Migrate A100 CUDA CI jobs to OSDC runners#20212
Merged
Merged
Conversation
Move the A100-dependent jobs in cuda.yml (export-model-cuda-artifact, test-model-cuda-e2e) and cuda-perf.yml (export-models, benchmark-cuda) from pytorch/test-infra linux_job_v2 (AWS) to linux_job_v3 (OSDC/ARC). Runner labels are remapped per pytorch/.github/arc.yaml: linux.aws.a100 -> mt-l-x86iavx512-11-125-a100 and the A10G fallback linux.g5.4xlarge.nvidia.gpu -> mt-l-x86aavx2-29-113-a10g. Jobs that never run on A100 stay on linux_job_v2 / linux.g5.4xlarge.nvidia.gpu. Authored with Claude Code.
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/20212
Note: Links to docs will display an error until the docs builds have been completed. ❌ 1 New Failure, 36 PendingAs of commit 1b44041 with merge base 129c687 ( NEW FAILURE - The following job has failed:
This comment was automatically generated by Dr. CI and updates every 15 minutes. |
meta-cla
Bot
added
the
CLA Signed
This PR needs a
|
OSDC runners can't reach the public PyPI CDN that download.pytorch.org's transitive deps resolve to, so the torch install in install_requirements.py fails fetching e.g. sympy from files.pythonhosted.org. Pre-install torch's pure-python deps from the in-cluster pypi-cache and clear PIP_EXTRA_INDEX_URL in the four migrated CUDA jobs, mirroring the torchtitan/ao OSDC workaround. Authored with Claude Code.
The example-deps install (torchvision==0.27.0 torchaudio==2.11.0) pulls pillow, which still resolved from files.pythonhosted.org and failed on OSDC. Add pillow to the pre-installed pure-python deps, matching the torchao OSDC list. Authored with Claude Code.
huydhn
temporarily deployed
to
upload-benchmark-results
GitHub Actions
Inactive
The examples install pulls datasets==3.6.0, which pins fsspec[http]<=2025.3.0. The unpinned pre-installed fsspec was newer, so pip tried to downgrade it via download.pytorch.org's pythonhosted link, which OSDC can't reach. Pre-install fsspec at <=2025.3.0 so only-if-needed leaves it in place. Authored with Claude Code.
Gasoonjia
approved these changes
Jun 11, 2026
huydhn
temporarily deployed
to
upload-benchmark-results
GitHub Actions
Inactive
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Moves the A100-dependent CUDA CI jobs from
pytorch/test-infralinux_job_v2(AWS) tolinux_job_v3(OSDC/ARC), and remaps their runner labels perpytorch/.github/arc.yaml.Migrated jobs (now on OSDC /
linux_job_v3)cuda.yml:export-model-cuda-artifact,test-model-cuda-e2ecuda-perf.yml:export-models,benchmark-cudaRunner label mapping
linux.aws.a100mt-l-x86iavx512-11-125-a100linux.g5.4xlarge.nvidia.gpu(A10G fallback branch)mt-l-x86aavx2-29-113-a10gThe A10G fallback branch in each conditional runner expression had to move to an OSDC label too, since
linux_job_v3requires ARC labels and that branch belongs to the same A100-dependent jobs.Left unchanged
Jobs that never run on A100 stay on
linux_job_v2/linux.g5.4xlarge.nvidia.gpu:test-cuda-builds,test-models-cuda,unittest-cuda,test-cuda-pybind.linux_job_v3resolves the docker image and--gpus allidentically to v2 for these jobs (none setdocker-image), so build/runtime behavior is unchanged.Authored with Claude Code.