forked from NVIDIA/cuda-python
-
Notifications
You must be signed in to change notification settings - Fork 0
277 lines (240 loc) · 11.2 KB
/
test-wheel-windows.yml
File metadata and controls
277 lines (240 loc) · 11.2 KB
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
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
# SPDX-FileCopyrightText: Copyright (c) 2024-2026 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
#
# SPDX-License-Identifier: Apache-2.0
name: "CI: Test wheels"
on:
workflow_call:
inputs:
build-type:
type: string
required: true
host-platform:
type: string
required: true
build-ctk-ver:
type: string
required: true
matrix_filter:
type: string
default: "."
nruns:
type: number
default: 1
jobs:
compute-matrix:
runs-on: ubuntu-latest
defaults:
run:
shell: bash --noprofile --norc -xeuo pipefail {0}
env:
BUILD_TYPE: ${{ inputs.build-type }}
ARCH: ${{ (inputs.host-platform == 'win-64' && 'amd64') }}
outputs:
MATRIX: ${{ steps.compute-matrix.outputs.MATRIX }}
steps:
- name: Checkout ${{ github.event.repository.name }}
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # v6.0.2
- name: Validate Test Type
run: |
if [[ "$BUILD_TYPE" != "pull-request" ]] && [[ "$BUILD_TYPE" != "nightly" ]] && [[ "$BUILD_TYPE" != "branch" ]]; then
echo "Invalid build type! Must be one of 'nightly', 'pull-request', or 'branch'."
exit 1
fi
- name: Compute Python Test Matrix
id: compute-matrix
run: |
# Use the nightly matrix for branch tests
MATRIX_TYPE="${BUILD_TYPE}"
if [[ "${MATRIX_TYPE}" == "branch" ]]; then
MATRIX_TYPE="nightly"
fi
# Read base matrix from YAML file for the specific architecture
TEST_MATRIX=$(yq -o json ".windows[\"${MATRIX_TYPE}\"] | map(select(.ARCH == \"${ARCH}\"))" ci/test-matrix.yml)
# Apply matrix filter and wrap in include structure
MATRIX=$(echo "$TEST_MATRIX" | jq -c '${{ inputs.matrix_filter }} | if (. | length) > 0 then {include: .} else "Error: Empty matrix\n" | halt_error(1) end')
echo "MATRIX=${MATRIX}" | tee --append "${GITHUB_OUTPUT}"
test:
name: py${{ matrix.PY_VER }}, ${{ matrix.CUDA_VER }}, ${{ (matrix.LOCAL_CTK == '1' && 'local') || 'wheels' }}, ${{ matrix.GPU }} (${{ matrix.DRIVER_MODE }})
# The build stage could fail but we want the CI to keep moving.
needs: compute-matrix
strategy:
fail-fast: false
matrix: ${{ fromJSON(needs.compute-matrix.outputs.MATRIX) }}
if: ${{ github.repository_owner == 'nvidia' && !cancelled() }}
runs-on: "windows-${{ matrix.ARCH }}-gpu-${{ matrix.GPU }}-${{ matrix.DRIVER }}-${{ matrix.GPU_COUNT }}"
steps:
- name: Checkout ${{ github.event.repository.name }}
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # v6.0.2
- name: Setup proxy cache
uses: nv-gha-runners/setup-proxy-cache@main
continue-on-error: true
with:
enable-apt: true
- name: Update driver
env:
DRIVER_MODE: ${{ matrix.DRIVER_MODE }}
GPU_TYPE: ${{ matrix.GPU }}
run: |
ci/tools/install_gpu_driver.ps1
- name: Ensure GPU is working
run: |
nvidia-smi
$mode_output = nvidia-smi | Select-String -Pattern "${{ matrix.DRIVER_MODE }}"
Write-Output "Driver mode check: $mode_output"
if ("$mode_output" -eq "") {
Write-Error "Switching to driver mode ${{ matrix.DRIVER_MODE }} failed!"
exit 1
}
Write-Output "Driver mode verified: ${{ matrix.DRIVER_MODE }}"
- name: Set environment variables
env:
BUILD_CUDA_VER: ${{ inputs.build-ctk-ver }}
CUDA_VER: ${{ matrix.CUDA_VER }}
HOST_PLATFORM: ${{ inputs.host-platform }}
LOCAL_CTK: ${{ matrix.LOCAL_CTK }}
PY_VER: ${{ matrix.PY_VER }}
SHA: ${{ github.sha }}
shell: bash --noprofile --norc -xeuo pipefail {0}
run: ./ci/tools/env-vars test
- name: Download cuda-pathfinder build artifacts
uses: actions/download-artifact@3e5f45b2cfb9172054b4087a40e8e0b5a5461e7c # v8.0.1
with:
name: cuda-pathfinder-wheel
path: ./cuda_pathfinder
- name: Download cuda-python build artifacts
if: ${{ env.SKIP_CUDA_BINDINGS_TEST == '0'}}
uses: actions/download-artifact@3e5f45b2cfb9172054b4087a40e8e0b5a5461e7c # v8.0.1
with:
name: cuda-python-wheel
path: .
