onnxruntime/tools/ci_build/github/azure-pipelines/py-cuda-packaging-pipeline.yml
Yi Zhang cd6d3aea45
Refactor Python CUDA packaging pipeline to fix random hangs in building (#19989)
### Description
1. Move building on CPU machine.
2. Optimize the pipeline
3. Since there isn't official ONNX package for python 12, the python 12
test stage uses the packages built with ONNX source in build stage.


### Motivation and Context
1. Resolve the random hang in compilation
4. Save a lot of GPU resources.

---------
2024-03-22 09:16:00 +08:00

50 lines
1.2 KiB
YAML

trigger: none
parameters:
- name: enable_linux_gpu
type: boolean
default: true
- name: enable_windows_gpu
type: boolean
default: true
- name: cmake_build_type
type: string
default: 'Release'
values:
- Debug
- Release
- RelWithDebInfo
- MinSizeRel
- name: cuda_version
type: string
default: '12.2'
values:
- 11.8
- 12.2
- name: SpecificArtifact
displayName: Use Specific Artifact
type: boolean
default: false
- name: BuildId
displayName: Specific Artifact's BuildId
type: string
default: '0'
resources:
repositories:
- repository: manylinux
type: Github
endpoint: Microsoft
name: pypa/manylinux
ref: 5eda9aded5462201e6310105728d33016e637ea7
stages:
- template: stages/py-cuda-packaging-stage.yml
parameters:
enable_linux_gpu: ${{ parameters.enable_linux_gpu }}
enable_windows_gpu: ${{ parameters.enable_windows_gpu }}
cmake_build_type: ${{ parameters.cmake_build_type }}
cuda_version: ${{ parameters.cuda_version }}
SpecificArtifact: ${{ parameters.SpecificArtifact }}
BuildId: ${{ parameters.BuildId }}