vortex/.github/workflows/cuda.yaml at 6fca3ff952b0214c2baf342e47aed4c59a811a9e · vortex-data/vortex · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
name: CUDA

# Concurrency control:
# - PRs: new commits on a feature branch will cancel in-progress (outdated) runs.
# - Push to develop: runs queue sequentially, never cancelled.
# - `workflow_dispatch`: groups by branch and queues if run on develop.
concurrency:
  group: ${{ github.workflow }}-${{ github.ref }}
  cancel-in-progress: ${{ github.ref != 'refs/heads/develop' }}
on:
  push:
    branches: [develop]
  pull_request: { }
  workflow_dispatch: { }

permissions:
  contents: read

env:
  CARGO_TERM_COLOR: always
  RUST_BACKTRACE: 1

jobs:
  cuda-build-lint:
    if: github.repository == 'vortex-data/vortex'
    name: "CUDA build & lint"
    timeout-minutes: 30
    runs-on: runs-on=${{ github.run_id }}/runner=gpu/tag=cuda-build
    steps:
      - uses: runs-on/action@v2
        with:
          sccache: s3
      - uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd  # v6
      - uses: ./.github/actions/setup-rust
        with:
          repo-token: ${{ secrets.GITHUB_TOKEN }}
      - uses: ./.github/actions/check-rebuild
        with:
          command: >-
            cargo build --profile ci --locked --all-features --all-targets
            -p vortex-cuda -p vortex-cub -p vortex-nvcomp
            -p gpu-scan-cli -p vortex-test-e2e-cuda
      - name: Clippy CUDA crates
        run: |
          cargo clippy --profile ci --locked --all-features --all-targets \
            -p vortex-cuda \
            -p vortex-cub \
            -p vortex-nvcomp \
            -p gpu-scan-cli \
            -p vortex-test-e2e-cuda \
            -- -D warnings

  cuda-test:
    if: github.repository == 'vortex-data/vortex'
    name: "CUDA tests"
    timeout-minutes: 30
    runs-on: runs-on=${{ github.run_id }}/runner=gpu/tag=cuda-tests
    steps:
      - uses: runs-on/action@v2
        with:
          sccache: s3
      - name: Display NVIDIA SMI details
        run: |
          nvidia-smi
          nvidia-smi -L
          nvidia-smi -q -d Memory
      - uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd  # v6
      - uses: ./.github/actions/setup-rust
        with:
          repo-token: ${{ secrets.GITHUB_TOKEN }}
      - name: Install nextest
        uses: taiki-e/install-action@711e1c3275189d76dcc4d34ddea63bf96ac49090  # v2
        with:
          tool: nextest
      - name: Rust Tests
        env:
          FLAT_LAYOUT_INLINE_ARRAY_NODE: true
        run: |
          cargo nextest run \
            --cargo-profile ci \
            --locked \
            -p vortex-file \
            -p vortex-cuda \
            -p vortex-cub \
            -p vortex-nvcomp \
            -p vortex-test-e2e-cuda \
            --all-features \
            --no-fail-fast \
            --target x86_64-unknown-linux-gnu \
            --verbose

  cuda-test-sanitizer:
    if: github.repository == 'vortex-data/vortex'
    name: "CUDA tests (${{ matrix.sanitizer }})"
    timeout-minutes: 30
    runs-on: runs-on=${{ github.run_id }}/runner=gpu/tag=cuda-test-sanitizer
    strategy:
      fail-fast: false
      matrix:
        include:
          - sanitizer: memcheck
            runner_flags: "--tool memcheck --leak-check=full --error-exitcode 1"
          # TODO(joe): try to re-enable racecheck, it is hanging in CI.
          # - sanitizer: racecheck
          #   runner_flags: "--tool racecheck --error-exitcode 1"
          - sanitizer: synccheck
            runner_flags: "--tool synccheck --error-exitcode 1"
          - sanitizer: initcheck
            runner_flags: "--tool initcheck --error-exitcode 1"
    steps:
      - uses: runs-on/action@v2
        with:
          sccache: s3
      - name: Display NVIDIA SMI details
        run: |
          nvidia-smi
          nvidia-smi -L
          nvidia-smi -q -d Memory
      - uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd  # v6
      - uses: ./.github/actions/setup-rust
        with:
          repo-token: ${{ secrets.GITHUB_TOKEN }}
      - name: Build tests
        run: cargo test --profile ci --locked -p vortex-cuda --all-features --target x86_64-unknown-linux-gnu --no-run
      - name: "CUDA - ${{ matrix.sanitizer }}"
        env:
          CARGO_TARGET_X86_64_UNKNOWN_LINUX_GNU_RUNNER: "compute-sanitizer ${{ matrix.runner_flags }}"
        run: cargo test --profile ci --locked -p vortex-cuda --all-features --target x86_64-unknown-linux-gnu

  cuda-test-cudf:
    if: github.repository == 'vortex-data/vortex'
    name: "CUDA tests (cudf)"
    timeout-minutes: 30
    runs-on: runs-on=${{ github.run_id }}/runner=gpu/tag=cuda-test-cudf
    steps:
      - uses: runs-on/action@v2
        with:
          sccache: s3
      - name: Display NVIDIA SMI details
        run: |
          nvidia-smi
          nvidia-smi -L
          nvidia-smi -q -d Memory
      - uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd  # v6
      - uses: ./.github/actions/setup-rust
        with:
          repo-token: ${{ secrets.GITHUB_TOKEN }}
      - name: Build cudf test library
        run: cargo build --profile ci --locked -p vortex-test-e2e-cuda --target x86_64-unknown-linux-gnu
      - name: Download and run cudf-test-harness
        run: |
          curl -fsSL https://github.com/vortex-data/cudf-test-harness/releases/latest/download/cudf-test-harness-x86_64.tar.gz | tar -xz
          cd cudf-test-harness-x86_64
          compute-sanitizer --tool memcheck --error-exitcode 1 ./cudf-test-harness check $GITHUB_WORKSPACE/target/x86_64-unknown-linux-gnu/ci/libvortex_test_e2e_cuda.so