Fix cuda 12.9 aarch64 GPU builds. Update CUDA_STABLE variable. #157641

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

Jump to bottom

Merged

atalman merged 1 commit into release/2.8 from cherry-pick-157630-by-pytorch_bot_bot_

Jul 4, 2025

Collaborator

pytorchbot commented Jul 4, 2025

This contains 2 fixes that required in main and will need to be cherry-picked to Release 2.8 branch:

The PR Add CUDA 12.9.1 sbsa nightly binaries #155819 missed to include triton change.
CUDA STABLE variable needs to be set to 12.8. Updating CUDA stable updates full static build


          Fix cuda 12.9 aarch64 GPU builds. Update CUDA_STABLE variable. (#157630)

d691716

This contains 2 fixes that required in main and will need to be cherry-picked to Release 2.8 branch:
1. The PR #155819 missed to include triton change.
2. CUDA STABLE variable needs to be set to 12.8. Updating CUDA stable updates full static build

Pull Request resolved: #157630
Approved by: https://github.com/Skylion007, https://github.com/jeanschmidt

(cherry picked from commit 7275f28)

pytorchbot requested a review from a team as a code owner

July 4, 2025 18:46

pytorchbot mentioned this pull request

[v.2.8.0] Release Tracker #156745

Open

pytorch-bot bot added the topic: not user facing label

pytorchbot mentioned this pull request

Fix cuda 12.9 aarch64 GPU builds. Update CUDA_STABLE variable. #157630

Closed

pytorch-bot bot commented Jul 4, 2025 •

edited

Loading

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/157641

📄 Preview Python docs built from this PR
📄 Preview C++ docs built from this PR
❓ Need help or want to give feedback on the CI? Visit the bot commands wiki or our office hours

Note: Links to docs will display an error until the docs builds have been completed.

❌ 50 Cancelled Jobs, 17 Unrelated Failures

As of commit d691716 with merge base 3a7ff82 ():

CANCELLED JOBS - The following jobs were cancelled. Please retry:

linux-aarch64-binary-manywheel / manywheel-py3_10-cuda-aarch64-12_9-build / build (gh)
##[error]The operation was canceled.
linux-aarch64-binary-manywheel / manywheel-py3_11-cuda-aarch64-12_9-build / build (gh)
##[error]The operation was canceled.
linux-aarch64-binary-manywheel / manywheel-py3_12-cuda-aarch64-12_9-build / build (gh)
##[error]The operation was canceled.
linux-aarch64-binary-manywheel / manywheel-py3_13-cuda-aarch64-12_9-build / build (gh)
##[error]The operation was canceled.
linux-aarch64-binary-manywheel / manywheel-py3_13t-cuda-aarch64-12_9-build / build (gh)
##[error]The operation was canceled.
linux-aarch64-binary-manywheel / manywheel-py3_9-cuda-aarch64-12_9-build / build (gh)
##[error]The operation was canceled.
linux-binary-libtorch / libtorch-rocm6_3-shared-with-deps-release-build / build (gh)
##[error]The operation was canceled.
linux-binary-libtorch / libtorch-rocm6_3-shared-with-deps-release-test (gh)
linux-binary-manywheel / manywheel-py3_10-rocm6_3-test (gh)
##[error]The operation was canceled.
linux-binary-manywheel / manywheel-py3_11-rocm6_3-build / build (gh)
##[error]The operation was canceled.
linux-binary-manywheel / manywheel-py3_11-rocm6_3-test (gh)
linux-binary-manywheel / manywheel-py3_12-rocm6_3-build / build (gh)
##[error]The operation was canceled.
linux-binary-manywheel / manywheel-py3_12-rocm6_3-test (gh)
linux-binary-manywheel / manywheel-py3_12-rocm6_4-build / build (gh)
##[error]The operation was canceled.
linux-binary-manywheel / manywheel-py3_12-rocm6_4-test (gh)
linux-binary-manywheel / manywheel-py3_13-rocm6_3-build / build (gh)
##[error]The operation was canceled.
linux-binary-manywheel / manywheel-py3_13-rocm6_3-test (gh)
linux-binary-manywheel / manywheel-py3_13-rocm6_4-build / build (gh)
##[error]The operation was canceled.
linux-binary-manywheel / manywheel-py3_13-rocm6_4-test (gh)
linux-binary-manywheel / manywheel-py3_13t-rocm6_3-build / build (gh)
##[error]The operation was canceled.
linux-binary-manywheel / manywheel-py3_13t-rocm6_3-test (gh)
linux-binary-manywheel / manywheel-py3_13t-rocm6_4-build / build (gh)
##[error]The operation was canceled.
linux-binary-manywheel / manywheel-py3_13t-rocm6_4-test (gh)
windows-binary-libtorch-release / libtorch-cuda12_8-shared-with-deps-release-test (gh)
windows-binary-wheel / wheel-py3_10-cuda12_8-test (gh)
windows-binary-wheel / wheel-py3_10-cuda12_9-build (gh)
sccache: error: couldn't connect to server
windows-binary-wheel / wheel-py3_10-cuda12_9-test (gh)
windows-binary-wheel / wheel-py3_10-xpu-build (gh)
windows-binary-wheel / wheel-py3_10-xpu-test (gh)
windows-binary-wheel / wheel-py3_11-cuda12_6-test (gh)
windows-binary-wheel / wheel-py3_11-cuda12_8-test (gh)
windows-binary-wheel / wheel-py3_11-cuda12_9-build (gh)
sccache: error: couldn't connect to server
windows-binary-wheel / wheel-py3_11-cuda12_9-test (gh)
windows-binary-wheel / wheel-py3_11-xpu-build (gh)
windows-binary-wheel / wheel-py3_11-xpu-test (gh)
windows-binary-wheel / wheel-py3_12-cuda12_8-build (gh)
sccache: error: couldn't connect to server
windows-binary-wheel / wheel-py3_12-cuda12_8-test (gh)
windows-binary-wheel / wheel-py3_12-cuda12_9-test (gh)
windows-binary-wheel / wheel-py3_12-xpu-build (gh)
windows-binary-wheel / wheel-py3_12-xpu-test (gh)
windows-binary-wheel / wheel-py3_13-cuda12_8-test (gh)
windows-binary-wheel / wheel-py3_13-cuda12_9-test (gh)
windows-binary-wheel / wheel-py3_13-xpu-build (gh)
windows-binary-wheel / wheel-py3_13-xpu-test (gh)
windows-binary-wheel / wheel-py3_13t-cuda12_8-test (gh)
windows-binary-wheel / wheel-py3_13t-xpu-build (gh)
windows-binary-wheel / wheel-py3_13t-xpu-test (gh)
windows-binary-wheel / wheel-py3_9-cuda12_8-test (gh)
windows-binary-wheel / wheel-py3_9-xpu-build (gh)
windows-binary-wheel / wheel-py3_9-xpu-test (gh)

