Skip to content

[CUDA] Upgrade build pipelines to use CUDA 12.8 + cuDNN 9.8 #26267

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
tianleiwu merged 14 commits into from
Oct 13, 2025
Merged

[CUDA] Upgrade build pipelines to use CUDA 12.8 + cuDNN 9.8 #26267

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
14 commits
Select commit Hold shift + click to select a range
6245011
CUDA build
Oct 8, 2025
218724f
update
Oct 8, 2025
c165a83
change base images
Oct 9, 2025
5a58b39
remove rdc to speed up build
tianleiwu Oct 9, 2025
77c5b46
Merge remote-tracking branch 'origin/main' into cuda128
Oct 9, 2025
5085600
Replace the Windows machine from onnxruntime-github-vs2022-mms to onn…
Oct 9, 2025
90a6a35
Update custom-nuget-packaging-pipeline.yml for Azure Pipelines
Oct 9, 2025
e851f63
Merge remote-tracking branch 'origin/main' into cuda128
Oct 9, 2025
1951b84
update
Oct 9, 2025
507ccc9
Merge remote-tracking branch 'origin/snnn/fl_pipeline' into cuda128
Oct 9, 2025
2e54006
update
Oct 9, 2025
fa9484f
disable failed onnx model tests
tianleiwu Oct 10, 2025
824554b
update
tianleiwu Oct 10, 2025
d01a3e5
updated disabled test names
tianleiwu Oct 11, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 1 addition & 1 deletion .github/workflows/publish-csharp-apidocs.yml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -20,7 +20,7 @@ permissions:

jobs:
build:
runs-on: ["self-hosted", "1ES.Pool=onnxruntime-github-vs2022-mms"]
runs-on: ["self-hosted", "1ES.Pool=onnxruntime-github-vs2022-latest"]
env:
DOCFXVERSION: 2.62.2
steps:
Expand Down
26 changes: 13 additions & 13 deletions .github/workflows/windows_cuda.yml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -19,7 +19,7 @@ concurrency:
jobs:
build:
name: Windows GPU CUDA CI Pipeline
runs-on: ["self-hosted", "1ES.Pool=onnxruntime-github-vs2022-mms"]
runs-on: ["self-hosted", "1ES.Pool=onnxruntime-github-vs2022-latest"]
steps:
- uses: actions/checkout@v5
with:
Expand All @@ -41,20 +41,20 @@ jobs:
working-directory: ${{ github.workspace }}
shell: cmd

- name: Download CUDA SDK v12.2
- name: Download CUDA SDK v12.8
working-directory: ${{ runner.temp }}
run: |
azcopy.exe cp --recursive "https://lotusscus.blob.core.windows.net/models/cuda_sdk/v12.2" .
azcopy.exe cp --recursive "https://lotusscus.blob.core.windows.net/models/cuda_sdk/v12.8" .
dir
shell: pwsh

- name: Add CUDA to PATH
shell: powershell
run: |
Write-Host "Adding CUDA to PATH"
Write-Host "CUDA Path: $env:RUNNER_TEMP\v12.2\bin"
Add-Content -Path $env:GITHUB_PATH -Value "$env:RUNNER_TEMP\v12.2\bin"
Add-Content -Path $env:GITHUB_PATH -Value "$env:RUNNER_TEMP\v12.2\extras\CUPTI\lib64"
Write-Host "CUDA Path: $env:RUNNER_TEMP\v12.8\bin"
Add-Content -Path $env:GITHUB_PATH -Value "$env:RUNNER_TEMP\v12.8\bin"
Add-Content -Path $env:GITHUB_PATH -Value "$env:RUNNER_TEMP\v12.8\extras\CUPTI\lib64"

