGS: "-C target-cpu=native"
RUST_BACKTRACE: "full"
RUST_MIN_STACK: "8388608"
SLACK_CHANNEL: ${{ secrets.SLACK_CHANNEL }}
SLACK_ICON: https://pbs.twimg.com/profile_images/1274014582265298945/OjBKP9kn_400x400.png
SLACK_USERNAME: ${{ secrets.BOT_USERNAME }}
SLACK_WEBHOOK: ${{ secrets.SLACK_WEBHOOK }}
SLACKIFY_MARKDOWN: true
IS_PULL_REQUEST: ${{ github.event_name == 'pull_request' }}
PULL_REQUEST_MD_LINK: ""
CHECKOUT_TOKEN: ${{ secrets.REPO_CHECKOUT_TOKEN || secrets.GITHUB_TOKEN }}
# Secrets will be available only to zama-ai organization members
SECRETS_AVAILABLE: ${{ secrets.JOB_SECRET != '' }}
EXTERNAL_CONTRIBUTION_RUNNER: "gpu_ubuntu-22.04"
on:
# Allows you to run this workflow manually from the Actions tab as an alternative.
pull_request:
types: [ labeled ]
workflow_dispatch:
permissions:
contents: read
# zizmor: ignore[concurrency-limits] concurrency is managed after instance setup to ensure safe provisioning
jobs:
should-run:
name: gpu_memory_sanitizer/should-run
runs-on: ubuntu-latest
permissions:
pull-requests: read # Needed to check for file change
outputs:
gpu_test: ${{ env.IS_PULL_REQUEST == 'false' || steps.changed-files.outputs.gpu_any_changed }}
steps:
- name: Checkout tfhe-rs
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd
with:
fetch-depth: 0
persist-credentials: 'false'
token: ${{ env.CHECKOUT_TOKEN }}
- name: Check for file changes
id: changed-files
uses: tj-actions/changed-files@9426d40962ed5378910ee2e21d5f8c6fcbf2dd96 # v47.0.6
with:
files_yaml: |
gpu:
- Cargo.toml
- tfhe/Cargo.toml
- tfhe/build.rs
- backends/tfhe-cuda-backend/**
- tfhe/src/core_crypto/gpu/**
- tfhe/src/integer/gpu/**
- tfhe/src/shortint/parameters/**
- tfhe/src/high_level_api/**
- '.github/workflows/gpu_memory_sanitizer.yml'
setup-instance:
name: gpu_memory_sanitizer/setup-instance
needs: should-run
runs-on: ubuntu-latest
if: github.event_name == 'workflow_dispatch' ||
(github.event.action == 'labeled' && github.event.label.name == 'approved' && needs.should-run.outputs.gpu_test == 'true')
outputs:
runner-name: ${{ steps.start-remote-instance.outputs.label || steps.start-github-instance.outputs.runner_group }}
steps:
- name: Start remote instance
id: start-remote-instance
if: env.SECRETS_AVAILABLE == 'true'
uses: zama-ai/slab-github-runner@5aee5d157f4a0201e5eaefc9cc648e5f9f5472a5 # v1.6.0
with:
mode: start
github-token: ${{ secrets.SLAB_ACTION_TOKEN }}
slab-url: ${{ secrets.SLAB_BASE_URL }}
job-secret: ${{ secrets.JOB_SECRET }}
backend: hyperstack
profile: gpu-test
# This instance will be spawned especially for pull-request from forked repository
- name: Start GitHub instance
id: start-github-instance
if: env.SECRETS_AVAILABLE == 'false'
run: |
echo "runner_group=${EXTERNAL_CONTRIBUTION_RUNNER}" >> "$GITHUB_OUTPUT"
cuda-tests-linux:
name: gpu_memory_sanitizer/cuda-tests-linux
needs: [ setup-instance ]
if: github.event_name != 'pull_request' ||
(github.event_name == 'pull_request' && needs.setup-instance.result != 'skipped')
concurrency:
group: ${{ github.workflow_ref }}
cancel-in-progress: ${{ github.ref != 'refs/heads/main' }}
runs-on: ${{ needs.setup-instance.outputs.runner-name }}
timeout-minutes: 240
strategy:
fail-fast: false
# explicit include-based build matrix, of known valid options
matrix:
include:
- os: ubuntu-22.04
cuda: "12.8"
gcc: 11
steps:
- name: Checkout tfhe-rs
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd
with:
persist-credentials: 'false'
token: ${{ env.CHECKOUT_TOKEN }}
- name: Setup Hyperstack dependencies
uses: ./.github/actions/gpu_setup
with:
cuda-version: ${{ matrix.cuda }}
gcc-version: ${{ matrix.gcc }}
github-instance: ${{ env.SECRETS_AVAILABLE == 'false' }}
- name: Find tools
run: |
find /usr -executable -name "compute-sanitizer"
- name: Install latest stable
uses: dtolnay/rust-toolchain@b3b07ba8b418998c39fb20f53e8b695cdcc8de1b # zizmor: ignore[stale-action-refs] this action doesn't create releases
with:
toolchain: stable
- name: Run memory sanitizer
run: |
make test_high_level_api_gpu_sanitizer
slack-notify:
name: gpu_memory_sanitizer/slack-notify
needs: [ setup-instance, cuda-tests-linux ]
runs-on: ubuntu-latest
if: ${{ always() && needs.cuda-tests-linux.result != 'skipped' && failure() }}
continue-on-error: true
steps:
- name: Set pull-request URL
if: env.SECRETS_AVAILABLE == 'true' && github.event_name == 'pull_request'
run: |
echo "PULL_REQUEST_MD_LINK=[pull-request](${PR_BASE_URL}${PR_NUMBER}), " >> "${GITHUB_ENV}"
env:
PR_BASE_URL: ${{ vars.PR_BASE_URL }}
PR_NUMBER: ${{ github.event.pull_request.number }}
- name: Send message
if: env.SECRETS_AVAILABLE == 'true'
uses: rtCamp/action-slack-notify@e31e87e03dd19038e411e38ae27cbad084a90661
env:
SLACK_COLOR: ${{ needs.cuda-tests-linux.result }}
SLACK_MESSAGE: "GPU Memory Checks tests finished with status: ${{ needs.cuda-tests-linux.result }}. (${{ env.PULL_REQUEST_MD_LINK }}[action run](${{ env.ACTION_RUN_URL }}))"
teardown-instance:
name: gpu_memory_sanitizer/teardown-instance
if: ${{ always() && needs.setup-instance.result == 'success' }}
needs: [ setup-instance, cuda-tests-linux ]
runs-on: ubuntu-latest
steps:
- name: Stop remote instance
id: stop-instance
if: env.SECRETS_AVAILABLE == 'true'
uses: zama-ai/slab-github-runner@5aee5d157f4a0201e5eaefc9cc648e5f9f5472a5 # v1.6.0
with:
mode: stop
github-token: ${{ secrets.SLAB_ACTION_TOKEN }}
slab-url: ${{ secrets.SLAB_BASE_URL }}
job-secret: ${{ secrets.JOB_SECRET }}
label: ${{ needs.setup-instance.outputs.runner-name }}
- name: Slack Notification
if: ${{ failure() }}
uses: rtCamp/action-slack-notify@e31e87e03dd19038e411e38ae27cbad084a90661
env:
SLACK_COLOR: ${{ job.status }}
SLACK_MESSAGE: "Instance teardown (gpu-memory-sanitizer) finished with status: ${{ job.status }}. (${{ env.ACTION_RUN_URL }})"

View File

@@ -1,182 +0,0 @@
# Compile and test tfhe-cuda-backend on an AWS instance
name: gpu_memory_sanitizer_h100
env:
CARGO_TERM_COLOR: always
ACTION_RUN_URL: ${{ github.server_url }}/${{ github.repository }}/actions/runs/${{ github.run_id }}
RUSTFLAGS: "-C target-cpu=native"
RUST_BACKTRACE: "full"
RUST_MIN_STACK: "8388608"
SLACK_CHANNEL: ${{ secrets.SLACK_CHANNEL }}
SLACK_ICON: https://pbs.twimg.com/profile_images/1274014582265298945/OjBKP9kn_400x400.png
SLACK_USERNAME: ${{ secrets.BOT_USERNAME }}
SLACK_WEBHOOK: ${{ secrets.SLACK_WEBHOOK }}
SLACKIFY_MARKDOWN: true
IS_PULL_REQUEST: ${{ github.event_name == 'pull_request' }}
PULL_REQUEST_MD_LINK: ""
CHECKOUT_TOKEN: ${{ secrets.REPO_CHECKOUT_TOKEN || secrets.GITHUB_TOKEN }}
# Secrets will be available only to zama-ai organization members
SECRETS_AVAILABLE: ${{ secrets.JOB_SECRET != '' }}
EXTERNAL_CONTRIBUTION_RUNNER: "gpu_ubuntu-22.04"
on:
# Allows you to run this workflow manually from the Actions tab as an alternative.
pull_request:
types: [ labeled ]
workflow_dispatch:
permissions:
contents: read
# zizmor: ignore[concurrency-limits] concurrency is managed after instance setup to ensure safe provisioning
jobs:
should-run:
name: gpu_memory_sanitizer_h100/should-run
runs-on: ubuntu-latest
permissions:
pull-requests: read # Needed to check for file change
outputs:
gpu_test: ${{ env.IS_PULL_REQUEST == 'false' || steps.changed-files.outputs.gpu_any_changed }}
steps:
- name: Checkout tfhe-rs
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd
with:
fetch-depth: 0
persist-credentials: 'false'
token: ${{ env.CHECKOUT_TOKEN }}
- name: Check for file changes
id: changed-files
uses: tj-actions/changed-files@9426d40962ed5378910ee2e21d5f8c6fcbf2dd96 # v47.0.6
with:
files_yaml: |
gpu:
- Cargo.toml
- tfhe/Cargo.toml
- tfhe/build.rs
- backends/tfhe-cuda-backend/**
- tfhe/src/core_crypto/gpu/**
- tfhe/src/integer/gpu/**
- tfhe/src/shortint/parameters/**
- tfhe/src/high_level_api/**
- '.github/workflows/gpu_memory_sanitizer_h100.yml'
setup-instance:
name: gpu_memory_sanitizer/setup-instance
needs: should-run
runs-on: ubuntu-latest
if: github.event_name == 'workflow_dispatch' ||
(github.event.action == 'labeled' && github.event.label.name == 'approved' && needs.should-run.outputs.gpu_test == 'true')
outputs:
runner-name: ${{ steps.start-remote-instance.outputs.label || steps.start-github-instance.outputs.runner_group }}
steps:
- name: Start remote instance
id: start-remote-instance
if: env.SECRETS_AVAILABLE == 'true'
uses: zama-ai/slab-github-runner@5aee5d157f4a0201e5eaefc9cc648e5f9f5472a5 # v1.6.0
with:
mode: start
github-token: ${{ secrets.SLAB_ACTION_TOKEN }}
slab-url: ${{ secrets.SLAB_BASE_URL }}
job-secret: ${{ secrets.JOB_SECRET }}
backend: hyperstack
profile: single-h100
# This instance will be spawned especially for pull-request from forked repository
- name: Start GitHub instance
id: start-github-instance
if: env.SECRETS_AVAILABLE == 'false'
run: |
echo "runner_group=${EXTERNAL_CONTRIBUTION_RUNNER}" >> "$GITHUB_OUTPUT"
cuda-tests-linux:
name: gpu_memory_sanitizer/cuda-tests-linux
needs: [ setup-instance ]
if: github.event_name != 'pull_request' ||
(github.event_name == 'pull_request' && needs.setup-instance.result != 'skipped')
concurrency:
group: ${{ github.workflow_ref }}
cancel-in-progress: ${{ github.ref != 'refs/heads/main' }}
runs-on: ${{ needs.setup-instance.outputs.runner-name }}
timeout-minutes: 240
strategy:
fail-fast: false
# explicit include-based build matrix, of known valid options
matrix:
include:
- os: ubuntu-22.04
cuda: "12.8"
gcc: 11
steps:
- name: Checkout tfhe-rs
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd
with:
persist-credentials: 'false'
token: ${{ env.CHECKOUT_TOKEN }}
- name: Setup Hyperstack dependencies
uses: ./.github/actions/gpu_setup
with:
cuda-version: ${{ matrix.cuda }}
gcc-version: ${{ matrix.gcc }}
github-instance: ${{ env.SECRETS_AVAILABLE == 'false' }}
- name: Find tools
run: |
find /usr -executable -name "compute-sanitizer"
- name: Install latest stable
uses: dtolnay/rust-toolchain@b3b07ba8b418998c39fb20f53e8b695cdcc8de1b # zizmor: ignore[stale-action-refs] this action doesn't create releases
with:
toolchain: stable
- name: Run memory sanitizer
run: |
make test_high_level_api_gpu_sanitizer
slack-notify:
name: gpu_memory_sanitizer/slack-notify
needs: [ setup-instance, cuda-tests-linux ]
runs-on: ubuntu-latest
if: ${{ always() && needs.cuda-tests-linux.result != 'skipped' && failure() }}
continue-on-error: true
steps:
- name: Set pull-request URL
if: env.SECRETS_AVAILABLE == 'true' && github.event_name == 'pull_request'
run: |
echo "PULL_REQUEST_MD_LINK=[pull-request](${PR_BASE_URL}${PR_NUMBER}), " >> "${GITHUB_ENV}"
env:
PR_BASE_URL: ${{ vars.PR_BASE_URL }}
PR_NUMBER: ${{ github.event.pull_request.number }}
- name: Send message
if: env.SECRETS_AVAILABLE == 'true'
uses: rtCamp/action-slack-notify@e31e87e03dd19038e411e38ae27cbad084a90661
env:
SLACK_COLOR: ${{ needs.cuda-tests-linux.result }}
SLACK_MESSAGE: "GPU Memory Checks tests finished with status: ${{ needs.cuda-tests-linux.result }}. (${{ env.PULL_REQUEST_MD_LINK }}[action run](${{ env.ACTION_RUN_URL }}))"
teardown-instance:
name: gpu_memory_sanitizer/teardown-instance
if: ${{ always() && needs.setup-instance.result == 'success' }}
needs: [ setup-instance, cuda-tests-linux ]
runs-on: ubuntu-latest
steps:
- name: Stop remote instance
id: stop-instance
if: env.SECRETS_AVAILABLE == 'true'
uses: zama-ai/slab-github-runner@5aee5d157f4a0201e5eaefc9cc648e5f9f5472a5 # v1.6.0
with:
mode: stop
github-token: ${{ secrets.SLAB_ACTION_TOKEN }}
slab-url: ${{ secrets.SLAB_BASE_URL }}
job-secret: ${{ secrets.JOB_SECRET }}
label: ${{ needs.setup-instance.outputs.runner-name }}
- name: Slack Notification
if: ${{ failure() }}
uses: rtCamp/action-slack-notify@e31e87e03dd19038e411e38ae27cbad084a90661
env:
SLACK_COLOR: ${{ job.status }}
SLACK_MESSAGE: "Instance teardown (gpu-memory-sanitizer-h100) finished with status: ${{ job.status }}. (${{ env.ACTION_RUN_URL }})"