FLAKY - The following jobs failed but were likely due to flakiness present on trunk:

linux-binary-libtorch / libtorch-rocm6_4-shared-with-deps-release-upload / upload (gh) (similar failure)
##[error]The operation was canceled.
linux-binary-manywheel / manywheel-py3_9-rocm6_3-upload / upload (gh) (similar failure)
##[error]The operation was canceled.
windows-binary-libtorch-release / libtorch-cuda12_8-shared-with-deps-release-build (gh) (matched win rule in flaky-rules.json)
##[error]The operation was canceled.
windows-binary-libtorch-release / libtorch-cuda12_9-shared-with-deps-release-test (gh) (matched win rule in flaky-rules.json)
##[error]The operation was canceled.
windows-binary-wheel / wheel-py3_10-cuda12_8-build (gh) (matched win rule in flaky-rules.json)
##[error]The operation was canceled.
windows-binary-wheel / wheel-py3_11-cuda12_8-build (gh) (matched win rule in flaky-rules.json)
##[error]The operation was canceled.
windows-binary-wheel / wheel-py3_12-cuda12_9-build (gh) (matched win rule in flaky-rules.json)
##[error]The operation was canceled.
windows-binary-wheel / wheel-py3_13-cuda12_6-test (gh) (matched win rule in flaky-rules.json)
##[error]The operation was canceled.
windows-binary-wheel / wheel-py3_13-cuda12_8-build (gh) (matched win rule in flaky-rules.json)
##[error]The operation was canceled.
windows-binary-wheel / wheel-py3_13-cuda12_9-build (gh) (matched win rule in flaky-rules.json)
##[error]The operation was canceled.
windows-binary-wheel / wheel-py3_13t-cuda12_6-test (gh) (matched win rule in flaky-rules.json)
##[error]The operation was canceled.
windows-binary-wheel / wheel-py3_13t-cuda12_8-build (gh) (matched win rule in flaky-rules.json)
##[error]The operation was canceled.
windows-binary-wheel / wheel-py3_13t-cuda12_9-test (gh) (matched win rule in flaky-rules.json)
##[error]The operation was canceled.
windows-binary-wheel / wheel-py3_9-cuda12_6-test (gh) (matched win rule in flaky-rules.json)
##[error]The operation was canceled.
windows-binary-wheel / wheel-py3_9-cuda12_8-build (gh) (matched win rule in flaky-rules.json)
##[error]The operation was canceled.
windows-binary-wheel / wheel-py3_9-cuda12_9-upload / upload (gh) (matched win rule in flaky-rules.json)
##[error]The operation was canceled.

UNSTABLE - The following job is marked as unstable, possibly due to flakiness on trunk:

pull / cuda12.8-py3.10-gcc9-sm75 / test (pr_time_benchmarks, 1, 1, linux.g4dn.metal.nvidia.gpu, unstable) (gh) (#153987)
MISSING REGRESSION TEST

This comment was automatically generated by Dr. CI and updates every 15 minutes.

pytorchbot added the open source label

atalman added the ciflow/binaries label

atalman approved these changes

View reviewed changes

atalman merged commit d91af20 into release/2.8

345 of 387 checks passed

atalman mentioned this pull request

Release 2.8.0 validations checklist and cherry-picks #158939

Closed

79 tasks

Contributor

atalman commented Aug 1, 2025

Validated. See: https://download.pytorch.org/whl/test/cu128_full/torch/index.html

github-actions bot deleted the cherry-pick-157630-by-pytorch_bot_bot_ branch

September 1, 2025 02:19

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

ciflow/binaries open source topic: not user facing