- uses: actions/setup-node@v5
with:
Expand Down Expand Up @@ -111,7 +111,7 @@ jobs:
exit $lastExitCode
}
# Execute the build process
python.exe ${{ github.workspace }}\tools\ci_build\build.py --update --build --config RelWithDebInfo --build_dir build --skip_submodule_sync --build_csharp --parallel --use_binskim_compliant_compile_flags --cmake_generator "Visual Studio 17 2022" --build_shared_lib --build_wheel --build_java --use_cuda --cuda_home="$env:RUNNER_TEMP\v12.2" --enable_cuda_profiling --use_vcpkg --use_vcpkg_ms_internal_asset_cache --enable_transformers_tool_test --cmake_extra_defines CMAKE_CUDA_ARCHITECTURES=86 --cmake_extra_defines onnxruntime_ENABLE_CUDA_EP_INTERNAL_TESTS=ON
python.exe ${{ github.workspace }}\tools\ci_build\build.py --update --build --config RelWithDebInfo --build_dir build --skip_submodule_sync --build_csharp --parallel --use_binskim_compliant_compile_flags --cmake_generator "Visual Studio 17 2022" --build_shared_lib --build_wheel --build_java --use_cuda --cuda_home="$env:RUNNER_TEMP\v12.8" --enable_cuda_profiling --use_vcpkg --use_vcpkg_ms_internal_asset_cache --enable_transformers_tool_test --cmake_extra_defines CMAKE_CUDA_ARCHITECTURES=86 --cmake_extra_defines onnxruntime_ENABLE_CUDA_EP_INTERNAL_TESTS=ON
if ($lastExitCode -ne 0) {
exit $lastExitCode
}
Expand Down Expand Up @@ -188,20 +188,20 @@ jobs:
working-directory: ${{ github.workspace }}
shell: cmd

- name: Download CUDA SDK v12.2
- name: Download CUDA SDK v12.8
working-directory: ${{ runner.temp }}
run: |
azcopy.exe cp --recursive "https://lotusscus.blob.core.windows.net/models/cuda_sdk/v12.2" .
azcopy.exe cp --recursive "https://lotusscus.blob.core.windows.net/models/cuda_sdk/v12.8" .
dir
shell: pwsh

- name: Add CUDA to PATH
shell: powershell
run: |
Write-Host "Adding CUDA to PATH"
Write-Host "CUDA Path: $env:RUNNER_TEMP\v12.2\bin"
Add-Content -Path $env:GITHUB_PATH -Value "$env:RUNNER_TEMP\v12.2\bin"
Add-Content -Path $env:GITHUB_PATH -Value "$env:RUNNER_TEMP\v12.2\extras\CUPTI\lib64"
Write-Host "CUDA Path: $env:RUNNER_TEMP\v12.8\bin"
Add-Content -Path $env:GITHUB_PATH -Value "$env:RUNNER_TEMP\v12.8\bin"
Add-Content -Path $env:GITHUB_PATH -Value "$env:RUNNER_TEMP\v12.8\extras\CUPTI\lib64"

- name: Set OnnxRuntimeBuildDirectory
shell: pwsh
Expand All @@ -227,7 +227,7 @@ jobs:
exit $lastExitCode
}