View File

@@ -1,5 +1,5 @@
# Perform tfhe-cuda-backend post-commit checks on an AWS instance
name: gpu_pcc
# Perfom tfhe-cuda-backend post-commit checks on an AWS instance
name: TFHE Cuda Backend - Post-commit Checks
env:
CARGO_TERM_COLOR: always
@@ -11,34 +11,20 @@ env:
SLACK_ICON: https://pbs.twimg.com/profile_images/1274014582265298945/OjBKP9kn_400x400.png
SLACK_USERNAME: ${{ secrets.BOT_USERNAME }}
SLACK_WEBHOOK: ${{ secrets.SLACK_WEBHOOK }}
SLACKIFY_MARKDOWN: true
PULL_REQUEST_MD_LINK: ""
CHECKOUT_TOKEN: ${{ secrets.REPO_CHECKOUT_TOKEN || secrets.GITHUB_TOKEN }}
# Secrets will be available only to zama-ai organization members
SECRETS_AVAILABLE: ${{ secrets.JOB_SECRET != '' }}
EXTERNAL_CONTRIBUTION_RUNNER: "large_ubuntu_16-22.04"
CUDA_KEYRING_PACKAGE: cuda-keyring_1.1-1_all.deb
CUDA_KEYRING_SHA: "d93190d50b98ad4699ff40f4f7af50f16a76dac3bb8da1eaaf366d47898ff8df"
on:
pull_request:
permissions:
contents: read
# zizmor: ignore[concurrency-limits] only Zama organization members can trigger this workflow (via manual approval for PR from forks)
jobs:
setup-instance:
name: gpu_pcc/setup-instance
name: Setup instance (cuda-pcc)
runs-on: ubuntu-latest
outputs:
runner-name: ${{ steps.start-remote-instance.outputs.label || steps.start-github-instance.outputs.runner_group }}
runner-name: ${{ steps.start-instance.outputs.label }}
steps:
- name: Start remote instance
id: start-remote-instance
if: env.SECRETS_AVAILABLE == 'true'
uses: zama-ai/slab-github-runner@5aee5d157f4a0201e5eaefc9cc648e5f9f5472a5 # v1.6.0
- name: Start instance
id: start-instance
uses: zama-ai/slab-github-runner@98f0788261a7323d5d695a883e20df36591a92b7
with:
mode: start
github-token: ${{ secrets.SLAB_ACTION_TOKEN }}
@@ -47,18 +33,11 @@ jobs:
backend: aws
profile: gpu-build
# This instance will be spawned especially for pull-request from forked repository
- name: Start GitHub instance
id: start-github-instance
if: env.SECRETS_AVAILABLE == 'false'
run: |
echo "runner_group=${EXTERNAL_CONTRIBUTION_RUNNER}" >> "$GITHUB_OUTPUT"
cuda-pcc:
name: gpu_pcc/cuda-pcc (bpr)
name: CUDA post-commit checks
needs: setup-instance
concurrency:
group: ${{ github.workflow_ref }}
group: ${{ github.workflow }}_${{ github.ref }}
cancel-in-progress: true
runs-on: ${{ needs.setup-instance.outputs.runner-name }}
strategy:
@@ -74,38 +53,17 @@ jobs:
steps:
- name: Checkout tfhe-rs
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd
uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683
with:
persist-credentials: 'false'
token: ${{ env.CHECKOUT_TOKEN }}
token: ${{ secrets.FHE_ACTIONS_TOKEN }}
- name: Install CUDA and other dependencies
if: env.SECRETS_AVAILABLE == 'false'
shell: bash
- name: Set up home
run: |
# Use Sed to extract a value from a string, this cannot be done with the ${variable//search/replace} pattern.
# shellcheck disable=SC2001
TOOLKIT_VERSION="$(echo "${CUDA_VERSION}" | sed 's/\(.*\)\.\(.*\)/\1-\2/')"
wget https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2204/x86_64/"${CUDA_KEYRING_PACKAGE}"
echo "${CUDA_KEYRING_SHA} ${CUDA_KEYRING_PACKAGE}" > checksum
sha256sum -c checksum
sudo dpkg -i "${CUDA_KEYRING_PACKAGE}"
# Disable unattended-upgrades to avoid lock issues
sudo systemctl mask --now unattended-upgrades
sudo systemctl stop --now unattended-upgrades
sudo apt-get clean
sudo rm -rf /var/lib/apt/lists/*
sudo apt purge -y unattended-upgrades
sudo apt update
sudo apt -y install "cuda-toolkit-${TOOLKIT_VERSION}" cmake-format python3-venv
env:
CUDA_VERSION: ${{ matrix.cuda }}
echo "HOME=/home/ubuntu" >> "${GITHUB_ENV}"
- name: Install latest stable
uses: dtolnay/rust-toolchain@e97e2d8cc328f1b50210efc529dca0028893a2d9 # zizmor: ignore[stale-action-refs] this action doesn't create releases
uses: dtolnay/rust-toolchain@315e265cd78dad1e1dcf3a5074f6d6c47029d5aa
with:
toolchain: stable
@@ -115,25 +73,18 @@ jobs:
echo "CUDA_PATH=$CUDA_PATH" >> "${GITHUB_ENV}"
echo "$CUDA_PATH/bin" >> "${GITHUB_PATH}"
echo "LD_LIBRARY_PATH=$CUDA_PATH/lib:$LD_LIBRARY_PATH" >> "${GITHUB_ENV}"
echo "CUDACXX=/usr/local/cuda-${CUDA_VERSION}/bin/nvcc" >> "${GITHUB_ENV}"
env:
CUDA_VERSION: ${{ matrix.cuda }}
echo "CUDACXX=/usr/local/cuda-${{ matrix.cuda }}/bin/nvcc" >> "${GITHUB_ENV}"
# Specify the correct host compilers
- name: Export gcc and g++ variables
if: ${{ !cancelled() }}
run: |
{
echo "CC=/usr/bin/gcc-${GCC_VERSION}";
echo "CXX=/usr/bin/g++-${GCC_VERSION}";
echo "CUDAHOSTCXX=/usr/bin/g++-${GCC_VERSION}";
echo "CC=/usr/bin/gcc-${{ matrix.gcc }}";
echo "CXX=/usr/bin/g++-${{ matrix.gcc }}";
echo "CUDAHOSTCXX=/usr/bin/g++-${{ matrix.gcc }}";
echo "HOME=/home/ubuntu";
} >> "${GITHUB_ENV}"
env:
GCC_VERSION: ${{ matrix.gcc }}
- name: Run semgrep and lint checks on CUDA code
run: |
make semgrep_and_lint_gpu_code
- name: Run fmt checks
run: |
@@ -143,40 +94,23 @@ jobs:
run: |
make pcc_gpu
- name: Run semver checks on tfhe-cuda-backend
run: |
make semver_check_cuda_backend
- name: Check build with hpu enabled
run: |
make clippy_gpu_hpu
- name: Set pull-request URL
if: ${{ failure() && github.event_name == 'pull_request' }}
run: |
echo "PULL_REQUEST_MD_LINK=[pull-request](${PR_BASE_URL}${PR_NUMBER}), " >> "${GITHUB_ENV}"
env:
PR_BASE_URL: ${{ vars.PR_BASE_URL }}
PR_NUMBER: ${{ github.event.pull_request.number }}
- name: Slack Notification
if: ${{ failure() && env.SECRETS_AVAILABLE == 'true' }}
if: ${{ failure() }}
continue-on-error: true
uses: rtCamp/action-slack-notify@e31e87e03dd19038e411e38ae27cbad084a90661
uses: rtCamp/action-slack-notify@c33737706dea87cd7784c687dadc9adf1be59990
env:
SLACK_COLOR: ${{ job.status }}
SLACK_MESSAGE: "CUDA AWS post-commit checks finished with status: ${{ job.status }}. (${{ env.PULL_REQUEST_MD_LINK }}[action run](${{ env.ACTION_RUN_URL }}))"
SLACK_MESSAGE: "CUDA AWS post-commit checks finished with status: ${{ job.status }}. (${{ env.ACTION_RUN_URL }})"
teardown-instance:
name: cuda_pcc/teardown-instance
if: ${{ always() && needs.setup-instance.result == 'success' }}
name: Teardown instance (cuda-pcc)
if: ${{ always() && needs.setup-instance.result != 'skipped' }}
needs: [ setup-instance, cuda-pcc ]
runs-on: ubuntu-latest
steps:
- name: Stop remote instance
- name: Stop instance
id: stop-instance
if: env.SECRETS_AVAILABLE == 'true'
uses: zama-ai/slab-github-runner@5aee5d157f4a0201e5eaefc9cc648e5f9f5472a5 # v1.6.0
uses: zama-ai/slab-github-runner@98f0788261a7323d5d695a883e20df36591a92b7
with:
mode: stop
github-token: ${{ secrets.SLAB_ACTION_TOKEN }}
@@ -186,7 +120,8 @@ jobs:
- name: Slack Notification
if: ${{ failure() }}
uses: rtCamp/action-slack-notify@e31e87e03dd19038e411e38ae27cbad084a90661
continue-on-error: true
uses: rtCamp/action-slack-notify@c33737706dea87cd7784c687dadc9adf1be59990
env:
SLACK_COLOR: ${{ job.status }}
SLACK_MESSAGE: "Instance teardown (cuda-pcc) finished with status: ${{ job.status }}. (${{ env.ACTION_RUN_URL }})"

View File

@@ -1,5 +1,5 @@
# Signed integer GPU tests on an RTXA6000 VM on hyperstack with classical PBS
name: gpu_signed_integer_classic_tests
name: TFHE Cuda Backend - Signed integer tests with classical PBS
env:
CARGO_TERM_COLOR: always
@@ -11,45 +11,33 @@ env:
SLACK_ICON: https://pbs.twimg.com/profile_images/1274014582265298945/OjBKP9kn_400x400.png
SLACK_USERNAME: ${{ secrets.BOT_USERNAME }}
SLACK_WEBHOOK: ${{ secrets.SLACK_WEBHOOK }}
SLACKIFY_MARKDOWN: true
IS_PULL_REQUEST: ${{ github.event_name == 'pull_request' }}
PULL_REQUEST_MD_LINK: ""
CHECKOUT_TOKEN: ${{ secrets.REPO_CHECKOUT_TOKEN || secrets.GITHUB_TOKEN }}
# Secrets will be available only to zama-ai organization members
SECRETS_AVAILABLE: ${{ secrets.JOB_SECRET != '' }}
EXTERNAL_CONTRIBUTION_RUNNER: "gpu_ubuntu-22.04"
on:
# Allows you to run this workflow manually from the Actions tab as an alternative.
workflow_dispatch:
pull_request:
types: [ labeled ]
permissions:
contents: read
# zizmor: ignore[concurrency-limits] concurrency is managed after instance setup to ensure safe provisioning
types: [ labeled ]
jobs:
should-run:
name: gpu_signed_integer_classic_tests/should-run
runs-on: ubuntu-latest
permissions:
pull-requests: read # Needed to check for file change
pull-requests: write
outputs:
gpu_test: ${{ env.IS_PULL_REQUEST == 'false' || steps.changed-files.outputs.gpu_any_changed }}
steps:
- name: Checkout tfhe-rs
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd
uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683
with:
fetch-depth: 0
persist-credentials: 'false'
token: ${{ env.CHECKOUT_TOKEN }}
token: ${{ secrets.FHE_ACTIONS_TOKEN }}
- name: Check for file changes
id: changed-files
uses: tj-actions/changed-files@9426d40962ed5378910ee2e21d5f8c6fcbf2dd96 # v47.0.6
uses: tj-actions/changed-files@bab30c2299617f6615ec02a68b9a40d10bd21366
with:
since_last_remote_commit: true
files_yaml: |
gpu:
- tfhe/Cargo.toml
@@ -63,23 +51,24 @@ jobs:
- tfhe/src/shortint/parameters/**
- tfhe/src/high_level_api/**
- tfhe/src/c_api/**
- 'tfhe/docs/**.md'
- '.github/workflows/gpu_signed_integer_classic_tests.yml'
- scripts/integer-tests.sh
- ci/slab.toml
setup-instance:
name: gpu_signed_integer_classic_tests/setup-instance
name: Setup instance (cuda-signed-classic-tests)
needs: should-run
if: github.event_name != 'pull_request' ||
(github.event.action != 'labeled' && needs.should-run.outputs.gpu_test == 'true') ||
(github.event.action == 'labeled' && github.event.label.name == 'approved' && needs.should-run.outputs.gpu_test == 'true')
runs-on: ubuntu-latest
outputs:
runner-name: ${{ steps.start-remote-instance.outputs.label || steps.start-github-instance.outputs.runner_group }}
runner-name: ${{ steps.start-instance.outputs.label }}
steps:
- name: Start remote instance
id: start-remote-instance
if: env.SECRETS_AVAILABLE == 'true'
uses: zama-ai/slab-github-runner@5aee5d157f4a0201e5eaefc9cc648e5f9f5472a5 # v1.6.0
- name: Start instance
id: start-instance
uses: zama-ai/slab-github-runner@98f0788261a7323d5d695a883e20df36591a92b7
with:
mode: start
github-token: ${{ secrets.SLAB_ACTION_TOKEN }}
@@ -88,20 +77,13 @@ jobs:
backend: hyperstack
profile: gpu-test
# This instance will be spawned especially for pull-request from forked repository
- name: Start GitHub instance
id: start-github-instance
if: env.SECRETS_AVAILABLE == 'false'
run: |
echo "runner_group=${EXTERNAL_CONTRIBUTION_RUNNER}" >> "$GITHUB_OUTPUT"
cuda-tests-linux:
name: gpu_signed_integer_classic_tests/cuda-tests-linux
name: CUDA signed integer tests with classical PBS
needs: [ should-run, setup-instance ]
if: github.event_name != 'pull_request' ||
(github.event_name == 'pull_request' && needs.setup-instance.result != 'skipped')
concurrency:
group: ${{ github.workflow_ref }}
group: ${{ github.workflow }}_${{ github.ref }}
cancel-in-progress: ${{ github.ref != 'refs/heads/main' }}
runs-on: ${{ needs.setup-instance.outputs.runner-name }}
strategy:
@@ -110,65 +92,53 @@ jobs:
matrix:
include:
- os: ubuntu-22.04
cuda: "12.8"
cuda: "12.2"
gcc: 11
steps:
- name: Checkout tfhe-rs
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd
with:
persist-credentials: 'false'
token: ${{ env.CHECKOUT_TOKEN }}
uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683
- name: Setup Hyperstack dependencies
uses: ./.github/actions/gpu_setup
uses: ./.github/actions/hyperstack_setup
with:
cuda-version: ${{ matrix.cuda }}
gcc-version: ${{ matrix.gcc }}
github-instance: ${{ env.SECRETS_AVAILABLE == 'false' }}
- name: Set up home
run: |
echo "HOME=/home/ubuntu" >> "${GITHUB_ENV}"
- name: Install latest stable
uses: dtolnay/rust-toolchain@e97e2d8cc328f1b50210efc529dca0028893a2d9 # zizmor: ignore[stale-action-refs] this action doesn't create releases
uses: dtolnay/rust-toolchain@315e265cd78dad1e1dcf3a5074f6d6c47029d5aa
with:
toolchain: stable
- name: Enable nvidia multi-process service
run: |
nvidia-cuda-mps-control -d
- name: Run signed integer tests
run: |
BIG_TESTS_INSTANCE=TRUE make test_signed_integer_gpu_ci
slack-notify:
name: gpu_signed_integer_classic_tests/slack-notify
name: Slack Notification
needs: [ setup-instance, cuda-tests-linux ]
runs-on: ubuntu-latest
if: ${{ always() && needs.cuda-tests-linux.result != 'skipped' && failure() }}
continue-on-error: true
steps:
- name: Set pull-request URL
if: env.SECRETS_AVAILABLE == 'true' && github.event_name == 'pull_request'
run: |
echo "PULL_REQUEST_MD_LINK=[pull-request](${PR_BASE_URL}${PR_NUMBER}), " >> "${GITHUB_ENV}"
env:
PR_BASE_URL: ${{ vars.PR_BASE_URL }}
PR_NUMBER: ${{ github.event.pull_request.number }}
- name: Send message
if: env.SECRETS_AVAILABLE == 'true'
uses: rtCamp/action-slack-notify@e31e87e03dd19038e411e38ae27cbad084a90661
uses: rtCamp/action-slack-notify@c33737706dea87cd7784c687dadc9adf1be59990
env:
SLACK_COLOR: ${{ needs.cuda-tests-linux.result }}
SLACK_MESSAGE: "Integer GPU signed integer tests with classical PBS finished with status: ${{ needs.cuda-tests-linux.result }}. (${{ env.PULL_REQUEST_MD_LINK }}[action run](${{ env.ACTION_RUN_URL }}))"
SLACK_MESSAGE: "Integer GPU signed integer tests with classical PBS finished with status: ${{ needs.cuda-tests-linux.result }}. (${{ env.ACTION_RUN_URL }})"
teardown-instance:
name: gpu_signed_integer_classic_tests/teardown-instance
if: ${{ always() && needs.setup-instance.result == 'success' }}
name: Teardown instance (cuda-signed-classic-tests)
if: ${{ always() && needs.setup-instance.result != 'skipped' }}
needs: [ setup-instance, cuda-tests-linux ]
runs-on: ubuntu-latest
steps:
- name: Stop remote instance
- name: Stop instance
id: stop-instance
if: env.SECRETS_AVAILABLE == 'true'
uses: zama-ai/slab-github-runner@5aee5d157f4a0201e5eaefc9cc648e5f9f5472a5 # v1.6.0
uses: zama-ai/slab-github-runner@98f0788261a7323d5d695a883e20df36591a92b7
with:
mode: stop
github-token: ${{ secrets.SLAB_ACTION_TOKEN }}
@@ -178,7 +148,8 @@ jobs:
- name: Slack Notification
if: ${{ failure() }}
uses: rtCamp/action-slack-notify@e31e87e03dd19038e411e38ae27cbad084a90661
continue-on-error: true
uses: rtCamp/action-slack-notify@c33737706dea87cd7784c687dadc9adf1be59990
env:
SLACK_COLOR: ${{ job.status }}
SLACK_MESSAGE: "Instance teardown (cuda-signed-classic-tests) finished with status: ${{ job.status }}. (${{ env.ACTION_RUN_URL }})"