- name: Download cuda.bindings build artifacts
if: ${{ env.SKIP_CUDA_BINDINGS_TEST == '0'}}
uses: actions/download-artifact@3e5f45b2cfb9172054b4087a40e8e0b5a5461e7c # v8.0.1
with:
name: ${{ env.CUDA_BINDINGS_ARTIFACT_NAME }}
path: ${{ env.CUDA_BINDINGS_ARTIFACTS_DIR }}
- name: Download cuda-python & cuda.bindings build artifacts from the prior branch
if: ${{ env.SKIP_CUDA_BINDINGS_TEST == '1'}}
env:
GH_TOKEN: ${{ secrets.GITHUB_TOKEN }}
run: |
$OLD_BRANCH = yq '.backport_branch' ci/versions.yml
$OLD_BASENAME = "cuda-bindings-python${env:PYTHON_VERSION_FORMATTED}-cuda*-${{ inputs.host-platform }}*"
$runData = gh run list -b $OLD_BRANCH -L 1 -w "ci.yml" -s success -R NVIDIA/cuda-python --json databaseId | ConvertFrom-Json
if (-not $runData -or $runData.Length -eq 0 -or -not $runData[0].databaseId -or [string]::IsNullOrEmpty($runData[0].databaseId)) {
Write-Host "LATEST_PRIOR_RUN_ID not found!"
exit 1
}
$LATEST_PRIOR_RUN_ID = $runData[0].databaseId
gh run download $LATEST_PRIOR_RUN_ID -p $OLD_BASENAME -R NVIDIA/cuda-python
Remove-Item -Recurse -Force "${OLD_BASENAME}-tests" # exclude cython test artifacts
Get-ChildItem -Path $OLD_BASENAME
New-Item -Path "${{ env.CUDA_BINDINGS_ARTIFACTS_DIR }}" -ItemType Directory -Force
Move-Item -Path "$OLD_BASENAME/*.whl" -Destination "${{ env.CUDA_BINDINGS_ARTIFACTS_DIR }}"
Remove-Item -Path $OLD_BASENAME -Force
gh run download $LATEST_PRIOR_RUN_ID -p cuda-python-wheel -R NVIDIA/cuda-python
Get-ChildItem -Path cuda-python-wheel
Move-Item -Path "cuda-python-wheel/*.whl" -Destination .
Remove-Item -Path cuda-python-wheel -Force
- name: Display structure of downloaded cuda-python artifacts
run: |
Get-Location
Get-ChildItem -Recurse -Force | Select-Object Mode, LastWriteTime, Length, FullName
- name: Display structure of downloaded cuda.bindings artifacts
run: |
Get-Location
Get-ChildItem -Recurse -Force $env:CUDA_BINDINGS_ARTIFACTS_DIR | Select-Object Mode, LastWriteTime, Length, FullName
- name: Download cuda.bindings Cython tests
if: ${{ env.SKIP_CYTHON_TEST == '0' }}
uses: actions/download-artifact@3e5f45b2cfb9172054b4087a40e8e0b5a5461e7c # v8.0.1
with:
name: ${{ env.CUDA_BINDINGS_ARTIFACT_NAME }}-tests
path: ${{ env.CUDA_BINDINGS_CYTHON_TESTS_DIR }}
- name: Display structure of downloaded cuda.bindings Cython tests
if: ${{ env.SKIP_CYTHON_TEST == '0' }}
run: |
Get-Location
Get-ChildItem -Recurse -Force $env:CUDA_BINDINGS_CYTHON_TESTS_DIR | Select-Object Mode, LastWriteTime, Length, FullName
- name: Download cuda.core build artifacts
uses: actions/download-artifact@3e5f45b2cfb9172054b4087a40e8e0b5a5461e7c # v8.0.1
with:
name: ${{ env.CUDA_CORE_ARTIFACT_NAME }}
path: ${{ env.CUDA_CORE_ARTIFACTS_DIR }}
- name: Display structure of downloaded cuda.core build artifacts
run: |
Get-Location
Get-ChildItem -Recurse -Force $env:CUDA_CORE_ARTIFACTS_DIR | Select-Object Mode, LastWriteTime, Length, FullName