python.exe ${{ github.workspace }}\tools\ci_build\build.py --test --config RelWithDebInfo --build_dir build --skip_submodule_sync --build_csharp --parallel --use_binskim_compliant_compile_flags --cmake_generator "Visual Studio 17 2022" --build_shared_lib --build_wheel --build_java --use_cuda --cuda_home="$env:RUNNER_TEMP\v12.2" --enable_cuda_profiling --use_vcpkg --use_vcpkg_ms_internal_asset_cache --enable_transformers_tool_test --cmake_extra_defines CMAKE_CUDA_ARCHITECTURES=86 --cmake_extra_defines onnxruntime_ENABLE_CUDA_EP_INTERNAL_TESTS=ON
python.exe ${{ github.workspace }}\tools\ci_build\build.py --test --config RelWithDebInfo --build_dir build --skip_submodule_sync --build_csharp --parallel --use_binskim_compliant_compile_flags --cmake_generator "Visual Studio 17 2022" --build_shared_lib --build_wheel --build_java --use_cuda --cuda_home="$env:RUNNER_TEMP\v12.8" --enable_cuda_profiling --use_vcpkg --use_vcpkg_ms_internal_asset_cache --enable_transformers_tool_test --cmake_extra_defines CMAKE_CUDA_ARCHITECTURES=86 --cmake_extra_defines onnxruntime_ENABLE_CUDA_EP_INTERNAL_TESTS=ON
if ($lastExitCode -ne 0) {
exit $lastExitCode
}
Expand Down
2 changes: 1 addition & 1 deletion .github/workflows/windows_openvino.yml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -18,7 +18,7 @@ concurrency:
jobs:
BUILD_OPENVINO_EP:
name: Windows OpenVINO CI Pipeline
runs-on: ["self-hosted", "1ES.Pool=onnxruntime-github-vs2022-mms"]
runs-on: ["self-hosted", "1ES.Pool=onnxruntime-github-vs2022-latest"]
timeout-minutes: 240
env:
AZCOPY_AUTO_LOGIN_TYPE: MSI
Expand Down
2 changes: 1 addition & 1 deletion .github/workflows/windows_qnn_x64.yml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -18,7 +18,7 @@ concurrency:
jobs:
build_test_qnn_ep:
name: Windows x64 QNN CI Pipeline (${{ matrix.QnnLibKind }})
runs-on: ["self-hosted", "1ES.Pool=onnxruntime-github-vs2022-mms"]
runs-on: ["self-hosted", "1ES.Pool=onnxruntime-github-vs2022-latest"]
timeout-minutes: 120
strategy:
matrix:
Expand Down
26 changes: 13 additions & 13 deletions .github/workflows/windows_tensorrt.yml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -19,7 +19,7 @@ concurrency:
jobs:
build:
name: Windows GPU TensorRT CI Pipeline
runs-on: ["self-hosted", "1ES.Pool=onnxruntime-github-vs2022-mms"]
runs-on: ["self-hosted", "1ES.Pool=onnxruntime-github-vs2022-latest"]
steps:
- uses: actions/checkout@v5
with:
Expand All @@ -41,10 +41,10 @@ jobs:
working-directory: ${{ github.workspace }}
shell: cmd

- name: Download CUDA SDK v12.2
- name: Download CUDA SDK v12.8
working-directory: ${{ runner.temp }}
run: |
azcopy.exe cp --recursive "https://lotusscus.blob.core.windows.net/models/cuda_sdk/v12.2" .
azcopy.exe cp --recursive "https://lotusscus.blob.core.windows.net/models/cuda_sdk/v12.8" .
dir
shell: pwsh

Expand All @@ -56,9 +56,9 @@ jobs:
shell: powershell
run: |
Write-Host "Adding CUDA to PATH"
Write-Host "CUDA Path: $env:RUNNER_TEMP\v12.2\bin"
Add-Content -Path $env:GITHUB_PATH -Value "$env:RUNNER_TEMP\v12.2\bin"
Add-Content -Path $env:GITHUB_PATH -Value "$env:RUNNER_TEMP\v12.2\extras\CUPTI\lib64"
Write-Host "CUDA Path: $env:RUNNER_TEMP\v12.8\bin"
Add-Content -Path $env:GITHUB_PATH -Value "$env:RUNNER_TEMP\v12.8\bin"
Add-Content -Path $env:GITHUB_PATH -Value "$env:RUNNER_TEMP\v12.8\extras\CUPTI\lib64"
Add-Content -Path $env:GITHUB_PATH -Value "$env:RUNNER_TEMP\TensorRT-10.9.0.34.Windows10.x86_64.cuda-12.8\lib"