View File

@@ -1,5 +1,5 @@
# Signed integer GPU tests on an H100 VM on hyperstack
name: gpu_signed_integer_h100_tests
name: TFHE Cuda Backend - Signed integer tests on H100
env:
CARGO_TERM_COLOR: always
@@ -11,46 +11,33 @@ env:
SLACK_ICON: https://pbs.twimg.com/profile_images/1274014582265298945/OjBKP9kn_400x400.png
SLACK_USERNAME: ${{ secrets.BOT_USERNAME }}
SLACK_WEBHOOK: ${{ secrets.SLACK_WEBHOOK }}
SLACKIFY_MARKDOWN: true
IS_PULL_REQUEST: ${{ github.event_name == 'pull_request' }}
PULL_REQUEST_MD_LINK: ""
CHECKOUT_TOKEN: ${{ secrets.REPO_CHECKOUT_TOKEN || secrets.GITHUB_TOKEN }}
# Secrets will be available only to zama-ai organization members
SECRETS_AVAILABLE: ${{ secrets.JOB_SECRET != '' }}
EXTERNAL_CONTRIBUTION_RUNNER: "gpu_ubuntu-22.04"
on:
# Allows you to run this workflow manually from the Actions tab as an alternative.
workflow_dispatch:
pull_request:
types: [ labeled, opened, synchronize ]
permissions:
contents: read
# zizmor: ignore[concurrency-limits] concurrency is managed after instance setup to ensure safe provisioning
types: [ labeled ]
jobs:
should-run:
name: gpu_signed_integer_h100_tests/should-run
runs-on: ubuntu-latest
permissions:
pull-requests: read # Needed to check for file change
pull-requests: write
outputs:
gpu_test: ${{ env.IS_PULL_REQUEST == 'false' || steps.changed-files.outputs.gpu_any_changed }}
core_crypto_changed: ${{ steps.changed-files.outputs.core_crypto_any_changed }}
steps:
- name: Checkout tfhe-rs
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd
uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683
with:
fetch-depth: 0
persist-credentials: 'false'
token: ${{ env.CHECKOUT_TOKEN }}
token: ${{ secrets.FHE_ACTIONS_TOKEN }}
- name: Check for file changes
id: changed-files
uses: tj-actions/changed-files@9426d40962ed5378910ee2e21d5f8c6fcbf2dd96 # v47.0.6
uses: tj-actions/changed-files@bab30c2299617f6615ec02a68b9a40d10bd21366
with:
since_last_remote_commit: true
files_yaml: |
gpu:
- tfhe/Cargo.toml
@@ -64,25 +51,24 @@ jobs:
- tfhe/src/shortint/parameters/**
- tfhe/src/high_level_api/**
- tfhe/src/c_api/**
- 'tfhe/docs/**.md'
- '.github/workflows/gpu_signed_integer_h100_tests.yml'
- scripts/integer-tests.sh
core_crypto:
- tfhe/src/core_crypto/gpu/**
- ci/slab.toml
setup-instance:
name: gpu_signed_integer_h100_tests/setup-instance
name: Setup instance (cuda-h100-tests)
needs: should-run
if: github.event_name != 'pull_request' ||
(github.event.action == 'labeled' && github.event.label.name == 'approved' && needs.should-run.outputs.gpu_test == 'true') ||
(github.event.action != 'labeled' && needs.should-run.outputs.core_crypto_changed == 'true')
(github.event.action != 'labeled' && needs.should-run.outputs.gpu_test == 'true') ||
(github.event.action == 'labeled' && github.event.label.name == 'approved' && needs.should-run.outputs.gpu_test == 'true')
runs-on: ubuntu-latest
outputs:
runner-name: ${{ steps.start-remote-instance.outputs.label || steps.start-github-instance.outputs.runner_group }}
runner-name: ${{ steps.start-instance.outputs.label }}
steps:
- name: Start remote instance
id: start-remote-instance
if: env.SECRETS_AVAILABLE == 'true'
uses: zama-ai/slab-github-runner@5aee5d157f4a0201e5eaefc9cc648e5f9f5472a5 # v1.6.0
- name: Start instance
id: start-instance
uses: zama-ai/slab-github-runner@98f0788261a7323d5d695a883e20df36591a92b7
with:
mode: start
github-token: ${{ secrets.SLAB_ACTION_TOKEN }}
@@ -91,20 +77,13 @@ jobs:
backend: hyperstack
profile: single-h100
# This instance will be spawned especially for pull-request from forked repository
- name: Start GitHub instance
id: start-github-instance
if: env.SECRETS_AVAILABLE == 'false'
run: |
echo "runner_group=${EXTERNAL_CONTRIBUTION_RUNNER}" >> "$GITHUB_OUTPUT"
cuda-tests-linux:
name: gpu_signed_integer_h100_tests/cuda-tests-linux
name: CUDA H100 signed integer tests
needs: [ should-run, setup-instance ]
if: github.event_name != 'pull_request' ||
(github.event_name == 'pull_request' && needs.setup-instance.result != 'skipped')
concurrency:
group: ${{ github.workflow_ref }}
group: ${{ github.workflow }}_${{ github.ref }}
cancel-in-progress: ${{ github.ref != 'refs/heads/main' }}
runs-on: ${{ needs.setup-instance.outputs.runner-name }}
strategy:
@@ -113,65 +92,53 @@ jobs:
matrix:
include:
- os: ubuntu-22.04
cuda: "12.8"
cuda: "12.2"
gcc: 11
steps:
- name: Checkout tfhe-rs
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd
with:
persist-credentials: 'false'
token: ${{ env.CHECKOUT_TOKEN }}
uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683
- name: Setup Hyperstack dependencies
uses: ./.github/actions/gpu_setup
uses: ./.github/actions/hyperstack_setup
with:
cuda-version: ${{ matrix.cuda }}
gcc-version: ${{ matrix.gcc }}
github-instance: ${{ env.SECRETS_AVAILABLE == 'false' }}
- name: Set up home
run: |
echo "HOME=/home/ubuntu" >> "${GITHUB_ENV}"
- name: Install latest stable
uses: dtolnay/rust-toolchain@e97e2d8cc328f1b50210efc529dca0028893a2d9 # zizmor: ignore[stale-action-refs] this action doesn't create releases
uses: dtolnay/rust-toolchain@315e265cd78dad1e1dcf3a5074f6d6c47029d5aa
with:
toolchain: stable
- name: Enable nvidia multi-process service
run: |
nvidia-cuda-mps-control -d
- name: Run signed integer multi-bit tests
run: |
BIG_TESTS_INSTANCE=TRUE make test_signed_integer_multi_bit_gpu_ci
slack-notify:
name: gpu_signed_integer_h100_tests/slack-notify
name: Slack Notification
needs: [ setup-instance, cuda-tests-linux ]
runs-on: ubuntu-latest
if: ${{ always() && needs.cuda-tests-linux.result != 'skipped' && failure() }}
continue-on-error: true
steps:
- name: Set pull-request URL
if: env.SECRETS_AVAILABLE == 'true' && github.event_name == 'pull_request'
run: |
echo "PULL_REQUEST_MD_LINK=[pull-request](${PR_BASE_URL}${PR_NUMBER}), " >> "${GITHUB_ENV}"
env:
PR_BASE_URL: ${{ vars.PR_BASE_URL }}
PR_NUMBER: ${{ github.event.pull_request.number }}
- name: Send message
if: env.SECRETS_AVAILABLE == 'true'
uses: rtCamp/action-slack-notify@e31e87e03dd19038e411e38ae27cbad084a90661
uses: rtCamp/action-slack-notify@c33737706dea87cd7784c687dadc9adf1be59990
env:
SLACK_COLOR: ${{ needs.cuda-tests-linux.result }}
SLACK_MESSAGE: "Integer GPU H100 tests finished with status: ${{ needs.cuda-tests-linux.result }}. (${{ env.PULL_REQUEST_MD_LINK }}[action run](${{ env.ACTION_RUN_URL }}))"
SLACK_MESSAGE: "Integer GPU H100 tests finished with status: ${{ needs.cuda-tests-linux.result }}. (${{ env.ACTION_RUN_URL }})"
teardown-instance:
name: gpu_signed_integer_h100_tests/teardown-instance
if: ${{ always() && needs.setup-instance.result == 'success' }}
name: Teardown instance (cuda-h100-tests)
if: ${{ always() && needs.setup-instance.result != 'skipped' }}
needs: [ setup-instance, cuda-tests-linux ]
runs-on: ubuntu-latest
steps:
- name: Stop remote instance
- name: Stop instance
id: stop-instance
if: env.SECRETS_AVAILABLE == 'true'
uses: zama-ai/slab-github-runner@5aee5d157f4a0201e5eaefc9cc648e5f9f5472a5 # v1.6.0
uses: zama-ai/slab-github-runner@98f0788261a7323d5d695a883e20df36591a92b7
with:
mode: stop
github-token: ${{ secrets.SLAB_ACTION_TOKEN }}
@@ -181,7 +148,8 @@ jobs:
- name: Slack Notification
if: ${{ failure() }}
uses: rtCamp/action-slack-notify@e31e87e03dd19038e411e38ae27cbad084a90661
continue-on-error: true
uses: rtCamp/action-slack-notify@c33737706dea87cd7784c687dadc9adf1be59990
env:
SLACK_COLOR: ${{ job.status }}
SLACK_MESSAGE: "Instance teardown (cuda-h100-tests) finished with status: ${{ job.status }}. (${{ env.ACTION_RUN_URL }})"

View File

@@ -1,5 +1,5 @@
# Compile and test tfhe-cuda-backend signed integer on an AWS instance
name: gpu_signed_integer_tests
name: TFHE Cuda Backend - Signed integer tests
env:
CARGO_TERM_COLOR: always
@@ -11,46 +11,40 @@ env:
SLACK_ICON: https://pbs.twimg.com/profile_images/1274014582265298945/OjBKP9kn_400x400.png
SLACK_USERNAME: ${{ secrets.BOT_USERNAME }}
SLACK_WEBHOOK: ${{ secrets.SLACK_WEBHOOK }}
SLACKIFY_MARKDOWN: true
FAST_TESTS: TRUE
NIGHTLY_TESTS: FALSE
IS_PULL_REQUEST: ${{ github.event_name == 'pull_request' }}
PULL_REQUEST_MD_LINK: ""
CHECKOUT_TOKEN: ${{ secrets.REPO_CHECKOUT_TOKEN || secrets.GITHUB_TOKEN }}
# Secrets will be available only to zama-ai organization members
SECRETS_AVAILABLE: ${{ secrets.JOB_SECRET != '' }}
EXTERNAL_CONTRIBUTION_RUNNER: "gpu_ubuntu-22.04"
on:
# Allows you to run this workflow manually from the Actions tab as an alternative.
workflow_dispatch:
pull_request:
permissions:
contents: read
# zizmor: ignore[concurrency-limits] concurrency is managed after instance setup to ensure safe provisioning
types:
- opened
- synchronize
schedule:
# Nightly tests @ 1AM after each work day
- cron: "0 1 * * MON-FRI"
jobs:
should-run:
name: gpu_signed_integer_tests/should-run
runs-on: ubuntu-latest
permissions:
pull-requests: read # Needed to check for file change
pull-requests: write
outputs:
gpu_test: ${{ env.IS_PULL_REQUEST == 'false' || steps.changed-files.outputs.gpu_any_changed }}
steps:
- name: Checkout tfhe-rs
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd
uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683
with:
fetch-depth: 0
persist-credentials: 'false'
token: ${{ env.CHECKOUT_TOKEN }}
token: ${{ secrets.FHE_ACTIONS_TOKEN }}
- name: Check for file changes
id: changed-files
uses: tj-actions/changed-files@9426d40962ed5378910ee2e21d5f8c6fcbf2dd96 # v47.0.6
uses: tj-actions/changed-files@bab30c2299617f6615ec02a68b9a40d10bd21366
with:
since_last_remote_commit: true
files_yaml: |
gpu:
- tfhe/Cargo.toml
@@ -64,23 +58,24 @@ jobs:
- tfhe/src/shortint/parameters/**
- tfhe/src/high_level_api/**
- tfhe/src/c_api/**
- 'tfhe/docs/**.md'
- '.github/workflows/gpu_signed_integer_tests.yml'
- scripts/integer-tests.sh
- ci/slab.toml
setup-instance:
name: gpu_signed_integer_tests/setup-instance
name: Setup instance (cuda-signed-integer-tests)
runs-on: ubuntu-latest
needs: should-run
if: (github.event_name == 'schedule' && github.repository == 'zama-ai/tfhe-rs') ||
github.event_name == 'workflow_dispatch' ||
needs.should-run.outputs.gpu_test == 'true'
outputs:
runner-name: ${{ steps.start-remote-instance.outputs.label || steps.start-github-instance.outputs.runner_group }}
runner-name: ${{ steps.start-instance.outputs.label }}
steps:
- name: Start remote instance
id: start-remote-instance
if: env.SECRETS_AVAILABLE == 'true'
uses: zama-ai/slab-github-runner@5aee5d157f4a0201e5eaefc9cc648e5f9f5472a5 # v1.6.0
- name: Start instance
id: start-instance
uses: zama-ai/slab-github-runner@98f0788261a7323d5d695a883e20df36591a92b7
with:
mode: start
github-token: ${{ secrets.SLAB_ACTION_TOKEN }}
@@ -89,20 +84,13 @@ jobs:
backend: hyperstack
profile: gpu-test
# This instance will be spawned especially for pull-request from forked repository
- name: Start GitHub instance
id: start-github-instance
if: env.SECRETS_AVAILABLE == 'false'
run: |
echo "runner_group=${EXTERNAL_CONTRIBUTION_RUNNER}" >> "$GITHUB_OUTPUT"
cuda-signed-integer-tests:
name: gpu_signed_integer_tests/cuda-signed-integer-tests
name: CUDA signed integer tests
needs: [ should-run, setup-instance ]
if: github.event_name != 'pull_request' ||
(github.event_name == 'pull_request' && needs.setup-instance.result != 'skipped')
concurrency:
group: ${{ github.workflow_ref }}
group: ${{ github.workflow }}_${{ github.ref }}
cancel-in-progress: ${{ github.ref != 'refs/heads/main' }}
runs-on: ${{ needs.setup-instance.outputs.runner-name }}
strategy:
@@ -111,29 +99,30 @@ jobs:
matrix:
include:
- os: ubuntu-22.04
cuda: "12.8"
gcc: 11
cuda: "12.2"
gcc: 11
steps:
- name: Checkout tfhe-rs
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd
uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683
with:
persist-credentials: 'false'
token: ${{ env.CHECKOUT_TOKEN }}
token: ${{ secrets.FHE_ACTIONS_TOKEN }}
- name: Setup Hyperstack dependencies
uses: ./.github/actions/gpu_setup
uses: ./.github/actions/hyperstack_setup
with:
cuda-version: ${{ matrix.cuda }}
gcc-version: ${{ matrix.gcc }}
github-instance: ${{ env.SECRETS_AVAILABLE == 'false' }}
- name: Set up home
run: |
echo "HOME=/home/ubuntu" >> "${GITHUB_ENV}"
- name: Install latest stable
uses: dtolnay/rust-toolchain@e97e2d8cc328f1b50210efc529dca0028893a2d9 # zizmor: ignore[stale-action-refs] this action doesn't create releases
uses: dtolnay/rust-toolchain@315e265cd78dad1e1dcf3a5074f6d6c47029d5aa
with:
toolchain: stable
- name: Enable nvidia multi-process service
run: |
nvidia-cuda-mps-control -d
- name: Should run nightly tests
if: github.event_name == 'schedule'
run: |
@@ -147,37 +136,27 @@ jobs:
make test_signed_integer_multi_bit_gpu_ci
slack-notify:
name: gpu_signed_integer_tests/slack-notify
name: Slack Notification
needs: [ setup-instance, cuda-signed-integer-tests ]
runs-on: ubuntu-latest
if: ${{ always() && needs.cuda-signed-integer-tests.result != 'skipped' && failure() }}
continue-on-error: true
steps:
- name: Set pull-request URL
if: env.SECRETS_AVAILABLE == 'true' && github.event_name == 'pull_request'
run: |
echo "PULL_REQUEST_MD_LINK=[pull-request](${PR_BASE_URL}${PR_NUMBER}), " >> "${GITHUB_ENV}"
env:
PR_BASE_URL: ${{ vars.PR_BASE_URL }}
PR_NUMBER: ${{ github.event.pull_request.number }}
- name: Send message
if: env.SECRETS_AVAILABLE == 'true'
uses: rtCamp/action-slack-notify@e31e87e03dd19038e411e38ae27cbad084a90661
uses: rtCamp/action-slack-notify@c33737706dea87cd7784c687dadc9adf1be59990
env:
SLACK_COLOR: ${{ needs.cuda-signed-integer-tests.result }}
SLACK_MESSAGE: "Signed GPU tests finished with status: ${{ needs.cuda-signed-integer-tests.result }}. (${{ env.PULL_REQUEST_MD_LINK }}[action run](${{ env.ACTION_RUN_URL }}))"
SLACK_MESSAGE: "Base GPU tests finished with status: ${{ needs.cuda-signed-integer-tests.result }}. (${{ env.ACTION_RUN_URL }})"
teardown-instance:
name: gpu_signed_integer_tests/teardown-instance
if: ${{ always() && needs.setup-instance.result == 'success' }}
name: Teardown instance (cuda-tests)
if: ${{ always() && needs.setup-instance.result != 'skipped' }}
needs: [ setup-instance, cuda-signed-integer-tests ]
runs-on: ubuntu-latest
steps:
- name: Stop remote instance
- name: Stop instance
id: stop-instance
if: env.SECRETS_AVAILABLE == 'true'
uses: zama-ai/slab-github-runner@5aee5d157f4a0201e5eaefc9cc648e5f9f5472a5 # v1.6.0
uses: zama-ai/slab-github-runner@98f0788261a7323d5d695a883e20df36591a92b7
with:
mode: stop
github-token: ${{ secrets.SLAB_ACTION_TOKEN }}
@@ -187,7 +166,8 @@ jobs:
- name: Slack Notification
if: ${{ failure() }}
uses: rtCamp/action-slack-notify@e31e87e03dd19038e411e38ae27cbad084a90661
continue-on-error: true
uses: rtCamp/action-slack-notify@c33737706dea87cd7784c687dadc9adf1be59990
env:
SLACK_COLOR: ${{ job.status }}
SLACK_MESSAGE: "Instance teardown (cuda-signed-integer-tests) finished with status: ${{ job.status }}. (${{ env.ACTION_RUN_URL }})"

