The label module: cuda is only applicable to issues and has been removed. Please only use this label on issues.

The label module: floatx (formerly float8) is only applicable to issues and has been removed. Please only use this label on issues.

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/167734

• 📄 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

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

❌ 7 New Failures, 10 Unrelated Failures

As of commit c162c57 with merge base 2c846bb ():

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

They need their own SM arch? ;-;

They need their own SM arch? ;-;

Yes, the arch specific instructions are not compatible between 12.0 and 12.1.

The executorch test failures are unrelated.

@pytorchbot merge -i

Merge started

Your change will be merged while ignoring the following 1 checks: trunk / linux-jammy-py3-clang12-executorch / test (executorch, 1, 1, lf.linux.2xlarge, unstable)

Learn more about merging in the wiki.

Questions? Feedback? Please reach out to the PyTorch DevX Team

@pytorchbot revert -m "fails on CUDA 12.8"

❌ 🤖 pytorchbot command failed:

@pytorchbot revert: error: the following arguments are required: -c/--classification

usage: @pytorchbot revert -m MESSAGE -c
                          {nosignal,ignoredsignal,landrace,weird,ghfirst,autorevert}

Try @pytorchbot --help for more info.

@pytorchbot revert -m "fails on CUDA 12.8" -c nosignal

@pytorchbot successfully started a revert job. Check the current status here.
Questions? Feedback? Please reach out to the PyTorch DevX Team

@Aidyn-A your PR has been successfully reverted.

Thanks for reverting, this should restore the nightly wheel from the below error. we should add the ciflow/binaries label next time..
nvcc fatal : Unsupported gpu architecture 'compute_121a'
cc @atalman , could you help restart the https://github.com/pytorch/pytorch/actions/runs/19421948119/job/55560842460 when you get a chance, I was not able to restart the 12.8 build

HI @tinglvv and @Aidyn-A thank you for revert. We will wait for tomorrow build to confirm.

Looks like it caused all CUDA 12.8 builds to fail, Linux x86, aarch64 and Windows x86

Yeah, those failures are certainly not related to the sm_121a flag.

@pytorchbot merge -i

Merge started

Your change will be merged while ignoring the following 17 checks: trunk / linux-jammy-py3-clang12-executorch / test (executorch, 1, 1, linux.2xlarge, unstable), trunk / win-vs2022-cpu-py3 / test (default, 1, 4, windows.4xlarge.nonephemeral), linux-binary-manywheel / manywheel-py3_10-rocm7_0-test, linux-binary-manywheel / manywheel-py3_13t-rocm7_0-test, linux-binary-manywheel / manywheel-py3_14-rocm7_0-test, linux-binary-manywheel / manywheel-py3_14t-rocm7_0-test, linux-binary-manywheel / manywheel-py3_11-rocm7_1-test, linux-binary-manywheel / manywheel-py3_14-rocm7_1-test, linux-binary-manywheel / manywheel-py3_13t-rocm7_1-test, linux-binary-manywheel / manywheel-py3_14t-rocm7_1-test, linux-binary-libtorch / libtorch-rocm7_1-shared-with-deps-release-build / build, linux-binary-libtorch / libtorch-cpu-shared-with-deps-release-test / test, linux-binary-libtorch / libtorch-cuda12_6-shared-with-deps-release-test / test, linux-binary-libtorch / libtorch-cuda13_0-shared-with-deps-release-test / test, linux-binary-libtorch / libtorch-cuda12_8-shared-with-deps-release-test / test, linux-binary-libtorch / libtorch-cuda12_9-shared-with-deps-release-test / test, linux-binary-libtorch / libtorch-rocm7_0-shared-with-deps-release-test

Learn more about merging in the wiki.

Questions? Feedback? Please reach out to the PyTorch DevX Team