- uses: actions/setup-node@v5
Expand Down Expand Up @@ -116,7 +116,7 @@ jobs:
exit $lastExitCode
}
# Execute the build process
python ${{ github.workspace }}\tools\ci_build\build.py --config RelWithDebInfo --parallel --use_binskim_compliant_compile_flags --build_dir build --skip_submodule_sync --build_shared_lib --build --update --cmake_generator "Visual Studio 17 2022" --build_wheel --enable_onnx_tests --use_tensorrt --tensorrt_home="${{ runner.temp }}\TensorRT-10.9.0.34.Windows10.x86_64.cuda-12.8" --cuda_home="${{ runner.temp }}\v12.2" --use_vcpkg --use_vcpkg_ms_internal_asset_cache --cmake_extra_defines CMAKE_CUDA_ARCHITECTURES=86
python ${{ github.workspace }}\tools\ci_build\build.py --config RelWithDebInfo --parallel --use_binskim_compliant_compile_flags --build_dir build --skip_submodule_sync --build_shared_lib --build --update --cmake_generator "Visual Studio 17 2022" --build_wheel --enable_onnx_tests --use_tensorrt --tensorrt_home="${{ runner.temp }}\TensorRT-10.9.0.34.Windows10.x86_64.cuda-12.8" --cuda_home="${{ runner.temp }}\v12.8" --use_vcpkg --use_vcpkg_ms_internal_asset_cache --cmake_extra_defines CMAKE_CUDA_ARCHITECTURES=86
if ($lastExitCode -ne 0) {
exit $lastExitCode
}
Expand Down Expand Up @@ -193,10 +193,10 @@ jobs:
working-directory: ${{ github.workspace }}
shell: cmd

- name: Download CUDA SDK v12.2
- name: Download CUDA SDK v12.8
working-directory: ${{ runner.temp }}
run: |
azcopy.exe cp --recursive "https://lotusscus.blob.core.windows.net/models/cuda_sdk/v12.2" .
azcopy.exe cp --recursive "https://lotusscus.blob.core.windows.net/models/cuda_sdk/v12.8" .
dir
shell: pwsh

Expand All @@ -208,9 +208,9 @@ jobs:
shell: powershell
run: |
Write-Host "Adding CUDA to PATH"
Write-Host "CUDA Path: $env:RUNNER_TEMP\v12.2\bin"
Add-Content -Path $env:GITHUB_PATH -Value "$env:RUNNER_TEMP\v12.2\bin"
Add-Content -Path $env:GITHUB_PATH -Value "$env:RUNNER_TEMP\v12.2\extras\CUPTI\lib64"
Write-Host "CUDA Path: $env:RUNNER_TEMP\v12.8\bin"
Add-Content -Path $env:GITHUB_PATH -Value "$env:RUNNER_TEMP\v12.8\bin"
Add-Content -Path $env:GITHUB_PATH -Value "$env:RUNNER_TEMP\v12.8\extras\CUPTI\lib64"
Add-Content -Path $env:GITHUB_PATH -Value "$env:RUNNER_TEMP\TensorRT-10.9.0.34.Windows10.x86_64.cuda-12.8\lib"

- name: Set OnnxRuntimeBuildDirectory
Expand All @@ -237,7 +237,7 @@ jobs:
exit $lastExitCode
}