View File

@@ -1,5 +1,5 @@
# Test unsigned integers on an RTXA6000 VM on hyperstack with the classical PBS
name: gpu_unsigned_integer_classic_tests
name: TFHE Cuda Backend - Unsigned integer tests with classical PBS
env:
CARGO_TERM_COLOR: always
@@ -11,45 +11,33 @@ env:
SLACK_ICON: https://pbs.twimg.com/profile_images/1274014582265298945/OjBKP9kn_400x400.png
SLACK_USERNAME: ${{ secrets.BOT_USERNAME }}
SLACK_WEBHOOK: ${{ secrets.SLACK_WEBHOOK }}
SLACKIFY_MARKDOWN: true
IS_PULL_REQUEST: ${{ github.event_name == 'pull_request' }}
PULL_REQUEST_MD_LINK: ""
CHECKOUT_TOKEN: ${{ secrets.REPO_CHECKOUT_TOKEN || secrets.GITHUB_TOKEN }}
# Secrets will be available only to zama-ai organization members
SECRETS_AVAILABLE: ${{ secrets.JOB_SECRET != '' }}
EXTERNAL_CONTRIBUTION_RUNNER: "gpu_ubuntu-22.04"
on:
# Allows you to run this workflow manually from the Actions tab as an alternative.
workflow_dispatch:
pull_request:
types: [ labeled ]
permissions:
contents: read
# zizmor: ignore[concurrency-limits] concurrency is managed after instance setup to ensure safe provisioning
types: [ labeled ]
jobs:
should-run:
name: gpu_unsigned_integer_classic_tests/should-run
runs-on: ubuntu-latest
permissions:
pull-requests: read # Needed to check for file change
pull-requests: write
outputs:
gpu_test: ${{ env.IS_PULL_REQUEST == 'false' || steps.changed-files.outputs.gpu_any_changed }}
steps:
- name: Checkout tfhe-rs
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd
uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683
with:
fetch-depth: 0
persist-credentials: 'false'
token: ${{ env.CHECKOUT_TOKEN }}
token: ${{ secrets.FHE_ACTIONS_TOKEN }}
- name: Check for file changes
id: changed-files
uses: tj-actions/changed-files@9426d40962ed5378910ee2e21d5f8c6fcbf2dd96 # v47.0.6
uses: tj-actions/changed-files@bab30c2299617f6615ec02a68b9a40d10bd21366
with:
since_last_remote_commit: true
files_yaml: |
gpu:
- tfhe/Cargo.toml
@@ -63,23 +51,24 @@ jobs:
- tfhe/src/shortint/parameters/**
- tfhe/src/high_level_api/**
- tfhe/src/c_api/**
- 'tfhe/docs/**.md'
- '.github/workflows/gpu_unsigned_integer_classic_tests.yml'
- scripts/integer-tests.sh
- ci/slab.toml
setup-instance:
name: gpu_unsigned_integer_classic_tests/setup-instance
name: Setup instance (cuda-unsigned-classic-tests)
needs: should-run
if: github.event_name == 'workflow_dispatch' ||
if: github.event_name != 'pull_request' ||
(github.event.action != 'labeled' && needs.should-run.outputs.gpu_test == 'true') ||
(github.event.action == 'labeled' && github.event.label.name == 'approved' && needs.should-run.outputs.gpu_test == 'true')
runs-on: ubuntu-latest
outputs:
runner-name: ${{ steps.start-remote-instance.outputs.label || steps.start-github-instance.outputs.runner_group }}
runner-name: ${{ steps.start-instance.outputs.label }}
steps:
- name: Start remote instance
id: start-remote-instance
if: env.SECRETS_AVAILABLE == 'true'
uses: zama-ai/slab-github-runner@5aee5d157f4a0201e5eaefc9cc648e5f9f5472a5 # v1.6.0
- name: Start instance
id: start-instance
uses: zama-ai/slab-github-runner@98f0788261a7323d5d695a883e20df36591a92b7
with:
mode: start
github-token: ${{ secrets.SLAB_ACTION_TOKEN }}
@@ -88,20 +77,13 @@ jobs:
backend: hyperstack
profile: gpu-test
# This instance will be spawned especially for pull-request from forked repository
- name: Start GitHub instance
id: start-github-instance
if: env.SECRETS_AVAILABLE == 'false'
run: |
echo "runner_group=${EXTERNAL_CONTRIBUTION_RUNNER}" >> "$GITHUB_OUTPUT"
cuda-tests-linux:
name: gpu_unsigned_integer_classic_tests/cuda-tests-linux
name: CUDA unsigned integer tests with classical PBS
needs: [ should-run, setup-instance ]
if: github.event_name != 'pull_request' ||
(github.event_name == 'pull_request' && needs.setup-instance.result != 'skipped')
concurrency:
group: ${{ github.workflow_ref }}
group: ${{ github.workflow }}_${{ github.ref }}
cancel-in-progress: ${{ github.ref != 'refs/heads/main' }}
runs-on: ${{ needs.setup-instance.outputs.runner-name }}
strategy:
@@ -110,65 +92,53 @@ jobs:
matrix:
include:
- os: ubuntu-22.04
cuda: "12.8"
cuda: "12.2"
gcc: 11
steps:
- name: Checkout tfhe-rs
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd
with:
persist-credentials: 'false'
token: ${{ env.CHECKOUT_TOKEN }}
uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683
- name: Setup Hyperstack dependencies
uses: ./.github/actions/gpu_setup
uses: ./.github/actions/hyperstack_setup
with:
cuda-version: ${{ matrix.cuda }}
gcc-version: ${{ matrix.gcc }}
github-instance: ${{ env.SECRETS_AVAILABLE == 'false' }}
- name: Set up home
run: |
echo "HOME=/home/ubuntu" >> "${GITHUB_ENV}"
- name: Install latest stable
uses: dtolnay/rust-toolchain@e97e2d8cc328f1b50210efc529dca0028893a2d9 # zizmor: ignore[stale-action-refs] this action doesn't create releases
uses: dtolnay/rust-toolchain@315e265cd78dad1e1dcf3a5074f6d6c47029d5aa
with:
toolchain: stable
- name: Enable nvidia multi-process service
run: |
nvidia-cuda-mps-control -d
- name: Run unsigned integer tests
run: |
BIG_TESTS_INSTANCE=TRUE make test_unsigned_integer_gpu_ci
slack-notify:
name: gpu_unsigned_integer_classic_tests/slack-notify
name: Slack Notification
needs: [ setup-instance, cuda-tests-linux ]
runs-on: ubuntu-latest
if: ${{ always() && needs.cuda-tests-linux.result != 'skipped' && failure() }}
continue-on-error: true
steps:
- name: Set pull-request URL
if: env.SECRETS_AVAILABLE == 'true' && github.event_name == 'pull_request'
run: |
echo "PULL_REQUEST_MD_LINK=[pull-request](${PR_BASE_URL}${PR_NUMBER}), " >> "${GITHUB_ENV}"
env:
PR_BASE_URL: ${{ vars.PR_BASE_URL }}
PR_NUMBER: ${{ github.event.pull_request.number }}
- name: Send message
if: env.SECRETS_AVAILABLE == 'true'
uses: rtCamp/action-slack-notify@e31e87e03dd19038e411e38ae27cbad084a90661
uses: rtCamp/action-slack-notify@c33737706dea87cd7784c687dadc9adf1be59990
env:
SLACK_COLOR: ${{ needs.cuda-tests-linux.result }}
SLACK_MESSAGE: "Unsigned integer GPU classic tests finished with status: ${{ needs.cuda-tests-linux.result }}. (${{ env.PULL_REQUEST_MD_LINK }}[action run](${{ env.ACTION_RUN_URL }}))"
SLACK_MESSAGE: "Unsigned integer GPU classic tests finished with status: ${{ needs.cuda-tests-linux.result }}. (${{ env.ACTION_RUN_URL }})"
teardown-instance:
name: gpu_unsigned_integer_classic_tests/teardown-instance
if: ${{ always() && needs.setup-instance.result == 'success' }}
name: Teardown instance (cuda-unsigned-classic-tests)
if: ${{ always() && needs.setup-instance.result != 'skipped' }}
needs: [ setup-instance, cuda-tests-linux ]
runs-on: ubuntu-latest
steps:
- name: Stop remote instance
- name: Stop instance
id: stop-instance
if: env.SECRETS_AVAILABLE == 'true'
uses: zama-ai/slab-github-runner@5aee5d157f4a0201e5eaefc9cc648e5f9f5472a5 # v1.6.0
uses: zama-ai/slab-github-runner@98f0788261a7323d5d695a883e20df36591a92b7
with:
mode: stop
github-token: ${{ secrets.SLAB_ACTION_TOKEN }}
@@ -178,7 +148,8 @@ jobs:
- name: Slack Notification
if: ${{ failure() }}
uses: rtCamp/action-slack-notify@e31e87e03dd19038e411e38ae27cbad084a90661
continue-on-error: true
uses: rtCamp/action-slack-notify@c33737706dea87cd7784c687dadc9adf1be59990
env:
SLACK_COLOR: ${{ job.status }}
SLACK_MESSAGE: "Instance teardown (cuda-unsigned-classic-tests) finished with status: ${{ job.status }}. (${{ env.ACTION_RUN_URL }})"

View File

@@ -1,5 +1,5 @@
# Test unsigned integers on an H100 VM on hyperstack
name: gpu_unsigned_integer_h100_tests/
name: TFHE Cuda Backend - Unsigned integer tests on H100
env:
CARGO_TERM_COLOR: always
@@ -11,46 +11,33 @@ env:
SLACK_ICON: https://pbs.twimg.com/profile_images/1274014582265298945/OjBKP9kn_400x400.png
SLACK_USERNAME: ${{ secrets.BOT_USERNAME }}
SLACK_WEBHOOK: ${{ secrets.SLACK_WEBHOOK }}
SLACKIFY_MARKDOWN: true
IS_PULL_REQUEST: ${{ github.event_name == 'pull_request' }}
PULL_REQUEST_MD_LINK: ""
CHECKOUT_TOKEN: ${{ secrets.REPO_CHECKOUT_TOKEN || secrets.GITHUB_TOKEN }}
# Secrets will be available only to zama-ai organization members
SECRETS_AVAILABLE: ${{ secrets.JOB_SECRET != '' }}
EXTERNAL_CONTRIBUTION_RUNNER: "gpu_ubuntu-22.04"
on:
# Allows you to run this workflow manually from the Actions tab as an alternative.
workflow_dispatch:
pull_request:
types: [ labeled, opened, synchronize ]
permissions:
contents: read
# zizmor: ignore[concurrency-limits] concurrency is managed after instance setup to ensure safe provisioning
types: [ labeled ]
jobs:
should-run:
name: gpu_unsigned_integer_h100_tests/should-run
runs-on: ubuntu-latest
permissions:
pull-requests: read # Needed to check for file change
pull-requests: write
outputs:
gpu_test: ${{ env.IS_PULL_REQUEST == 'false' || steps.changed-files.outputs.gpu_any_changed }}
core_crypto_changed: ${{ steps.changed-files.outputs.core_crypto_any_changed }}
steps:
- name: Checkout tfhe-rs
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd
uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683
with:
fetch-depth: 0
persist-credentials: 'false'
token: ${{ env.CHECKOUT_TOKEN }}
token: ${{ secrets.FHE_ACTIONS_TOKEN }}
- name: Check for file changes
id: changed-files
uses: tj-actions/changed-files@9426d40962ed5378910ee2e21d5f8c6fcbf2dd96 # v47.0.6
uses: tj-actions/changed-files@bab30c2299617f6615ec02a68b9a40d10bd21366
with:
since_last_remote_commit: true
files_yaml: |
gpu:
- tfhe/Cargo.toml
@@ -64,25 +51,24 @@ jobs:
- tfhe/src/shortint/parameters/**
- tfhe/src/high_level_api/**
- tfhe/src/c_api/**
- 'tfhe/docs/**.md'
- '.github/workflows/gpu_unsigned_integer_h100_tests.yml'
- scripts/integer-tests.sh
core_crypto:
- tfhe/src/core_crypto/gpu/**
- ci/slab.toml
setup-instance:
name: gpu_unsigned_integer_h100_tests/setup-instance
name: Setup instance (cuda-h100-tests)
needs: should-run
if: github.event_name == 'workflow_dispatch' ||
(github.event.action == 'labeled' && github.event.label.name == 'approved' && needs.should-run.outputs.gpu_test == 'true') ||
(github.event.action != 'labeled' && needs.should-run.outputs.core_crypto_changed == 'true')
if: github.event_name != 'pull_request' ||
(github.event.action != 'labeled' && needs.should-run.outputs.gpu_test == 'true') ||
(github.event.action == 'labeled' && github.event.label.name == 'approved' && needs.should-run.outputs.gpu_test == 'true')
runs-on: ubuntu-latest
outputs:
runner-name: ${{ steps.start-remote-instance.outputs.label || steps.start-github-instance.outputs.runner_group }}
runner-name: ${{ steps.start-instance.outputs.label }}
steps:
- name: Start remote instance
id: start-remote-instance
if: env.SECRETS_AVAILABLE == 'true'
uses: zama-ai/slab-github-runner@5aee5d157f4a0201e5eaefc9cc648e5f9f5472a5 # v1.6.0
- name: Start instance
id: start-instance
uses: zama-ai/slab-github-runner@98f0788261a7323d5d695a883e20df36591a92b7
with:
mode: start
github-token: ${{ secrets.SLAB_ACTION_TOKEN }}
@@ -91,20 +77,13 @@ jobs:
backend: hyperstack
profile: single-h100
# This instance will be spawned especially for pull-request from forked repository
- name: Start GitHub instance
id: start-github-instance
if: env.SECRETS_AVAILABLE == 'false'
run: |
echo "runner_group=${EXTERNAL_CONTRIBUTION_RUNNER}" >> "$GITHUB_OUTPUT"
cuda-tests-linux:
name: gpu_unsigned_integer_h100_tests/cuda-tests-linux
name: CUDA H100 unsigned integer tests
needs: [ should-run, setup-instance ]
if: github.event_name != 'pull_request' ||
(github.event_name == 'pull_request' && needs.setup-instance.result != 'skipped')
concurrency:
group: ${{ github.workflow_ref }}
group: ${{ github.workflow }}_${{ github.ref }}
cancel-in-progress: ${{ github.ref != 'refs/heads/main' }}
runs-on: ${{ needs.setup-instance.outputs.runner-name }}
strategy:
@@ -113,65 +92,53 @@ jobs:
matrix:
include:
- os: ubuntu-22.04
cuda: "12.8"
cuda: "12.2"
gcc: 11
steps:
- name: Checkout tfhe-rs
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd
with:
persist-credentials: 'false'
token: ${{ env.CHECKOUT_TOKEN }}
uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683
- name: Setup Hyperstack dependencies
uses: ./.github/actions/gpu_setup
uses: ./.github/actions/hyperstack_setup
with:
cuda-version: ${{ matrix.cuda }}
gcc-version: ${{ matrix.gcc }}
github-instance: ${{ env.SECRETS_AVAILABLE == 'false' }}
- name: Set up home
run: |
echo "HOME=/home/ubuntu" >> "${GITHUB_ENV}"
- name: Install latest stable
uses: dtolnay/rust-toolchain@e97e2d8cc328f1b50210efc529dca0028893a2d9 # zizmor: ignore[stale-action-refs] this action doesn't create releases
uses: dtolnay/rust-toolchain@315e265cd78dad1e1dcf3a5074f6d6c47029d5aa
with:
toolchain: stable
- name: Enable nvidia multi-process service
run: |
nvidia-cuda-mps-control -d
- name: Run unsigned integer multi-bit tests
run: |
BIG_TESTS_INSTANCE=TRUE make test_unsigned_integer_multi_bit_gpu_ci
slack-notify:
name: gpu_unsigned_integer_h100_tests/slack-notify
name: Slack Notification
needs: [ setup-instance, cuda-tests-linux ]
runs-on: ubuntu-latest
if: ${{ always() && needs.cuda-tests-linux.result != 'skipped' && failure() }}
continue-on-error: true
steps:
- name: Set pull-request URL
if: env.SECRETS_AVAILABLE == 'true' && github.event_name == 'pull_request'
run: |
echo "PULL_REQUEST_MD_LINK=[pull-request](${PR_BASE_URL}${PR_NUMBER}), " >> "${GITHUB_ENV}"
env:
PR_BASE_URL: ${{ vars.PR_BASE_URL }}
PR_NUMBER: ${{ github.event.pull_request.number }}
- name: Send message
if: env.SECRETS_AVAILABLE == 'true'
uses: rtCamp/action-slack-notify@e31e87e03dd19038e411e38ae27cbad084a90661
uses: rtCamp/action-slack-notify@c33737706dea87cd7784c687dadc9adf1be59990
env:
SLACK_COLOR: ${{ needs.cuda-tests-linux.result }}
SLACK_MESSAGE: "Unsigned integer GPU H100 tests finished with status: ${{ needs.cuda-tests-linux.result }}. (${{ env.PULL_REQUEST_MD_LINK }}[action run](${{ env.ACTION_RUN_URL }}))"
SLACK_MESSAGE: "Unsigned integer GPU H100 tests finished with status: ${{ needs.cuda-tests-linux.result }}. (${{ env.ACTION_RUN_URL }})"
teardown-instance:
name: gpu_unsigned_integer_h100_tests/teardown-instance
if: ${{ always() && needs.setup-instance.result == 'success' }}
name: Teardown instance (cuda-h100-tests)
if: ${{ always() && needs.setup-instance.result != 'skipped' }}
needs: [ setup-instance, cuda-tests-linux ]
runs-on: ubuntu-latest
steps:
- name: Stop remote instance
- name: Stop instance
id: stop-instance
if: env.SECRETS_AVAILABLE == 'true'
uses: zama-ai/slab-github-runner@5aee5d157f4a0201e5eaefc9cc648e5f9f5472a5 # v1.6.0
uses: zama-ai/slab-github-runner@98f0788261a7323d5d695a883e20df36591a92b7
with:
mode: stop
github-token: ${{ secrets.SLAB_ACTION_TOKEN }}
@@ -181,7 +148,8 @@ jobs:
- name: Slack Notification
if: ${{ failure() }}
uses: rtCamp/action-slack-notify@e31e87e03dd19038e411e38ae27cbad084a90661
continue-on-error: true
uses: rtCamp/action-slack-notify@c33737706dea87cd7784c687dadc9adf1be59990
env:
SLACK_COLOR: ${{ job.status }}
SLACK_MESSAGE: "Instance teardown (cuda-h100-tests) finished with status: ${{ job.status }}. (${{ env.ACTION_RUN_URL }})"