- name: Download cuda.core Cython tests
if: ${{ env.SKIP_CYTHON_TEST == '0' }}
uses: actions/download-artifact@3e5f45b2cfb9172054b4087a40e8e0b5a5461e7c # v8.0.1
with:
name: ${{ env.CUDA_CORE_ARTIFACT_NAME }}-tests
path: ${{ env.CUDA_CORE_CYTHON_TESTS_DIR }}
- name: Display structure of downloaded cuda.core Cython tests
if: ${{ env.SKIP_CYTHON_TEST == '0' }}
run: |
Get-Location
Get-ChildItem -Recurse -Force $env:CUDA_CORE_CYTHON_TESTS_DIR | Select-Object Mode, LastWriteTime, Length, FullName
- name: Set up Python ${{ matrix.PY_VER }}
uses: actions/setup-python@a309ff8b426b58ec0e2a45f0f869d46889d02405 # v6.2.0
with:
python-version: ${{ matrix.PY_VER }}
- name: Verify LongPathsEnabled
run: |
$val = (Get-ItemProperty -Path 'HKLM:\SYSTEM\CurrentControlSet\Control\FileSystem' -Name 'LongPathsEnabled').LongPathsEnabled
echo "LongPathsEnabled = $val"
if ($val -ne 1) {
echo "::error::LongPathsEnabled is not set to 1 (see issue #1820)"
exit 1
}
- name: Set up mini CTK
if: ${{ matrix.LOCAL_CTK == '1' }}
uses: ./.github/actions/fetch_ctk
continue-on-error: false
with:
host-platform: ${{ inputs.host-platform }}
cuda-version: ${{ matrix.CUDA_VER }}
- name: Set up test repetition on nightly runs
shell: bash --noprofile --norc -xeuo pipefail {0}
run: echo "PYTEST_ADDOPTS=\"--count=${{ inputs.nruns }}\"" >> "$GITHUB_ENV"
- name: Run cuda.pathfinder tests with see_what_works
env:
CUDA_PATHFINDER_TEST_LOAD_NVIDIA_DYNAMIC_LIB_STRICTNESS: see_what_works
CUDA_PATHFINDER_TEST_FIND_NVIDIA_HEADERS_STRICTNESS: see_what_works
CUDA_PATHFINDER_TEST_FIND_NVIDIA_BITCODE_LIB_STRICTNESS: see_what_works
shell: bash --noprofile --norc -xeuo pipefail {0}
run: run-tests pathfinder
- name: Run cuda.bindings tests
if: ${{ env.SKIP_CUDA_BINDINGS_TEST == '0' }}
env:
CUDA_VER: ${{ matrix.CUDA_VER }}
LOCAL_CTK: ${{ matrix.LOCAL_CTK }}
shell: bash --noprofile --norc -xeuo pipefail {0}
run: run-tests bindings
- name: Run cuda.core tests
env:
CUDA_VER: ${{ matrix.CUDA_VER }}
LOCAL_CTK: ${{ matrix.LOCAL_CTK }}
shell: bash --noprofile --norc -xeuo pipefail {0}
run: run-tests core
- name: Ensure cuda-python installable
run: |
if ('${{ matrix.LOCAL_CTK }}' -eq '1') {
pip install --only-binary=:all: (Get-ChildItem -Filter cuda_python*.whl).FullName
} else {
pip install --only-binary=:all: "$((Get-ChildItem -Filter cuda_python*.whl).FullName)[all]"
}
- name: Install cuda.pathfinder extra wheels for testing
shell: bash --noprofile --norc -xeuo pipefail {0}
run: |
pushd cuda_pathfinder
pip install --only-binary=:all: -v ./*.whl --group "test-cu${TEST_CUDA_MAJOR}"
pip list
popd
- name: Run cuda.pathfinder tests with all_must_work
env:
CUDA_PATHFINDER_TEST_LOAD_NVIDIA_DYNAMIC_LIB_STRICTNESS: all_must_work
CUDA_PATHFINDER_TEST_FIND_NVIDIA_HEADERS_STRICTNESS: all_must_work
CUDA_PATHFINDER_TEST_FIND_NVIDIA_BITCODE_LIB_STRICTNESS: all_must_work
shell: bash --noprofile --norc -xeuo pipefail {0}
run: run-tests pathfinder