python ${{ github.workspace }}\tools\ci_build\build.py --config RelWithDebInfo --parallel --use_binskim_compliant_compile_flags --build_dir build --skip_submodule_sync --build_shared_lib --test --cmake_generator "Visual Studio 17 2022" --build_wheel --enable_onnx_tests --use_tensorrt --tensorrt_home="${{ runner.temp }}\TensorRT-10.9.0.34.Windows10.x86_64.cuda-12.8" --cuda_home="${{ runner.temp }}\v12.2" --use_vcpkg --use_vcpkg_ms_internal_asset_cache --cmake_extra_defines CMAKE_CUDA_ARCHITECTURES=86
python ${{ github.workspace }}\tools\ci_build\build.py --config RelWithDebInfo --parallel --use_binskim_compliant_compile_flags --build_dir build --skip_submodule_sync --build_shared_lib --test --cmake_generator "Visual Studio 17 2022" --build_wheel --enable_onnx_tests --use_tensorrt --tensorrt_home="${{ runner.temp }}\TensorRT-10.9.0.34.Windows10.x86_64.cuda-12.8" --cuda_home="${{ runner.temp }}\v12.8" --use_vcpkg --use_vcpkg_ms_internal_asset_cache --cmake_extra_defines CMAKE_CUDA_ARCHITECTURES=86
if ($lastExitCode -ne 0) {
exit $lastExitCode
}
Expand Down
2 changes: 1 addition & 1 deletion .github/workflows/windows_x64_debug_build_x64_debug.yml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -13,7 +13,7 @@ concurrency:

jobs:
build_x64_debug:
runs-on: ["self-hosted", "1ES.Pool=onnxruntime-github-vs2022-mms"]
runs-on: ["self-hosted", "1ES.Pool=onnxruntime-github-vs2022-latest"]
timeout-minutes: 300

steps:
Expand Down
2 changes: 1 addition & 1 deletion .github/workflows/windows_x64_release_build_x64_release.yml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -13,7 +13,7 @@ concurrency:

jobs:
build_x64_release:
runs-on: ["self-hosted", "1ES.Pool=onnxruntime-github-vs2022-mms"]
runs-on: ["self-hosted", "1ES.Pool=onnxruntime-github-vs2022-latest"]
timeout-minutes: 300

steps:
Expand Down
2 changes: 1 addition & 1 deletion ...flows/windows_x64_release_ep_generic_interface_build_x64_release_ep_generic_interface.yml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -13,7 +13,7 @@ concurrency:

jobs:
build_x64_release_ep_generic_interface:
runs-on: ["self-hosted", "1ES.Pool=onnxruntime-github-vs2022-mms"]
runs-on: ["self-hosted", "1ES.Pool=onnxruntime-github-vs2022-latest"]
timeout-minutes: 300

steps:
Expand Down
2 changes: 1 addition & 1 deletion .github/workflows/windows_x64_release_vitisai_build_x64_release.yml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -13,7 +13,7 @@ concurrency:

jobs:
build_x64_release_vitisai:
runs-on: ["self-hosted", "1ES.Pool=onnxruntime-github-vs2022-mms"]
runs-on: ["self-hosted", "1ES.Pool=onnxruntime-github-vs2022-latest"]
timeout-minutes: 300

steps:
Expand Down
2 changes: 1 addition & 1 deletion .github/workflows/windows_x64_release_xnnpack.yml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -13,7 +13,7 @@ concurrency:

jobs:
build_x64_release_xnnpack:
runs-on: ["self-hosted", "1ES.Pool=onnxruntime-github-vs2022-mms"]
runs-on: ["self-hosted", "1ES.Pool=onnxruntime-github-vs2022-latest"]
timeout-minutes: 300

steps:
Expand Down
2 changes: 1 addition & 1 deletion .github/workflows/windows_x86.yml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -13,7 +13,7 @@ concurrency:

jobs:
build_x86_release:
runs-on: ["self-hosted", "1ES.Pool=onnxruntime-github-vs2022-mms"]
runs-on: ["self-hosted", "1ES.Pool=onnxruntime-github-vs2022-latest"]
timeout-minutes: 300

steps:
Expand Down
4 changes: 2 additions & 2 deletions cmake/onnxruntime_providers_cuda.cmake
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -182,8 +182,8 @@