View File

@@ -1,5 +1,5 @@
# Compile and test tfhe-cuda-backend unsigned integer on an AWS instance
name: gpu_unsigned_integer_tests
name: TFHE Cuda Backend - Unsigned integer tests
env:
CARGO_TERM_COLOR: always
@@ -11,46 +11,39 @@ env:
SLACK_ICON: https://pbs.twimg.com/profile_images/1274014582265298945/OjBKP9kn_400x400.png
SLACK_USERNAME: ${{ secrets.BOT_USERNAME }}
SLACK_WEBHOOK: ${{ secrets.SLACK_WEBHOOK }}
SLACKIFY_MARKDOWN: true
FAST_TESTS: TRUE
NIGHTLY_TESTS: FALSE
IS_PULL_REQUEST: ${{ github.event_name == 'pull_request' }}
PULL_REQUEST_MD_LINK: ""
CHECKOUT_TOKEN: ${{ secrets.REPO_CHECKOUT_TOKEN || secrets.GITHUB_TOKEN }}
# Secrets will be available only to zama-ai organization members
SECRETS_AVAILABLE: ${{ secrets.JOB_SECRET != '' }}
EXTERNAL_CONTRIBUTION_RUNNER: "gpu_ubuntu-22.04"
on:
# Allows you to run this workflow manually from the Actions tab as an alternative.
workflow_dispatch:
pull_request:
permissions:
contents: read
# zizmor: ignore[concurrency-limits] concurrency is managed after instance setup to ensure safe provisioning
types:
- opened
- synchronize
schedule:
# Nightly tests @ 1AM after each work day
- cron: "0 1 * * MON-FRI"
jobs:
should-run:
name: gpu_unsigned_integer_tests/should-run
runs-on: ubuntu-latest
permissions:
pull-requests: read # Needed to check for file change
pull-requests: write
outputs:
gpu_test: ${{ env.IS_PULL_REQUEST == 'false' || steps.changed-files.outputs.gpu_any_changed }}
steps:
- name: Checkout tfhe-rs
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd
uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683
with:
fetch-depth: 0
persist-credentials: 'false'
token: ${{ env.CHECKOUT_TOKEN }}
token: ${{ secrets.FHE_ACTIONS_TOKEN }}
- name: Check for file changes
id: changed-files
uses: tj-actions/changed-files@9426d40962ed5378910ee2e21d5f8c6fcbf2dd96 # v47.0.6
uses: tj-actions/changed-files@bab30c2299617f6615ec02a68b9a40d10bd21366
with:
since_last_remote_commit: true
files_yaml: |
gpu:
- tfhe/Cargo.toml
@@ -64,23 +57,24 @@ jobs:
- tfhe/src/shortint/parameters/**
- tfhe/src/high_level_api/**
- tfhe/src/c_api/**
- 'tfhe/docs/**.md'
- '.github/workflows/gpu_unsigned_integer_tests.yml'
- scripts/integer-tests.sh
- ci/slab.toml
setup-instance:
name: gpu_unsigned_integer_tests/setup-instance
runs-on: ubuntu-latest
name: Setup instance (cuda-unsigned-integer-tests)
needs: should-run
if: (github.event_name == 'schedule' && github.repository == 'zama-ai/tfhe-rs') ||
github.event_name == 'workflow_dispatch' ||
needs.should-run.outputs.gpu_test == 'true'
runs-on: ubuntu-latest
outputs:
runner-name: ${{ steps.start-remote-instance.outputs.label || steps.start-github-instance.outputs.runner_group }}
runner-name: ${{ steps.start-instance.outputs.label }}
steps:
- name: Start remote instance
id: start-remote-instance
if: env.SECRETS_AVAILABLE == 'true'
uses: zama-ai/slab-github-runner@5aee5d157f4a0201e5eaefc9cc648e5f9f5472a5 # v1.6.0
- name: Start instance
id: start-instance
uses: zama-ai/slab-github-runner@98f0788261a7323d5d695a883e20df36591a92b7
with:
mode: start
github-token: ${{ secrets.SLAB_ACTION_TOKEN }}
@@ -89,20 +83,13 @@ jobs:
backend: hyperstack
profile: gpu-test
# This instance will be spawned especially for pull-request from forked repository
- name: Start GitHub instance
id: start-github-instance
if: env.SECRETS_AVAILABLE == 'false'
run: |
echo "runner_group=${EXTERNAL_CONTRIBUTION_RUNNER}" >> "$GITHUB_OUTPUT"
cuda-unsigned-integer-tests:
name: gpu_unsigned_integer_tests/cuda-unsigned-integer-tests
name: CUDA unsigned integer tests
needs: [ should-run, setup-instance ]
if: github.event_name != 'pull_request' ||
(github.event_name == 'pull_request' && needs.setup-instance.result != 'skipped')
concurrency:
group: ${{ github.workflow_ref }}
group: ${{ github.workflow }}_${{ github.ref }}
cancel-in-progress: ${{ github.ref != 'refs/heads/main' }}
runs-on: ${{ needs.setup-instance.outputs.runner-name }}
strategy:
@@ -111,29 +98,27 @@ jobs:
matrix:
include:
- os: ubuntu-22.04
cuda: "12.8"
cuda: "12.2"
gcc: 11
steps:
- name: Checkout tfhe-rs
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd
with:
persist-credentials: 'false'
token: ${{ env.CHECKOUT_TOKEN }}
uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683
- name: Setup Hyperstack dependencies
uses: ./.github/actions/gpu_setup
uses: ./.github/actions/hyperstack_setup
with:
cuda-version: ${{ matrix.cuda }}
gcc-version: ${{ matrix.gcc }}
github-instance: ${{ env.SECRETS_AVAILABLE == 'false' }}
- name: Set up home
run: |
echo "HOME=/home/ubuntu" >> "${GITHUB_ENV}"
- name: Install latest stable
uses: dtolnay/rust-toolchain@e97e2d8cc328f1b50210efc529dca0028893a2d9 # zizmor: ignore[stale-action-refs] this action doesn't create releases
uses: dtolnay/rust-toolchain@315e265cd78dad1e1dcf3a5074f6d6c47029d5aa
with:
toolchain: stable
- name: Enable nvidia multi-process service
run: |
nvidia-cuda-mps-control -d
- name: Should run nightly tests
if: github.event_name == 'schedule'
run: |
@@ -147,37 +132,27 @@ jobs:
make test_unsigned_integer_multi_bit_gpu_ci
slack-notify:
name: gpu_unsigned_integer_tests/slack-notify
name: Slack Notification
needs: [ setup-instance, cuda-unsigned-integer-tests ]
runs-on: ubuntu-latest
if: ${{ always() && needs.cuda-unsigned-integer-tests.result != 'skipped' && failure() }}
continue-on-error: true
steps:
- name: Set pull-request URL
if: env.SECRETS_AVAILABLE == 'true' && github.event_name == 'pull_request'
run: |
echo "PULL_REQUEST_MD_LINK=[pull-request](${PR_BASE_URL}${PR_NUMBER}), " >> "${GITHUB_ENV}"
env:
PR_BASE_URL: ${{ vars.PR_BASE_URL }}
PR_NUMBER: ${{ github.event.pull_request.number }}
- name: Send message
if: env.SECRETS_AVAILABLE == 'true'
uses: rtCamp/action-slack-notify@e31e87e03dd19038e411e38ae27cbad084a90661
uses: rtCamp/action-slack-notify@c33737706dea87cd7784c687dadc9adf1be59990
env:
SLACK_COLOR: ${{ needs.cuda-unsigned-integer-tests.result }}
SLACK_MESSAGE: "Unsigned integer GPU tests finished with status: ${{ needs.cuda-unsigned-integer-tests.result }}. (${{ env.PULL_REQUEST_MD_LINK }}[action run](${{ env.ACTION_RUN_URL }}))"
SLACK_MESSAGE: "Unsigned integer GPU tests finished with status: ${{ needs.cuda-unsigned-integer-tests.result }}. (${{ env.ACTION_RUN_URL }})"
teardown-instance:
name: gpu_unsigned_integer_tests/teardown-instance
if: ${{ always() && needs.setup-instance.result == 'success' }}
name: Teardown instance (cuda-tests)
if: ${{ always() && needs.setup-instance.result != 'skipped' }}
needs: [ setup-instance, cuda-unsigned-integer-tests ]
runs-on: ubuntu-latest
steps:
- name: Stop instance
id: stop-instance
if: env.SECRETS_AVAILABLE == 'true'
uses: zama-ai/slab-github-runner@5aee5d157f4a0201e5eaefc9cc648e5f9f5472a5 # v1.6.0
uses: zama-ai/slab-github-runner@98f0788261a7323d5d695a883e20df36591a92b7
with:
mode: stop
github-token: ${{ secrets.SLAB_ACTION_TOKEN }}
@@ -187,7 +162,8 @@ jobs:
- name: Slack Notification
if: ${{ failure() }}
uses: rtCamp/action-slack-notify@e31e87e03dd19038e411e38ae27cbad084a90661
continue-on-error: true
uses: rtCamp/action-slack-notify@c33737706dea87cd7784c687dadc9adf1be59990
env:
SLACK_COLOR: ${{ job.status }}
SLACK_MESSAGE: "Instance teardown (cuda-unsigned-integer-tests) finished with status: ${{ job.status }}. (${{ env.ACTION_RUN_URL }})"

View File

@@ -1,183 +0,0 @@
# Compile and test zk-cuda-backend
name: gpu_zk_tests
env:
CARGO_TERM_COLOR: always
ACTION_RUN_URL: ${{ github.server_url }}/${{ github.repository }}/actions/runs/${{ github.run_id }}
RUSTFLAGS: "-C target-cpu=native"
RUST_BACKTRACE: "full"
RUST_MIN_STACK: "8388608"
SLACK_CHANNEL: ${{ secrets.SLACK_CHANNEL }}
SLACK_ICON: https://pbs.twimg.com/profile_images/1274014582265298945/OjBKP9kn_400x400.png
SLACK_USERNAME: ${{ secrets.BOT_USERNAME }}
SLACK_WEBHOOK: ${{ secrets.SLACK_WEBHOOK }}
SLACKIFY_MARKDOWN: true
IS_PULL_REQUEST: ${{ github.event_name == 'pull_request' }}
PULL_REQUEST_MD_LINK: ""
CHECKOUT_TOKEN: ${{ secrets.REPO_CHECKOUT_TOKEN || secrets.GITHUB_TOKEN }}
# Secrets will be available only to zama-ai organization members
SECRETS_AVAILABLE: ${{ secrets.JOB_SECRET != '' }}
EXTERNAL_CONTRIBUTION_RUNNER: "gpu_ubuntu-22.04"
on:
# Allows you to run this workflow manually from the Actions tab as an alternative.
workflow_dispatch:
pull_request:
permissions:
contents: read
# zizmor: ignore[concurrency-limits] concurrency is managed after instance setup to ensure safe provisioning
jobs:
should-run:
name: gpu_zk_tests/should-run
runs-on: ubuntu-latest
permissions:
pull-requests: read # Needed to check for file change
outputs:
gpu_test: ${{ env.IS_PULL_REQUEST == 'false' || steps.changed-files.outputs.gpu_any_changed }}
steps:
- name: Checkout tfhe-rs
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd
with:
fetch-depth: 0
persist-credentials: 'false'
token: ${{ env.CHECKOUT_TOKEN }}
- name: Check for file changes
id: changed-files
uses: tj-actions/changed-files@9426d40962ed5378910ee2e21d5f8c6fcbf2dd96 # v47.0.6
with:
files_yaml: |
gpu:
- tfhe/Cargo.toml
- tfhe/build.rs
- backends/tfhe-cuda-backend/**
- backends/zk-cuda-backend/**
- tfhe/src/shortint/parameters/**
- tfhe/src/zk/**
- tfhe-zk-pok/**
- '.github/workflows/gpu_zk_tests.yml'
- ci/slab.toml
setup-instance:
name: gpu_zk_tests/setup-instance
needs: should-run
if: github.event_name == 'workflow_dispatch' ||
needs.should-run.outputs.gpu_test == 'true'
runs-on: ubuntu-latest
outputs:
runner-name: ${{ steps.start-remote-instance.outputs.label || steps.start-github-instance.outputs.runner_group }}
steps:
- name: Start remote instance
id: start-remote-instance
if: env.SECRETS_AVAILABLE == 'true'
uses: zama-ai/slab-github-runner@5aee5d157f4a0201e5eaefc9cc648e5f9f5472a5 # v1.6.0
with:
mode: start
github-token: ${{ secrets.SLAB_ACTION_TOKEN }}
slab-url: ${{ secrets.SLAB_BASE_URL }}
job-secret: ${{ secrets.JOB_SECRET }}
backend: hyperstack
profile: gpu-test
# This instance will be spawned especially for pull-request from forked repository
- name: Start GitHub instance
id: start-github-instance
if: env.SECRETS_AVAILABLE == 'false'
run: |
echo "runner_group=${EXTERNAL_CONTRIBUTION_RUNNER}" >> "$GITHUB_OUTPUT"
cuda-tests-linux:
name: gpu_zk_tests/cuda-tests-linux
needs: [ should-run, setup-instance ]
if: github.event_name != 'pull_request' ||
(github.event_name == 'pull_request' && needs.setup-instance.result != 'skipped')
concurrency:
group: ${{ github.workflow_ref }}
cancel-in-progress: ${{ github.ref != 'refs/heads/main' }}
runs-on: ${{ needs.setup-instance.outputs.runner-name }}
strategy:
fail-fast: false
# explicit include-based build matrix, of known valid options
matrix:
include:
- os: ubuntu-22.04
cuda: "12.8"
gcc: 11
steps:
- name: Checkout tfhe-rs
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd
with:
persist-credentials: 'false'
token: ${{ env.CHECKOUT_TOKEN }}
- name: Setup Hyperstack dependencies
uses: ./.github/actions/gpu_setup
with:
cuda-version: ${{ matrix.cuda }}
gcc-version: ${{ matrix.gcc }}
github-instance: ${{ env.SECRETS_AVAILABLE == 'false' }}
- name: Install latest stable
uses: dtolnay/rust-toolchain@e97e2d8cc328f1b50210efc529dca0028893a2d9 # zizmor: ignore[stale-action-refs] this action doesn't create releases
with:
toolchain: stable
- name: Enable nvidia multi-process service
run: |
nvidia-cuda-mps-control -d
- name: Run zk-cuda-backend integration tests
run: |
make test_zk_cuda_backend
make test_zk_pok_experimental_gpu
make test_integer_zk_gpu
make test_integer_zk_experimental_gpu
slack-notify:
name: gpu_zk_tests/slack-notify
needs: [ setup-instance, cuda-tests-linux ]
runs-on: ubuntu-latest
if: ${{ always() && needs.cuda-tests-linux.result != 'skipped' && failure() }}
continue-on-error: true
steps:
- name: Set pull-request URL
if: env.SECRETS_AVAILABLE == 'true' && github.event_name == 'pull_request'
run: |
echo "PULL_REQUEST_MD_LINK=[pull-request](${PR_BASE_URL}${PR_NUMBER}), " >> "${GITHUB_ENV}"
env:
PR_BASE_URL: ${{ vars.PR_BASE_URL }}
PR_NUMBER: ${{ github.event.pull_request.number }}
- name: Send message
if: env.SECRETS_AVAILABLE == 'true'
uses: rtCamp/action-slack-notify@e31e87e03dd19038e411e38ae27cbad084a90661
env:
SLACK_COLOR: ${{ needs.cuda-tests-linux.result }}
SLACK_MESSAGE: "ZK GPU tests finished with status: ${{ needs.cuda-tests-linux.result }}. (${{ env.PULL_REQUEST_MD_LINK }}[action run](${{ env.ACTION_RUN_URL }}))"
teardown-instance:
name: gpu_zk_tests/teardown-instance
if: ${{ always() && needs.setup-instance.result == 'success' }}
needs: [ setup-instance, cuda-tests-linux ]
runs-on: ubuntu-latest
steps:
- name: Stop remote instance
id: stop-instance
if: env.SECRETS_AVAILABLE == 'true'
uses: zama-ai/slab-github-runner@5aee5d157f4a0201e5eaefc9cc648e5f9f5472a5 # v1.6.0
with:
mode: stop
github-token: ${{ secrets.SLAB_ACTION_TOKEN }}
slab-url: ${{ secrets.SLAB_BASE_URL }}
job-secret: ${{ secrets.JOB_SECRET }}
label: ${{ needs.setup-instance.outputs.runner-name }}
- name: Slack Notification
if: ${{ failure() }}
uses: rtCamp/action-slack-notify@e31e87e03dd19038e411e38ae27cbad084a90661
env:
SLACK_COLOR: ${{ job.status }}
SLACK_MESSAGE: "Instance teardown (cuda-tests) finished with status: ${{ job.status }}. (${{ env.ACTION_RUN_URL }})"

View File

@@ -1,81 +0,0 @@
# Test HPU backend HLAPI layer
name: hpu_hlapi_tests
on:
workflow_dispatch:
pull_request:
push:
branches:
- main
env:
CARGO_TERM_COLOR: always
ACTION_RUN_URL: ${{ github.server_url }}/${{ github.repository }}/actions/runs/${{ github.run_id }}
IS_PULL_REQUEST: ${{ github.event_name == 'pull_request' }}
CHECKOUT_TOKEN: ${{ secrets.REPO_CHECKOUT_TOKEN || secrets.GITHUB_TOKEN }}
concurrency:
group: ${{ github.workflow }}-${{ github.head_ref }}${{ github.ref == 'refs/heads/main' && github.sha || '' }}
cancel-in-progress: true
permissions: {}
jobs:
should-run:
name: hpu_hlapi_tests/should-run
if: github.event_name != 'push' ||
(github.event_name == 'push' && github.repository == 'zama-ai/tfhe-rs')
runs-on: ubuntu-latest
permissions:
pull-requests: read # Needed to check for file change
outputs:
hpu_test: ${{ env.IS_PULL_REQUEST == 'false' || steps.changed-files.outputs.hpu_any_changed }}
steps:
- name: Checkout tfhe-rs
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd
with:
fetch-depth: 0
persist-credentials: 'false'
token: ${{ env.CHECKOUT_TOKEN }}
- name: Check for file changes
id: changed-files
uses: tj-actions/changed-files@9426d40962ed5378910ee2e21d5f8c6fcbf2dd96 # v47.0.6
with:
files_yaml: |
hpu:
- tfhe/Cargo.toml
- Makefile
- backends/tfhe-hpu-backend/**
- mockups/tfhe-hpu-mockup/**
cargo-tests-hpu:
name: hpu_hlapi_tests/cargo-tests-hpu (bpr)
needs: should-run
if:
needs.should-run.outputs.hpu_test == 'true' &&
((github.event_name == 'push' && github.repository == 'zama-ai/tfhe-rs') ||github.event_name == 'pull_request')
runs-on: "runs-on=${{ github.run_id }}/runner=cpu-big"
steps:
- uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd
with:
persist-credentials: 'false'
token: ${{ env.CHECKOUT_TOKEN }}
- name: Install Rust
uses: dtolnay/rust-toolchain@e97e2d8cc328f1b50210efc529dca0028893a2d9 # zizmor: ignore[stale-action-refs] this action doesn't create releases
with:
toolchain: stable
override: true
- name: Install Just
run: |
cargo install just
- name: Test HLAPI HPU
run: |
source setup_hpu.sh
just -f mockups/tfhe-hpu-mockup/Justfile BUILD_PROFILE=release mockup &
make HPU_CONFIG=sim test_high_level_api_hpu
make HPU_CONFIG=sim test_user_doc_hpu

