Skip to content
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.

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

Add tensorrt test workflow #3266

Open
wants to merge 20 commits into
base: lluo/decomposition_upstream_change
Choose a base branch
from

Conversation

lanluo-nvidia
Copy link
Collaborator

@lanluo-nvidia lanluo-nvidia commented Oct 28, 2024

Description

Added the tensorrt test workflow:
example run:
https://github.com/pytorch/TensorRT/actions/runs/11675576873/job/32510381587?pr=3266

Type of change

Please delete options that are not relevant and/or add your own.

  • Bug fix (non-breaking change which fixes an issue)
  • New feature (non-breaking change which adds functionality)
  • Breaking change (fix or feature that would cause existing functionality to not work as expected)
  • This change requires a documentation update

Checklist:

  • My code follows the style guidelines of this project (You can use the linters)
  • I have performed a self-review of my own code
  • I have commented my code, particularly in hard-to-understand areas and hacks
  • I have made corresponding changes to the documentation
  • I have added tests to verify my fix or my feature
  • New and existing unit tests pass locally with my changes
  • I have added the relevant labels to my PR in so that relevant reviewers are notified

@lanluo-nvidia lanluo-nvidia added the WIP Work is in progress, pull request should not be merged yet label Oct 28, 2024
@lanluo-nvidia lanluo-nvidia self-assigned this Oct 28, 2024
@lanluo-nvidia lanluo-nvidia removed the WIP Work is in progress, pull request should not be merged yet label Nov 5, 2024
@lanluo-nvidia lanluo-nvidia changed the title Add tensorrt test workflow (WIP) Add tensorrt test workflow Nov 5, 2024
@lanluo-nvidia lanluo-nvidia marked this pull request as ready for review November 5, 2024 03:31
if [[ ${TENSORRT_VERSION} != "" ]]; then
# this is the upgraded TensorRT version, replace current tensorrt version to the upgrade tensorRT version in the pyproject.toml
current_version=$(cat dev_dep_versions.yml | grep __tensorrt_version__ | sed 's/__tensorrt_version__: //g' | sed 's/"//g')
sed -i -e "s/tensorrt-cu12==${current_version}/tensorrt-cu12==${TENSORRT_VERSION}/g" \
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Is CUDA 11.8 still in the support matrix?

Copy link
Collaborator Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

yes, cuda 11.8 still in the support matrix in tensorRT10.6.0

-e "s/tensorrt-cu12-libs==${current_version}/tensorrt-cu12-libs==${TENSORRT_VERSION}/g" \
pyproject.toml
fi

if [[ "${CU_VERSION::4}" < "cu12" ]]; then
# replace dependencies from tensorrt-cu12-bindings/libs to tensorrt-cu11-bindings/libs
sed -i -e "s/tensorrt-cu12==/tensorrt-${CU_VERSION::4}==/g" \
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Or does it get handled here?

Copy link
Collaborator Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

yes, the cuda 11.8 is handled here.

Copy link
Collaborator

@narendasan narendasan left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Mostly seems fine

@github-actions github-actions bot added component: lowering Issues re: The lowering / preprocessing passes component: api [Python] Issues re: Python API component: dynamo Issues relating to the `torch.compile` or `torch._dynamo.export` paths labels Nov 12, 2024
@lanluo-nvidia lanluo-nvidia changed the base branch from main to lluo/decomposition_upstream_change November 12, 2024 22:44
@narendasan
Copy link
Collaborator

Seems like 12.1 got dropped is that us or PyTorch?

@lanluo-nvidia
Copy link
Collaborator Author

Seems like 12.1 got dropped is that us or PyTorch?

@narendasan
yes, cuda 12.1 is deprecated by pytorch, it was merged in just a few hours ago.
pytorch/test-infra#5904

Very wierd is that cuda11.8 is still supported, they only deprecated cuda12.1

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
cla signed component: api [Python] Issues re: Python API component: dynamo Issues relating to the `torch.compile` or `torch._dynamo.export` paths component: lowering Issues re: The lowering / preprocessing passes
Projects
None yet
Development

Successfully merging this pull request may close these issues.

3 participants