# Since CUDA 12.8, compiling diagnostics become stricter
if (CMAKE_CUDA_COMPILER_VERSION VERSION_GREATER_EQUAL 12.8)
target_compile_options(${target} PRIVATE "$<$<COMPILE_LANGUAGE:CUDA>:--relocatable-device-code=true>")
set_target_properties(${target} PROPERTIES CUDA_SEPARABLE_COMPILATION ON)
target_compile_options(${target} PRIVATE "$<$<COMPILE_LANGUAGE:CUDA>:--static-global-template-stub=false>")

if (MSVC)
target_compile_options(${target} PRIVATE "$<$<COMPILE_LANGUAGE:CUDA>:SHELL:-Xcompiler /wd4505>")
endif()
Expand Down
2 changes: 1 addition & 1 deletion onnxruntime/test/contrib_ops/gather_block_quantized_op_test.cc
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -82,7 +82,7 @@ void CheckDataAndShape(const std::vector<T>& data, const std::vector<int64_t>& s

ORT_ENFORCE(static_cast<int64_t>(data.size()) == total_elements, "Data size does not match the shape",
"Data size: ", data.size(), ", Expected size: ", total_elements,
", Shape: ", VectorToString(shape), " Name:", name, " Type:", typeid(T).name());
", Shape: ", VectorToString(shape), " Name:", name);
}

// Combinations: types, gather_axis, quantize_axis, block_size, indices, scale shape vs data shape
Expand Down
9 changes: 8 additions & 1 deletion onnxruntime/test/providers/cpu/model_tests.cc
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -678,7 +678,14 @@ ::std::vector<::std::basic_string<ORTCHAR_T>> GetParameterStrings() {
ORT_TSTR("fp16_coreml_FNS-Candy"),
ORT_TSTR("fp16_test_tiny_yolov2"),
ORT_TSTR("fp16_test_shufflenet"),
ORT_TSTR("keras2coreml_SimpleRNN_ImageNet")};
ORT_TSTR("keras2coreml_SimpleRNN_ImageNet"),
// models from model zoo. #26274: cuDNN frontend no valid engine
ORT_TSTR("YOLOv3"),
ORT_TSTR("YOLOv3-12"),
ORT_TSTR("YOLOv4"),
ORT_TSTR("SSD-MobilenetV1"),
ORT_TSTR("SSD-MobilenetV1-12")};

// For ROCm EP, also disable the following tests due to flakiness,
// mainly with precision issue and random memory access fault.
static const ORTCHAR_T* rocm_disabled_tests[] = {ORT_TSTR("bvlc_alexnet"),
Expand Down
2 changes: 1 addition & 1 deletion onnxruntime/test/providers/cpu/tensor/cast_op_test.cc
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1477,7 +1477,7 @@ template <typename F4>
void CastOpTestFloatFloat4(std::vector<int64_t> shape,
std::vector<float> float_data,
bool is_fp4_input = false) {
size_t num_pairs = float_data.size() / 2;
int num_pairs = static_cast<int>(float_data.size()) / 2;
int num_fp4_elements = static_cast<int>((float_data.size() + 1) / 2);
bool is_odd_count = (float_data.size() % 2 != 0);

Expand Down
2 changes: 1 addition & 1 deletion tools/ci_build/github/azure-pipelines/build-perf-test-binaries-pipeline.yml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -31,5 +31,5 @@ stages:
machine_pool: 'onnxruntime-Ubuntu2404-AMD-CPU'
extra_build_arg: ''
cmake_build_type: Release
cuda_version: 12.2
cuda_version: 12.8
docker_base_image: onnxruntimebuildcache.azurecr.io/internal/azureml/onnxruntime/build/cuda12_x64_almalinux8_gcc12:20250714.2
6 changes: 3 additions & 3 deletions tools/ci_build/github/azure-pipelines/c-api-noopenmp-packaging-pipelines.yml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -73,12 +73,12 @@ variables:
- name: ReleaseVersionSuffix
value: ''
- name: win_trt_version
value: 12.2
value: 12.8