View File

@@ -1,4 +1,4 @@
name: integer_long_run_tests
name: AWS Long Run Tests on CPU
env:
CARGO_TERM_COLOR: always
@@ -15,33 +15,45 @@ on:
# Allows you to run this workflow manually from the Actions tab as an alternative.
workflow_dispatch:
schedule:
# Weekly tests will be triggered each Friday at 9p.m.
- cron: "0 21 * * 5"
permissions: {}
# zizmor: ignore[concurrency-limits] concurrency is managed after instance setup to ensure safe provisioning
# Weekly tests will be triggered each Friday at 1a.m.
- cron: '0 1 * * FRI'
jobs:
cpu-tests:
name: integer_long_run_tests/cpu-tests
setup-instance:
name: Setup instance (cpu-tests)
if: github.event_name != 'schedule' ||
(github.event_name == 'schedule' && github.repository == 'zama-ai/tfhe-rs')
runs-on: ubuntu-latest
outputs:
runner-name: ${{ steps.start-instance.outputs.label }}
steps:
- name: Start instance
id: start-instance
uses: zama-ai/slab-github-runner@98f0788261a7323d5d695a883e20df36591a92b7
with:
mode: start
github-token: ${{ secrets.SLAB_ACTION_TOKEN }}
slab-url: ${{ secrets.SLAB_BASE_URL }}
job-secret: ${{ secrets.JOB_SECRET }}
backend: aws
profile: cpu-big
cpu-tests:
name: Long run CPU tests
needs: [ setup-instance ]
concurrency:
group: ${{ github.workflow_ref }}_${{github.event_name}}
group: ${{ github.workflow }}_${{github.event_name}}_${{ github.ref }}
cancel-in-progress: true
runs-on: "runs-on=${{ github.run_id }}/runner=cpu-big"
timeout-minutes: 4320 # 72 hours
runs-on: ${{ needs.setup-instance.outputs.runner-name }}
steps:
- name: Checkout tfhe-rs
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd
uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683
with:
persist-credentials: 'false'
token: ${{ secrets.REPO_CHECKOUT_TOKEN }}
token: ${{ secrets.FHE_ACTIONS_TOKEN }}
- name: Install latest stable
uses: dtolnay/rust-toolchain@e97e2d8cc328f1b50210efc529dca0028893a2d9 # zizmor: ignore[stale-action-refs] this action doesn't create releases
uses: dtolnay/rust-toolchain@315e265cd78dad1e1dcf3a5074f6d6c47029d5aa
with:
toolchain: stable
@@ -50,9 +62,33 @@ jobs:
make test_integer_long_run
- name: Slack Notification
if: ${{ failure() || (cancelled() && github.event_name != 'pull_request') }}
if: ${{ failure() }}
continue-on-error: true
uses: rtCamp/action-slack-notify@e31e87e03dd19038e411e38ae27cbad084a90661
uses: rtCamp/action-slack-notify@c33737706dea87cd7784c687dadc9adf1be59990
env:
SLACK_COLOR: ${{ job.status }}
SLACK_MESSAGE: "CPU long run tests finished with status: ${{ job.status }}. (${{ env.ACTION_RUN_URL }})"
teardown-instance:
name: Teardown instance (cpu-tests)
if: ${{ always() && needs.setup-instance.result != 'skipped' }}
needs: [ setup-instance, cpu-tests ]
runs-on: ubuntu-latest
steps:
- name: Stop instance
id: stop-instance
uses: zama-ai/slab-github-runner@98f0788261a7323d5d695a883e20df36591a92b7
with:
mode: stop
github-token: ${{ secrets.SLAB_ACTION_TOKEN }}
slab-url: ${{ secrets.SLAB_BASE_URL }}
job-secret: ${{ secrets.JOB_SECRET }}
label: ${{ needs.setup-instance.outputs.runner-name }}
- name: Slack Notification
if: ${{ failure() }}
continue-on-error: true
uses: rtCamp/action-slack-notify@c33737706dea87cd7784c687dadc9adf1be59990
env:
SLACK_COLOR: ${{ job.status }}
SLACK_MESSAGE: "Instance teardown (cpu-long-run-tests) finished with status: ${{ job.status }}. (${{ env.ACTION_RUN_URL }})"

View File

@@ -1,9 +1,9 @@
name: m1_tests
name: Tests on M1 CPU
on:
workflow_dispatch:
pull_request:
types: [ labeled ]
types: [labeled]
# Have a nightly build for M1 tests
schedule:
# * is a special character in YAML so you have to quote this string
@@ -21,33 +21,25 @@ env:
# We clear the cache to reduce memory pressure because of the numerous processes of cargo
# nextest
TFHE_RS_CLEAR_IN_MEMORY_KEY_CACHE: "1"
CHECKOUT_TOKEN: ${{ secrets.REPO_CHECKOUT_TOKEN || secrets.GITHUB_TOKEN }}
concurrency:
group: ${{ github.workflow_ref }}
group: ${{ github.workflow }}-${{ github.head_ref }}
cancel-in-progress: true
permissions:
contents: read
jobs:
cargo-builds-m1:
name: m1_tests/cargo-builds-m1
if: ${{ (github.event_name == 'schedule' && github.repository == 'zama-ai/tfhe-rs') ||
github.event_name == 'workflow_dispatch' ||
contains(github.event.label.name, 'm1_test') }}
if: ${{ (github.event_name == 'schedule' && github.repository == 'zama-ai/tfhe-rs') || github.event_name == 'workflow_dispatch' || contains(github.event.label.name, 'm1_test') }}
runs-on: ["self-hosted", "m1mac"]
# 12 hours, default is 6 hours, hopefully this is more than enough
timeout-minutes: 720
steps:
- uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd
- uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683
with:
persist-credentials: "false"
token: ${{ env.CHECKOUT_TOKEN }}
- name: Install latest stable
uses: dtolnay/rust-toolchain@e97e2d8cc328f1b50210efc529dca0028893a2d9 # zizmor: ignore[stale-action-refs] this action doesn't create releases
uses: dtolnay/rust-toolchain@315e265cd78dad1e1dcf3a5074f6d6c47029d5aa
with:
toolchain: stable
@@ -67,7 +59,9 @@ jobs:
run: |
make test_fft
make test_fft_serde
make test_fft_nightly
make test_fft_no_std
make test_fft_no_std_nightly
# we don't run the js stuff here as it's causing issues with the M1 config
- name: Run pcc NTT checks
@@ -177,7 +171,7 @@ jobs:
make test_integer_multi_bit_ci
remove_label:
name: m1_tests/remove_label
name: Remove m1_test label
runs-on: ubuntu-latest
needs:
- cargo-builds-m1
@@ -191,7 +185,8 @@ jobs:
- name: Slack Notification
if: ${{ needs.cargo-builds-m1.result != 'skipped' }}
uses: rtCamp/action-slack-notify@e31e87e03dd19038e411e38ae27cbad084a90661
continue-on-error: true
uses: rtCamp/action-slack-notify@c33737706dea87cd7784c687dadc9adf1be59990
env:
SLACK_COLOR: ${{ needs.cargo-builds-m1.result }}
SLACK_CHANNEL: ${{ secrets.SLACK_CHANNEL }}
@@ -199,5 +194,3 @@ jobs:
SLACK_MESSAGE: "M1 tests finished with status: ${{ needs.cargo-builds-m1.result }}. (${{ env.ACTION_RUN_URL }})"
SLACK_USERNAME: ${{ secrets.BOT_USERNAME }}
SLACK_WEBHOOK: ${{ secrets.SLACK_WEBHOOK }}
MSG_MINIMAL: event,action url,commit
BRANCH: ${{ github.ref }}

165
.github/workflows/make_release.yml vendored Normal file
View File

@@ -0,0 +1,165 @@
# Publish new release of tfhe-rs on various platform.
name: Publish release
on:
workflow_dispatch:
inputs:
dry_run:
description: "Dry-run"
type: boolean
default: true
push_to_crates:
description: "Push to crate"
type: boolean
default: true
push_web_package:
description: "Push web js package"
type: boolean
default: true
push_node_package:
description: "Push node js package"
type: boolean
default: true
npm_latest_tag:
description: "Set NPM tag as latest"
type: boolean
default: false
env:
ACTION_RUN_URL: ${{ github.server_url }}/${{ github.repository }}/actions/runs/${{ github.run_id }}
NPM_TAG: ""
jobs:
verify_tag:
uses: ./.github/workflows/verify_tagged_commit.yml
secrets:
RELEASE_TEAM: ${{ secrets.RELEASE_TEAM }}
READ_ORG_TOKEN: ${{ secrets.READ_ORG_TOKEN }}
package:
runs-on: ubuntu-latest
needs: verify_tag
outputs:
hash: ${{ steps.hash.outputs.hash }}
steps:
- name: Checkout
uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683
with:
fetch-depth: 0
token: ${{ secrets.FHE_ACTIONS_TOKEN }}
- name: Prepare package
run: |
cargo package -p tfhe
- uses: actions/upload-artifact@b4b15b8c7c6ac21ea08fcf65892d2ee8f75cf882 # v4.4.3
with:
name: crate
path: target/package/*.crate
- name: generate hash
id: hash
run: cd target/package && echo "hash=$(sha256sum ./*.crate | base64 -w0)" >> "${GITHUB_OUTPUT}"
provenance:
if: ${{ !inputs.dry_run }}
needs: [package]
uses: slsa-framework/slsa-github-generator/.github/workflows/generator_generic_slsa3.yml@v2.0.0
permissions:
# Needed to detect the GitHub Actions environment
actions: read
# Needed to create the provenance via GitHub OIDC
id-token: write
# Needed to upload assets/artifacts
contents: write
with:
# SHA-256 hashes of the Crate package.
base64-subjects: ${{ needs.package.outputs.hash }}
publish_release:
name: Publish Release
needs: [package] # for comparing hashes
runs-on: ubuntu-latest
permissions:
contents: read
id-token: write
steps:
- name: Checkout
uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683
with:
fetch-depth: 0
token: ${{ secrets.FHE_ACTIONS_TOKEN }}
- name: Create NPM version tag
if: ${{ inputs.npm_latest_tag }}
run: |
echo "NPM_TAG=latest" >> "${GITHUB_ENV}"
- name: Download artifact
uses: actions/download-artifact@fa0a91b85d4f404e444e00e005971372dc801d16 # v4.1.8
with:
name: crate
path: target/package
- name: Publish crate.io package
if: ${{ inputs.push_to_crates }}
env:
CRATES_TOKEN: ${{ secrets.CARGO_REGISTRY_TOKEN }}
DRY_RUN: ${{ inputs.dry_run && '--dry-run' || '' }}
run: |
cargo publish -p tfhe --token ${{ env.CRATES_TOKEN }} ${{ env.DRY_RUN }}
- name: Generate hash
id: published_hash
run: cd target/package && echo "pub_hash=$(sha256sum ./*.crate | base64 -w0)" >> "${GITHUB_OUTPUT}"
- name: Slack notification (hashes comparison)
if: ${{ needs.package.outputs.hash != steps.published_hash.outputs.pub_hash }}
continue-on-error: true
uses: rtCamp/action-slack-notify@c33737706dea87cd7784c687dadc9adf1be59990
env:
SLACK_COLOR: failure
SLACK_CHANNEL: ${{ secrets.SLACK_CHANNEL }}
SLACK_ICON: https://pbs.twimg.com/profile_images/1274014582265298945/OjBKP9kn_400x400.png
SLACK_MESSAGE: "SLSA tfhe crate - hash comparison failure: (${{ env.ACTION_RUN_URL }})"
SLACK_USERNAME: ${{ secrets.BOT_USERNAME }}
SLACK_WEBHOOK: ${{ secrets.SLACK_WEBHOOK }}
- name: Build web package
if: ${{ inputs.push_web_package }}
run: |
make build_web_js_api_parallel
- name: Publish web package
if: ${{ inputs.push_web_package }}
uses: JS-DevTools/npm-publish@19c28f1ef146469e409470805ea4279d47c3d35c
with:
token: ${{ secrets.NPM_TOKEN }}
package: tfhe/pkg/package.json
dry-run: ${{ inputs.dry_run }}
tag: ${{ env.NPM_TAG }}
provenance: true
- name: Build Node package
if: ${{ inputs.push_node_package }}
run: |
rm -rf tfhe/pkg
make build_node_js_api
sed -i 's/"tfhe"/"node-tfhe"/g' tfhe/pkg/package.json
- name: Publish Node package
if: ${{ inputs.push_node_package }}
uses: JS-DevTools/npm-publish@19c28f1ef146469e409470805ea4279d47c3d35c
with:
token: ${{ secrets.NPM_TOKEN }}
package: tfhe/pkg/package.json
dry-run: ${{ inputs.dry_run }}
tag: ${{ env.NPM_TAG }}
provenance: true
- name: Slack Notification
if: ${{ failure() }}
continue-on-error: true
uses: rtCamp/action-slack-notify@c33737706dea87cd7784c687dadc9adf1be59990
env:
SLACK_COLOR: ${{ job.status }}
SLACK_CHANNEL: ${{ secrets.SLACK_CHANNEL }}
SLACK_ICON: https://pbs.twimg.com/profile_images/1274014582265298945/OjBKP9kn_400x400.png
SLACK_MESSAGE: "tfhe release failed: (${{ env.ACTION_RUN_URL }})"
SLACK_USERNAME: ${{ secrets.BOT_USERNAME }}
SLACK_WEBHOOK: ${{ secrets.SLACK_WEBHOOK }}

View File

@@ -1,142 +0,0 @@
# Common workflow to make crate release
name: make_release_common
on:
workflow_call:
inputs:
package-name:
type: string
required: true
dry-run:
type: boolean
default: true
secrets:
REPO_CHECKOUT_TOKEN:
required: true
SLACK_CHANNEL:
required: true
BOT_USERNAME:
required: true
SLACK_WEBHOOK:
required: true
ALLOWED_TEAM:
required: true
READ_ORG_TOKEN:
required: true
env:
ACTION_RUN_URL: ${{ github.server_url }}/${{ github.repository }}/actions/runs/${{ github.run_id }}
SLACK_CHANNEL: ${{ secrets.SLACK_CHANNEL }}
SLACK_ICON: https://pbs.twimg.com/profile_images/1274014582265298945/OjBKP9kn_400x400.png
SLACK_USERNAME: ${{ secrets.BOT_USERNAME }}
SLACK_WEBHOOK: ${{ secrets.SLACK_WEBHOOK }}
permissions: {}
# zizmor: ignore[concurrency-limits] caller workflow is responsible for the concurrency
jobs:
verify-triggering-actor:
name: make_release_common/verify-triggering-actor
if: startsWith(github.ref, 'refs/tags/')
uses: ./.github/workflows/verify_triggering_actor.yml
secrets:
ALLOWED_TEAM: ${{ secrets.ALLOWED_TEAM }}
READ_ORG_TOKEN: ${{ secrets.READ_ORG_TOKEN }}
package:
name: make_release_common/package
runs-on: ubuntu-latest
needs: verify-triggering-actor
outputs:
hash: ${{ steps.hash.outputs.hash }}
steps:
- name: Checkout
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # v6.0.2
with:
fetch-depth: 0
persist-credentials: 'false'
token: ${{ secrets.REPO_CHECKOUT_TOKEN }}
- name: Prepare package
env:
PACKAGE: ${{ inputs.package-name }}
run: |
cargo package -p "${PACKAGE}"
- uses: actions/upload-artifact@043fb46d1a93c77aae656e7c1c64a875d1fc6a0a # v7.0.1
with:
name: crate-${{ inputs.package-name }}
path: target/package/*.crate
- name: generate hash
id: hash
run: cd target/package && echo "hash=$(sha256sum ./*.crate | base64 -w0)" >> "${GITHUB_OUTPUT}"
provenance:
name: make_release_common/provenance
if: ${{ !inputs.dry-run }}
needs: package
# This action cannot be pinned to a specific commit (see https://github.com/slsa-framework/slsa-github-generator/blob/main/README.md#referencing-slsa-builders-and-generators)
uses: slsa-framework/slsa-github-generator/.github/workflows/generator_generic_slsa3.yml@v2.1.0 # zizmor: ignore[unpinned-uses] as said above SLSA cannot be pinned by tag today
permissions:
actions: read # Needed to detect the GitHub Actions environment
id-token: write # Needed to create the provenance via GitHub OIDC
contents: write # Needed to upload assets/artifacts
with:
# SHA-256 hashes of the Crate package.
base64-subjects: ${{ needs.package.outputs.hash }}
publish_release:
name: make_release_common/publish-release
needs: package
runs-on: ubuntu-latest
permissions:
id-token: write # Needed for OIDC token exchange on crates.io
steps:
- name: Checkout
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # v6.0.2
with:
fetch-depth: 0
persist-credentials: 'false'
token: ${{ secrets.REPO_CHECKOUT_TOKEN }}
- name: Download artifact
uses: actions/download-artifact@3e5f45b2cfb9172054b4087a40e8e0b5a5461e7c # v8.0.1
with:
name: crate-${{ inputs.package-name }}
path: target/package
- name: Authenticate on registry
uses: rust-lang/crates-io-auth-action@bbd81622f20ce9e2dd9622e3218b975523e45bbe # v1.0.4
id: auth
- name: Publish crate.io package
env:
CARGO_REGISTRY_TOKEN: ${{ steps.auth.outputs.token }}
PACKAGE: ${{ inputs.package-name }}
DRY_RUN: ${{ inputs.dry-run && '--dry-run' || '' }}
run: |
# DRY_RUN expansion cannot be double quoted when variable contains empty string otherwise cargo publish
# would fail. This is safe since DRY_RUN is handled in the env section above.
# shellcheck disable=SC2086
cargo publish -p "${PACKAGE}" ${DRY_RUN}
- name: Generate hash
id: published_hash
run: cd target/package && echo "pub_hash=$(sha256sum ./*.crate | base64 -w0)" >> "${GITHUB_OUTPUT}"
- name: Slack notification (hashes comparison)
if: ${{ needs.package.outputs.hash != steps.published_hash.outputs.pub_hash }}
continue-on-error: true
uses: rtCamp/action-slack-notify@e31e87e03dd19038e411e38ae27cbad084a90661 # v2.3.3
env:
SLACK_COLOR: failure
SLACK_MESSAGE: "SLSA ${{ inputs.package-name }} - hash comparison failure: (${{ env.ACTION_RUN_URL }})"
- name: Slack Notification
if: ${{ failure() }}
continue-on-error: true
uses: rtCamp/action-slack-notify@e31e87e03dd19038e411e38ae27cbad084a90661 # v2.3.3
env:
SLACK_COLOR: ${{ job.status }}
SLACK_MESSAGE: "${{ inputs.package-name }} release finished with status: ${{ job.status }}. (${{ env.ACTION_RUN_URL }})"

