Two improvements in pipelines (#20449)

### Description
1. Update the image name to avoid docker image wouldn't be overwrite.
there was an mistake that variables.CUDA_VERSION_MAJOR is always empty

14fcf0a52d/tools/ci_build/github/azure-pipelines/stages/nuget-linux-cuda-packaging-stage.yml (L120)
3. set one artifact name as variable to make the job rerunnable



### Motivation and Context
<!-- - Why is this change required? What problem does it solve?
- If it fixes an open issue, please link to the issue here. -->
This commit is contained in:
Yi Zhang 2024-04-25 10:15:40 +08:00 committed by GitHub
parent 218b6b0a73
commit e5947f5729
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
2 changed files with 11 additions and 4 deletions

View file

@ -224,10 +224,12 @@ stages:
workingDirectory: $(Build.SourcesDirectory)
# For verification we will check the generated image looks .
# Because the artifact isn't used by other jobs, we set the artifact name to a varabile value.
# So the job can be rerun without the exception that artifact has been published.
- task: PublishPipelineArtifact@0
displayName: 'Publish code coverage report'
displayName: 'Publish Generated Image Artifact'
inputs:
artifactName: "Generated-Image"
artifactName: Generated-Image-$(System.JobAttempt)
targetPath: '$(GenerateImage_DIR)'
- task: Cache@2

View file

@ -88,7 +88,12 @@ stages:
workspace:
clean: all
pool: 'Onnxruntime-Linux-GPU'
variables:
- name: CUDA_VERSION_MAJOR
${{ if eq(parameters.CudaVersion, '11.8') }}:
value: '11'
${{ if eq(parameters.CudaVersion, '12.2') }}:
value: '12'
steps:
- checkout: self # due to checkout multiple repos, the root directory is $(Build.SourcesDirectory)/onnxruntime
submodules: false
@ -180,4 +185,4 @@ stages:
artifactName: 'onnxruntime-linux-x64-gpu'
- template: ../templates/component-governance-component-detection-steps.yml
parameters:
condition: 'succeeded'
condition: 'succeeded'