- name: win_trt_home
value: $(Agent.TempDirectory)\${{ variables.win_trt_folder_cuda12 }}
- name: win_cuda_home
value: $(Agent.TempDirectory)\v12.2
value: $(Agent.TempDirectory)\v12.8
extends:
# The pipeline extends the 1ES PT which will inject different SDL and compliance tasks.
# For non-production pipelines, use "Unofficial" as defined below.
Expand Down Expand Up @@ -142,7 +142,7 @@ extends:

- template: stages/nuget-combine-cuda-stage.yml
parameters:
CudaVersion: 12.2
CudaVersion: 12.8
RunOnnxRuntimeTests: ${{ parameters.RunOnnxRuntimeTests }}
UseIncreasedTimeoutForTests: ${{ parameters.UseIncreasedTimeoutForTests }}
win_trt_home: ${{ variables.win_trt_home }}
Expand Down
12 changes: 6 additions & 6 deletions tools/ci_build/github/azure-pipelines/c-api-noopenmp-test-pipelines.yml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -127,7 +127,7 @@ stages:
NugetPackageName: 'Microsoft.ML.OnnxRuntime.Gpu'
ArtifactSuffix: 'GPU'
StageSuffix: 'GPU'
CudaVersion: 12.2
CudaVersion: 12.8

- template: nuget/templates/test_win.yml
parameters:
Expand All @@ -136,15 +136,15 @@ stages:
ArtifactSuffix: 'GPU'
StageSuffix: 'GPU'
MoreSuffix: '_Windows'
CudaVersion: 12.2
CudaVersion: 12.8

- template: nuget/templates/test_linux.yml
parameters:
AgentPool: Onnxruntime-Linux-GPU-A10
ArtifactSuffix: 'GPU'
StageSuffix: 'GPU'
NugetPackageName: 'Microsoft.ML.OnnxRuntime.Gpu'
CudaVersion: 12.2
CudaVersion: 12.8

- template: nuget/templates/test_linux.yml
parameters:
Expand All @@ -153,7 +153,7 @@ stages:
StageSuffix: 'GPU'
MoreSuffix: '_Linux'
NugetPackageName: 'Microsoft.ML.OnnxRuntime.Gpu.Linux'
CudaVersion: 12.2
CudaVersion: 12.8



Expand Down Expand Up @@ -202,7 +202,7 @@ stages:

- template: templates/jobs/download_win_gpu_library.yml
parameters:
CudaVersion: 12.2
CudaVersion: 12.8
DownloadCUDA: true
DownloadTRT: true

Expand Down Expand Up @@ -257,7 +257,7 @@ stages:

- template: templates/jobs/download_win_gpu_library.yml
parameters:
CudaVersion: 12.2
CudaVersion: 12.8
DownloadCUDA: true
DownloadTRT: true

Expand Down
10 changes: 5 additions & 5 deletions tools/ci_build/github/azure-pipelines/cuda-packaging-pipeline.yml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -48,9 +48,9 @@ parameters:
- name: CudaVersion
displayName: CUDA version
type: string
default: '12.2'
default: '12.8'
values:
- 12.2
- 12.8

variables:
- template: templates/common-variables.yml
Expand All @@ -59,13 +59,13 @@ variables:
- name: win_trt_home
${{ if eq(parameters.CudaVersion, '11.8') }}:
value: $(Agent.TempDirectory)\${{ variables.win_trt_folder_cuda11 }}
${{ if eq(parameters.CudaVersion, '12.2') }}:
${{ if eq(parameters.CudaVersion, '12.8') }}:
value: $(Agent.TempDirectory)\${{ variables.win_trt_folder_cuda12 }}
- name: win_cuda_home
${{ if eq(parameters.CudaVersion, '11.8') }}:
value: $(Agent.TempDirectory)\v11.8
${{ if eq(parameters.CudaVersion, '12.2') }}:
value: $(Agent.TempDirectory)\v12.2
${{ if eq(parameters.CudaVersion, '12.8') }}:
value: $(Agent.TempDirectory)\v12.8

resources:
repositories:
Expand Down
Loading
Loading