View File

@@ -1,268 +0,0 @@
# Common workflow to make crate release for CUDA backend
name: make_release_common_cuda
on:
workflow_call:
inputs:
package-name:
type: string
required: true
dry-run:
type: boolean
default: true
secrets:
REPO_CHECKOUT_TOKEN:
required: true
SLAB_ACTION_TOKEN:
required: true
SLAB_BASE_URL:
required: true
SLAB_URL:
required: true
JOB_SECRET:
required: true
SLACK_CHANNEL:
required: true
BOT_USERNAME:
required: true
SLACK_WEBHOOK:
required: true
ALLOWED_TEAM:
required: true
READ_ORG_TOKEN:
required: true
env:
ACTION_RUN_URL: ${{ github.server_url }}/${{ github.repository }}/actions/runs/${{ github.run_id }}
SLACK_CHANNEL: ${{ secrets.SLACK_CHANNEL }}
SLACK_ICON: https://pbs.twimg.com/profile_images/1274014582265298945/OjBKP9kn_400x400.png
SLACK_USERNAME: ${{ secrets.BOT_USERNAME }}
SLACK_WEBHOOK: ${{ secrets.SLACK_WEBHOOK }}
permissions: {}
# zizmor: ignore[concurrency-limits] only Zama organization members can trigger this workflow
jobs:
verify-triggering-actor:
name: make_release_common_cuda/verify-triggering-actor
if: startsWith(github.ref, 'refs/tags/')
uses: ./.github/workflows/verify_triggering_actor.yml
secrets:
ALLOWED_TEAM: ${{ secrets.ALLOWED_TEAM }}
READ_ORG_TOKEN: ${{ secrets.READ_ORG_TOKEN }}
setup-instance:
name: make_release_common_cuda/setup-instance
needs: verify-triggering-actor
runs-on: ubuntu-latest
outputs:
runner-name: ${{ steps.start-instance.outputs.label }}
steps:
- name: Start instance
id: start-instance
uses: zama-ai/slab-github-runner@5aee5d157f4a0201e5eaefc9cc648e5f9f5472a5 # v1.6.0
with:
mode: start
github-token: ${{ secrets.SLAB_ACTION_TOKEN }}
slab-url: ${{ secrets.SLAB_BASE_URL }}
job-secret: ${{ secrets.JOB_SECRET }}
backend: aws
profile: gpu-build
package:
name: make_release_common_cuda/package
needs: setup-instance
runs-on: ${{ needs.setup-instance.outputs.runner-name }}
outputs:
hash: ${{ steps.hash.outputs.hash }}
strategy:
fail-fast: false
# explicit include-based build matrix, of known valid options
matrix:
include:
- os: ubuntu-22.04
cuda: "12.2"
gcc: 9
env:
CUDA_PATH: /usr/local/cuda-${{ matrix.cuda }}
steps:
- name: Checkout
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # v6.0.2
with:
fetch-depth: 0
persist-credentials: "false"
token: ${{ secrets.REPO_CHECKOUT_TOKEN }}
- name: Install latest stable
uses: dtolnay/rust-toolchain@e97e2d8cc328f1b50210efc529dca0028893a2d9 # zizmor: ignore[stale-action-refs] this action doesn't create releases
with:
toolchain: stable
- name: Export CUDA variables
run: |
echo "$CUDA_PATH/bin" >> "${GITHUB_PATH}"
{
echo "CUDA_PATH=$CUDA_PATH";
echo "LD_LIBRARY_PATH=$CUDA_PATH/lib:$LD_LIBRARY_PATH";
echo "CUDACXX=/usr/local/cuda-${CUDA_VERSION}/bin/nvcc";
} >> "${GITHUB_ENV}"
env:
CUDA_VERSION: ${{ matrix.cuda }}
# Specify the correct host compilers
- name: Export gcc and g++ variables
run: |
{
echo "CC=/usr/bin/gcc-${GCC_VERSION}";
echo "CXX=/usr/bin/g++-${GCC_VERSION}";
echo "CUDAHOSTCXX=/usr/bin/g++-${GCC_VERSION}";
echo "HOME=/home/ubuntu";
} >> "${GITHUB_ENV}"
env:
GCC_VERSION: ${{ matrix.gcc }}
- name: Prepare package
env:
PACKAGE: ${{ inputs.package-name }}
run: |
cargo package -p "${PACKAGE}"
- uses: actions/upload-artifact@043fb46d1a93c77aae656e7c1c64a875d1fc6a0a # v7.0.1
with:
name: crate-${{ inputs.package-name }}
path: target/package/*.crate
- name: generate hash
id: hash
run: cd target/package && echo "hash=$(sha256sum ./*.crate | base64 -w0)" >> "${GITHUB_OUTPUT}"
provenance:
name: make_release_common_cuda/provenance
if: ${{ !inputs.dry-run }}
needs: [package]
# This action cannot be pinned to a specific commit (see https://github.com/slsa-framework/slsa-github-generator/blob/main/README.md#referencing-slsa-builders-and-generators)
uses: slsa-framework/slsa-github-generator/.github/workflows/generator_generic_slsa3.yml@v2.1.0 # zizmor: ignore[unpinned-uses] as said above SLSA cannot be pinned by tag today
permissions:
actions: read # Needed to detect the GitHub Actions environment
id-token: write # Needed to create the provenance via GitHub OIDC
contents: write # Needed to upload assets/artifacts
with:
# SHA-256 hashes of the Crate package.
base64-subjects: ${{ needs.package.outputs.hash }}
publish-cuda-release:
name: make_release_common_cuda/publish-cuda-release
needs: [setup-instance, package] # for comparing hashes
runs-on: ${{ needs.setup-instance.outputs.runner-name }}
permissions:
id-token: write # Needed for OIDC token exchange on crates.io
strategy:
fail-fast: false
# explicit include-based build matrix, of known valid options
matrix:
include:
- os: ubuntu-22.04
cuda: "12.2"
gcc: 9
env:
CUDA_PATH: /usr/local/cuda-${{ matrix.cuda }}
steps:
- name: Install latest stable
uses: dtolnay/rust-toolchain@e97e2d8cc328f1b50210efc529dca0028893a2d9 # zizmor: ignore[stale-action-refs] this action doesn't create releases
with:
toolchain: stable
- name: Export CUDA variables
run: |
echo "$CUDA_PATH/bin" >> "${GITHUB_PATH}"
{
echo "CUDA_PATH=$CUDA_PATH";
echo "LD_LIBRARY_PATH=$CUDA_PATH/lib:$LD_LIBRARY_PATH";
echo "CUDACXX=/usr/local/cuda-${CUDA_VERSION}/bin/nvcc";
} >> "${GITHUB_ENV}"
env:
CUDA_VERSION: ${{ matrix.cuda }}
# Specify the correct host compilers
- name: Export gcc and g++ variables
run: |
{
echo "CC=/usr/bin/gcc-${GCC_VERSION}";
echo "CXX=/usr/bin/g++-${GCC_VERSION}";
echo "CUDAHOSTCXX=/usr/bin/g++-${GCC_VERSION}";
echo "HOME=/home/ubuntu";
} >> "${GITHUB_ENV}"
env:
GCC_VERSION: ${{ matrix.gcc }}
- name: Checkout
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # v6.0.2
with:
fetch-depth: 0
persist-credentials: "false"
token: ${{ secrets.REPO_CHECKOUT_TOKEN }}
- name: Download artifact
uses: actions/download-artifact@3e5f45b2cfb9172054b4087a40e8e0b5a5461e7c # v8.0.1
with:
name: crate-${{ inputs.package-name }}
path: target/package
- name: Authenticate on registry
uses: rust-lang/crates-io-auth-action@bbd81622f20ce9e2dd9622e3218b975523e45bbe # v1.0.4
id: auth
- name: Publish crate.io package
env:
CARGO_REGISTRY_TOKEN: ${{ steps.auth.outputs.token }}
PACKAGE: ${{ inputs.package-name }}
DRY_RUN: ${{ inputs.dry-run && '--dry-run' || '' }}
run: |
# DRY_RUN expansion cannot be double quoted when variable contains empty string otherwise cargo publish
# would fail. This is safe since DRY_RUN is handled in the env section above.
# shellcheck disable=SC2086
cargo publish -p "${PACKAGE}" ${DRY_RUN}
- name: Generate hash
id: published_hash
run: cd target/package && echo "pub_hash=$(sha256sum ./*.crate | base64 -w0)" >> "${GITHUB_OUTPUT}"
- name: Slack notification (hashes comparison)
if: ${{ needs.package.outputs.hash != steps.published_hash.outputs.pub_hash }}
continue-on-error: true
uses: rtCamp/action-slack-notify@e31e87e03dd19038e411e38ae27cbad084a90661 # v2.3.3
env:
SLACK_COLOR: failure
SLACK_MESSAGE: "SLSA ${{ inputs.package-name }} crate - hash comparison failure: (${{ env.ACTION_RUN_URL }})"
- name: Slack Notification
if: ${{ failure() || (cancelled() && github.event_name != 'pull_request') }}
continue-on-error: true
uses: rtCamp/action-slack-notify@e31e87e03dd19038e411e38ae27cbad084a90661 # v2.3.3
env:
SLACK_COLOR: ${{ job.status }}
SLACK_MESSAGE: "${{ inputs.package-name }} release finished with status: ${{ job.status }}. (${{ env.ACTION_RUN_URL }})"
teardown-instance:
name: make_release_common_cuda/teardown-instance
if: ${{ always() && needs.setup-instance.result == 'success' }}
needs: [setup-instance, publish-cuda-release]
runs-on: ubuntu-latest
steps:
- name: Stop instance
id: stop-instance
uses: zama-ai/slab-github-runner@5aee5d157f4a0201e5eaefc9cc648e5f9f5472a5 # v1.6.0
with:
mode: stop
github-token: ${{ secrets.SLAB_ACTION_TOKEN }}
slab-url: ${{ secrets.SLAB_BASE_URL }}
job-secret: ${{ secrets.JOB_SECRET }}
label: ${{ needs.setup-instance.outputs.runner-name }}
- name: Slack Notification
if: ${{ failure() }}
uses: rtCamp/action-slack-notify@e31e87e03dd19038e411e38ae27cbad084a90661 # v2.3.3
env:
SLACK_COLOR: ${{ job.status }}
SLACK_MESSAGE: "Instance teardown (${{ inputs.package-name }} release) finished with status: ${{ job.status }}. (${{ env.ACTION_RUN_URL }})"

View File

@@ -0,0 +1,49 @@
name: Publish tfhe-csprng release
on:
workflow_dispatch:
inputs:
dry_run:
description: "Dry-run"
type: boolean
default: true
env:
ACTION_RUN_URL: ${{ github.server_url }}/${{ github.repository }}/actions/runs/${{ github.run_id }}
jobs:
verify_tag:
uses: ./.github/workflows/verify_tagged_commit.yml
secrets:
RELEASE_TEAM: ${{ secrets.RELEASE_TEAM }}
READ_ORG_TOKEN: ${{ secrets.READ_ORG_TOKEN }}
publish_release:
name: Publish tfhe-csprng Release
needs: verify_tag
runs-on: ubuntu-latest
steps:
- name: Checkout
uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683
with:
fetch-depth: 0
token: ${{ secrets.FHE_ACTIONS_TOKEN }}
- name: Publish crate.io package
env:
CRATES_TOKEN: ${{ secrets.CARGO_REGISTRY_TOKEN }}
DRY_RUN: ${{ inputs.dry_run && '--dry-run' || '' }}
run: |
cargo publish -p tfhe-csprng --token ${{ env.CRATES_TOKEN }} ${{ env.DRY_RUN }}
- name: Slack Notification
if: ${{ failure() }}
continue-on-error: true
uses: rtCamp/action-slack-notify@c33737706dea87cd7784c687dadc9adf1be59990
env:
SLACK_COLOR: ${{ job.status }}
SLACK_CHANNEL: ${{ secrets.SLACK_CHANNEL }}
SLACK_ICON: https://pbs.twimg.com/profile_images/1274014582265298945/OjBKP9kn_400x400.png
SLACK_MESSAGE: "tfhe-csprng release finished with status: ${{ job.status }}. (${{ env.ACTION_RUN_URL }})"
SLACK_USERNAME: ${{ secrets.BOT_USERNAME }}
SLACK_WEBHOOK: ${{ secrets.SLACK_WEBHOOK }}

137
.github/workflows/make_release_cuda.yml vendored Normal file
View File

@@ -0,0 +1,137 @@
# Publish new release of tfhe-cuda-backend on crates.io.
name: Publish CUDA release
on:
workflow_dispatch:
inputs:
dry_run:
description: "Dry-run"
type: boolean
default: true
push_to_crates:
description: "Push to crate"
type: boolean
default: true
env:
ACTION_RUN_URL: ${{ github.server_url }}/${{ github.repository }}/actions/runs/${{ github.run_id }}
SLACK_CHANNEL: ${{ secrets.SLACK_CHANNEL }}
SLACK_ICON: https://pbs.twimg.com/profile_images/1274014582265298945/OjBKP9kn_400x400.png
SLACK_USERNAME: ${{ secrets.BOT_USERNAME }}
SLACK_WEBHOOK: ${{ secrets.SLACK_WEBHOOK }}
jobs:
verify_tag:
uses: ./.github/workflows/verify_tagged_commit.yml
secrets:
RELEASE_TEAM: ${{ secrets.RELEASE_TEAM }}
READ_ORG_TOKEN: ${{ secrets.READ_ORG_TOKEN }}
setup-instance:
name: Setup instance (publish-cuda-release)
needs: verify_tag
runs-on: ubuntu-latest
outputs:
runner-name: ${{ steps.start-instance.outputs.label }}
steps:
- name: Start instance
id: start-instance
uses: zama-ai/slab-github-runner@98f0788261a7323d5d695a883e20df36591a92b7
with:
mode: start
github-token: ${{ secrets.SLAB_ACTION_TOKEN }}
slab-url: ${{ secrets.SLAB_BASE_URL }}
job-secret: ${{ secrets.JOB_SECRET }}
backend: aws
profile: gpu-build
publish-cuda-release:
name: Publish CUDA Release
needs: setup-instance
runs-on: ${{ needs.setup-instance.outputs.runner-name }}
strategy:
fail-fast: false
# explicit include-based build matrix, of known valid options
matrix:
include:
- os: ubuntu-22.04
cuda: "12.2"
gcc: 9
env:
CUDA_PATH: /usr/local/cuda-${{ matrix.cuda }}
steps:
- name: Checkout
uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683
with:
fetch-depth: 0
token: ${{ secrets.FHE_ACTIONS_TOKEN }}
- name: Set up home
run: |
echo "HOME=/home/ubuntu" >> "${GITHUB_ENV}"
- name: Install latest stable
uses: dtolnay/rust-toolchain@315e265cd78dad1e1dcf3a5074f6d6c47029d5aa
with:
toolchain: stable
- name: Export CUDA variables
if: ${{ !cancelled() }}
run: |
echo "$CUDA_PATH/bin" >> "${GITHUB_PATH}"
{
echo "CUDA_PATH=$CUDA_PATH";
echo "LD_LIBRARY_PATH=$CUDA_PATH/lib:$LD_LIBRARY_PATH";
echo "CUDACXX=/usr/local/cuda-${{ matrix.cuda }}/bin/nvcc";
} >> "${GITHUB_ENV}"
# Specify the correct host compilers
- name: Export gcc and g++ variables
if: ${{ !cancelled() }}
run: |
{
echo "CC=/usr/bin/gcc-${{ matrix.gcc }}";
echo "CXX=/usr/bin/g++-${{ matrix.gcc }}";
echo "CUDAHOSTCXX=/usr/bin/g++-${{ matrix.gcc }}";
echo "HOME=/home/ubuntu";
} >> "${GITHUB_ENV}"
- name: Publish crate.io package
if: ${{ inputs.push_to_crates }}
env:
CRATES_TOKEN: ${{ secrets.CARGO_REGISTRY_TOKEN }}
DRY_RUN: ${{ inputs.dry_run && '--dry-run' || '' }}
run: |
cargo publish -p tfhe-cuda-backend --token ${{ env.CRATES_TOKEN }} ${{ env.DRY_RUN }}
- name: Slack Notification
if: ${{ failure() }}
continue-on-error: true
uses: rtCamp/action-slack-notify@c33737706dea87cd7784c687dadc9adf1be59990
env:
SLACK_COLOR: ${{ job.status }}
SLACK_MESSAGE: "tfhe-cuda-backend release finished with status: ${{ job.status }}. (${{ env.ACTION_RUN_URL }})"
teardown-instance:
name: Teardown instance (publish-release)
if: ${{ always() && needs.setup-instance.result != 'skipped' }}
needs: [ setup-instance, publish-cuda-release ]
runs-on: ubuntu-latest
steps:
- name: Stop instance
id: stop-instance
uses: zama-ai/slab-github-runner@98f0788261a7323d5d695a883e20df36591a92b7
with:
mode: stop
github-token: ${{ secrets.SLAB_ACTION_TOKEN }}
slab-url: ${{ secrets.SLAB_BASE_URL }}
job-secret: ${{ secrets.JOB_SECRET }}
label: ${{ needs.setup-instance.outputs.runner-name }}
- name: Slack Notification
if: ${{ failure() }}
continue-on-error: true
uses: rtCamp/action-slack-notify@c33737706dea87cd7784c687dadc9adf1be59990
env:
SLACK_COLOR: ${{ job.status }}
SLACK_MESSAGE: "Instance teardown (publish-cuda-release) finished with status: ${{ job.status }}. (${{ env.ACTION_RUN_URL }})"

View File

@@ -1,32 +0,0 @@
name: make_release_hpu
on:
workflow_dispatch:
inputs:
dry_run:
description: "Dry-run"
type: boolean
default: true
permissions: {}
# zizmor: ignore[concurrency-limits] only Zama organization members can trigger this workflow
jobs:
make-release:
name: make_release_hpu/make-release
uses: ./.github/workflows/make_release_common.yml
with:
package-name: "tfhe-hpu-backend"
dry-run: ${{ inputs.dry_run }}
permissions:
actions: read # Needed to detect the GitHub Actions environment
id-token: write # Needed to create the provenance via GitHub OIDC
contents: write # Needed to upload assets/artifacts
secrets:
BOT_USERNAME: ${{ secrets.BOT_USERNAME }}
SLACK_CHANNEL: ${{ secrets.SLACK_CHANNEL }}
SLACK_WEBHOOK: ${{ secrets.SLACK_WEBHOOK }}
REPO_CHECKOUT_TOKEN: ${{ secrets.REPO_CHECKOUT_TOKEN }}
ALLOWED_TEAM: ${{ secrets.RELEASE_TEAM }}
READ_ORG_TOKEN: ${{ secrets.READ_ORG_TOKEN }}

View File

@@ -1,146 +0,0 @@
# Publish new release of tfhe-rs on various platform.
name: make_release_tfhe
on:
workflow_dispatch:
inputs:
dry_run:
description: "Dry-run"
type: boolean
default: true
push_to_crates:
description: "Push to crate"
type: boolean
default: true
push_web_package:
description: "Push web js package"
type: boolean
default: true
push_web_compat_package:
description: "Push web compat (cross-origin) js package"
type: boolean
default: true
push_node_package:
description: "Push node js package"
type: boolean
default: true
npm_latest_tag:
description: "Set NPM tag as latest"
type: boolean
default: false
env:
ACTION_RUN_URL: ${{ github.server_url }}/${{ github.repository }}/actions/runs/${{ github.run_id }}
NPM_TAG: ""
SLACK_CHANNEL: ${{ secrets.SLACK_CHANNEL }}
SLACK_ICON: https://pbs.twimg.com/profile_images/1274014582265298945/OjBKP9kn_400x400.png
SLACK_USERNAME: ${{ secrets.BOT_USERNAME }}
SLACK_WEBHOOK: ${{ secrets.SLACK_WEBHOOK }}
permissions: {}
# zizmor: ignore[concurrency-limits] only Zama organization members can trigger this workflow
jobs:
make-release:
name: make_release_tfhe/make-release
uses: ./.github/workflows/make_release_common.yml
if: ${{ inputs.push_to_crates }}
with:
package-name: "tfhe"
dry-run: ${{ inputs.dry_run }}
permissions:
actions: read # Needed to detect the GitHub Actions environment
id-token: write # Needed to create the provenance via GitHub OIDC
contents: write # Needed to upload assets/artifacts
secrets:
BOT_USERNAME: ${{ secrets.BOT_USERNAME }}
SLACK_CHANNEL: ${{ secrets.SLACK_CHANNEL }}
SLACK_WEBHOOK: ${{ secrets.SLACK_WEBHOOK }}
REPO_CHECKOUT_TOKEN: ${{ secrets.REPO_CHECKOUT_TOKEN }}
ALLOWED_TEAM: ${{ secrets.RELEASE_TEAM }}
READ_ORG_TOKEN: ${{ secrets.READ_ORG_TOKEN }}
make-release-js:
name: make_release_tfhe/make-release-js
needs: make-release
if: ${{ always() && needs.make-release.result != 'failure' }}
runs-on: ubuntu-latest
# For provenance of npmjs publish
permissions:
contents: read
id-token: write # also needed for OIDC token exchange on crates.io and npmjs.com
steps:
- name: Checkout
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # v6.0.2
with:
fetch-depth: 0
persist-credentials: 'false'
token: ${{ secrets.REPO_CHECKOUT_TOKEN }}
- name: Create NPM version tag
if: ${{ inputs.npm_latest_tag }}
run: |
echo "NPM_TAG=latest" >> "${GITHUB_ENV}"
- name: Build web package
if: ${{ inputs.push_web_package }}
run: |
make build_web_js_api_parallel
- name: Authenticate on NPM
uses: actions/setup-node@48b55a011bda9f5d6aeb4c2d9c7362e8dae4041e # v6.4.0
with:
node-version: '24'
registry-url: 'https://registry.npmjs.org'
- name: Publish web package
if: ${{ inputs.push_web_package }}
uses: JS-DevTools/npm-publish@0fd2f4369c5d6bcfcde6091a7c527d810b9b5c3f
with:
package: tfhe/pkg/package.json
dry-run: ${{ inputs.dry_run }}
tag: ${{ env.NPM_TAG }}
provenance: true
- name: Build web compat (cross-origin) package
if: ${{ inputs.push_web_compat_package }}
run: |
rm -rf tfhe/pkg
make build_web_js_api
sed -i 's/"tfhe"/"tfhe-compat"/g' tfhe/pkg/package.json
- name: Publish web compat (cross-origin) package
if: ${{ inputs.push_web_compat_package }}
uses: JS-DevTools/npm-publish@0fd2f4369c5d6bcfcde6091a7c527d810b9b5c3f
with:
package: tfhe/pkg/package.json
dry-run: ${{ inputs.dry_run }}
tag: ${{ env.NPM_TAG }}
provenance: true
- name: Build Node package
if: ${{ inputs.push_node_package }}
run: |
rm -rf tfhe/pkg
make build_node_js_api
sed -i 's/"tfhe"/"node-tfhe"/g' tfhe/pkg/package.json
- name: Publish Node package
if: ${{ inputs.push_node_package }}
uses: JS-DevTools/npm-publish@0fd2f4369c5d6bcfcde6091a7c527d810b9b5c3f
with:
package: tfhe/pkg/package.json
dry-run: ${{ inputs.dry_run }}
tag: ${{ env.NPM_TAG }}
provenance: true
- name: Slack Notification
if: ${{ failure() }}
continue-on-error: true
uses: rtCamp/action-slack-notify@e31e87e03dd19038e411e38ae27cbad084a90661 # v2.3.3
env:
SLACK_COLOR: ${{ job.status }}
SLACK_MESSAGE: "tfhe release finished with status: ${{ job.status }}. (${{ env.ACTION_RUN_URL }})"

View File

@@ -1,32 +0,0 @@
name: make_release_tfhe_csprng
on:
workflow_dispatch:
inputs:
dry_run:
description: "Dry-run"
type: boolean
default: true
permissions: {}
# zizmor: ignore[concurrency-limits] only Zama organization members can trigger this workflow
jobs:
make-release:
name: make_release_tfhe_csprng/make-release
uses: ./.github/workflows/make_release_common.yml
with:
package-name: "tfhe-csprng"
dry-run: ${{ inputs.dry_run }}
permissions:
actions: read # Needed to detect the GitHub Actions environment
id-token: write # Needed to create the provenance via GitHub OIDC
contents: write # Needed to upload assets/artifacts
secrets:
BOT_USERNAME: ${{ secrets.BOT_USERNAME }}
SLACK_CHANNEL: ${{ secrets.SLACK_CHANNEL }}
SLACK_WEBHOOK: ${{ secrets.SLACK_WEBHOOK }}
REPO_CHECKOUT_TOKEN: ${{ secrets.REPO_CHECKOUT_TOKEN }}
ALLOWED_TEAM: ${{ secrets.RELEASE_TEAM }}
READ_ORG_TOKEN: ${{ secrets.READ_ORG_TOKEN }}

View File

@@ -1,44 +0,0 @@
# Publish new release of tfhe-rs CUDA backend on crates.io.
name: make_release_tfhe_cuda
on:
workflow_dispatch:
inputs:
dry_run:
description: "Dry-run"
type: boolean
default: true
env:
ACTION_RUN_URL: ${{ github.server_url }}/${{ github.repository }}/actions/runs/${{ github.run_id }}
SLACK_CHANNEL: ${{ secrets.SLACK_CHANNEL }}
SLACK_ICON: https://pbs.twimg.com/profile_images/1274014582265298945/OjBKP9kn_400x400.png
SLACK_USERNAME: ${{ secrets.BOT_USERNAME }}
SLACK_WEBHOOK: ${{ secrets.SLACK_WEBHOOK }}
permissions: {}
# zizmor: ignore[concurrency-limits] only Zama organization members can trigger this workflow
jobs:
make-release:
name: make_release_tfhe_cuda/make-release
uses: ./.github/workflows/make_release_common_cuda.yml
with:
package-name: "tfhe-cuda-backend"
dry-run: ${{ inputs.dry_run }}
permissions:
actions: read # Needed to detect the GitHub Actions environment
id-token: write # Needed to create the provenance via GitHub OIDC
contents: write # Needed to upload assets/artifacts
secrets:
BOT_USERNAME: ${{ secrets.BOT_USERNAME }}
SLACK_CHANNEL: ${{ secrets.SLACK_CHANNEL }}
SLACK_WEBHOOK: ${{ secrets.SLACK_WEBHOOK }}
REPO_CHECKOUT_TOKEN: ${{ secrets.REPO_CHECKOUT_TOKEN }}
ALLOWED_TEAM: ${{ secrets.RELEASE_TEAM }}
READ_ORG_TOKEN: ${{ secrets.READ_ORG_TOKEN }}
SLAB_ACTION_TOKEN: ${{ secrets.SLAB_ACTION_TOKEN }}
SLAB_BASE_URL: ${{ secrets.SLAB_BASE_URL }}
SLAB_URL: ${{ secrets.SLAB_URL }}
JOB_SECRET: ${{ secrets.JOB_SECRET }}

View File

@@ -1,5 +1,5 @@
# Publish new release of tfhe-fft
name: make_release_tfhe_fft
name: Publish tfhe-fft release
on:
workflow_dispatch:
@@ -9,25 +9,41 @@ on:
type: boolean
default: true
permissions: {}
# zizmor: ignore[concurrency-limits] only Zama organization members can trigger this workflow
env:
ACTION_RUN_URL: ${{ github.server_url }}/${{ github.repository }}/actions/runs/${{ github.run_id }}
jobs:
make-release:
name: make_release_tfhe_fft/make-release
uses: ./.github/workflows/make_release_common.yml
with:
package-name: "tfhe-fft"
dry-run: ${{ inputs.dry_run }}
permissions:
actions: read # Needed to detect the GitHub Actions environment
id-token: write # Needed to create the provenance via GitHub OIDC
contents: write # Needed to upload assets/artifacts
verify_tag:
uses: ./.github/workflows/verify_tagged_commit.yml
secrets:
BOT_USERNAME: ${{ secrets.BOT_USERNAME }}
SLACK_CHANNEL: ${{ secrets.SLACK_CHANNEL }}
SLACK_WEBHOOK: ${{ secrets.SLACK_WEBHOOK }}
REPO_CHECKOUT_TOKEN: ${{ secrets.REPO_CHECKOUT_TOKEN }}
ALLOWED_TEAM: ${{ secrets.RELEASE_TEAM }}
RELEASE_TEAM: ${{ secrets.RELEASE_TEAM }}
READ_ORG_TOKEN: ${{ secrets.READ_ORG_TOKEN }}
publish_release:
name: Publish tfhe-fft Release
runs-on: ubuntu-latest
needs: verify_tag
steps:
- name: Checkout
uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683
with:
fetch-depth: 0
- name: Publish crate.io package
env:
CRATES_TOKEN: ${{ secrets.CARGO_REGISTRY_TOKEN }}
DRY_RUN: ${{ inputs.dry_run && '--dry-run' || '' }}
run: |
cargo publish -p tfhe-fft --token ${{ env.CRATES_TOKEN }} ${{ env.DRY_RUN }}
- name: Slack Notification
if: ${{ failure() }}
continue-on-error: true
uses: rtCamp/action-slack-notify@c33737706dea87cd7784c687dadc9adf1be59990
env:
SLACK_COLOR: ${{ job.status }}
SLACK_CHANNEL: ${{ secrets.SLACK_CHANNEL }}
SLACK_ICON: https://pbs.twimg.com/profile_images/1274014582265298945/OjBKP9kn_400x400.png
SLACK_MESSAGE: "tfhe-fft release failed: (${{ env.ACTION_RUN_URL }})"
SLACK_USERNAME: ${{ secrets.BOT_USERNAME }}
SLACK_WEBHOOK: ${{ secrets.SLACK_WEBHOOK }}

View File

@@ -1,5 +1,5 @@
# Publish new release of tfhe-ntt
name: make_release_tfhe_ntt
name: Publish tfhe-ntt release
on:
workflow_dispatch:
@@ -9,25 +9,41 @@ on:
type: boolean
default: true
permissions: {}
# zizmor: ignore[concurrency-limits] only Zama organization members can trigger this workflow
env:
ACTION_RUN_URL: ${{ github.server_url }}/${{ github.repository }}/actions/runs/${{ github.run_id }}
jobs:
make-release:
name: make_release_tfhe_ntt/make-release
uses: ./.github/workflows/make_release_common.yml
with:
package-name: "tfhe-ntt"
dry-run: ${{ inputs.dry_run }}
permissions:
actions: read # Needed to detect the GitHub Actions environment
id-token: write # Needed to create the provenance via GitHub OIDC
contents: write # Needed to upload assets/artifacts
verify_tag:
uses: ./.github/workflows/verify_tagged_commit.yml
secrets:
BOT_USERNAME: ${{ secrets.BOT_USERNAME }}
SLACK_CHANNEL: ${{ secrets.SLACK_CHANNEL }}
SLACK_WEBHOOK: ${{ secrets.SLACK_WEBHOOK }}
REPO_CHECKOUT_TOKEN: ${{ secrets.REPO_CHECKOUT_TOKEN }}
ALLOWED_TEAM: ${{ secrets.RELEASE_TEAM }}
RELEASE_TEAM: ${{ secrets.RELEASE_TEAM }}
READ_ORG_TOKEN: ${{ secrets.READ_ORG_TOKEN }}
publish_release:
name: Publish tfhe-ntt Release
runs-on: ubuntu-latest
needs: verify_tag
steps:
- name: Checkout
uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683
with:
fetch-depth: 0
- name: Publish crate.io package
env:
CRATES_TOKEN: ${{ secrets.CARGO_REGISTRY_TOKEN }}
DRY_RUN: ${{ inputs.dry_run && '--dry-run' || '' }}
run: |
cargo publish -p tfhe-ntt --token ${{ env.CRATES_TOKEN }} ${{ env.DRY_RUN }}
- name: Slack Notification
if: ${{ failure() }}
continue-on-error: true
uses: rtCamp/action-slack-notify@c33737706dea87cd7784c687dadc9adf1be59990
env:
SLACK_COLOR: ${{ job.status }}
SLACK_CHANNEL: ${{ secrets.SLACK_CHANNEL }}
SLACK_ICON: https://pbs.twimg.com/profile_images/1274014582265298945/OjBKP9kn_400x400.png
SLACK_MESSAGE: "tfhe-ntt release failed: (${{ env.ACTION_RUN_URL }})"
SLACK_USERNAME: ${{ secrets.BOT_USERNAME }}
SLACK_WEBHOOK: ${{ secrets.SLACK_WEBHOOK }}

View File

@@ -1,32 +0,0 @@
name: make_release_tfhe_safe_serialize
on:
workflow_dispatch:
inputs:
dry_run:
description: "Dry-run"
type: boolean
default: true
permissions: {}
# zizmor: ignore[concurrency-limits] only Zama organization members can trigger this workflow
jobs:
make-release:
name: make_release_tfhe_safe_serialize/make-release
uses: ./.github/workflows/make_release_common.yml
with:
package-name: "tfhe-safe-serialize"
dry-run: ${{ inputs.dry_run }}
permissions:
actions: read # Needed to detect the GitHub Actions environment
id-token: write # Needed to create the provenance via GitHub OIDC
contents: write # Needed to upload assets/artifacts
secrets:
BOT_USERNAME: ${{ secrets.BOT_USERNAME }}
SLACK_CHANNEL: ${{ secrets.SLACK_CHANNEL }}
SLACK_WEBHOOK: ${{ secrets.SLACK_WEBHOOK }}
REPO_CHECKOUT_TOKEN: ${{ secrets.REPO_CHECKOUT_TOKEN }}
ALLOWED_TEAM: ${{ secrets.RELEASE_TEAM }}
READ_ORG_TOKEN: ${{ secrets.READ_ORG_TOKEN }}

Some files were not shown because too many files have changed in this diff Show More