[CI] CUTILE_SMOKE_TEST GB300 failed
Open
Nobody has claimed this yet.
bug
- Dominant language
- Cuda
- Stars
- 854
- Forks
- 236
- Avg merge
- 3d 3h
- Merged PRs (30d)
- 62
Description
release/26.10 and main Nighlies that use GB300 runners are failing with:
https://github.com/NVIDIA/cuvs/actions/runs/34517645748/job/103007042888#step:12:3119
5/38 Test #5: CUTILE_SMOKE_TEST ..........................***Failed 0.13 sec
'./../../..//bin/gtests/libcuvs/CUTILE_SMOKE_TEST'
Running main() from /tmp/conda-bld-output/bld/rattler-build_libcuvs-headers/work/cpp/build/_deps/gtest-src/googletest/src/gtest_main.cc
[==========] Running 2 tests from 1 test suite.
[----------] Global test environment set-up.
[----------] 2 tests from CutileSmoke
[ RUN ] CutileSmoke.ResolvesEveryEmbeddedArchitecture
[ OK ] CutileSmoke.ResolvesEveryEmbeddedArchitecture (0 ms)
[ RUN ] CutileSmoke.LaunchesCompatibleCubin
/tmp/conda-bld-output/bld/rattler-build_libcuvs-headers/work/cpp/tests/detail/jit_lto/cutile_smoke.cu:86: Failure
Expected: (launcher) != (nullptr), actual: (nullptr) vs (nullptr)
cc @divyegala
Contributor guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
Start with tests/detail/jit_lto/cutile_smoke.cu, especially the failing CutileSmoke.LaunchesCompatibleCubin test, and inspect the linked GitHub Actions job on a GB300 runner. Reproduce CUTILE_SMOKE_TEST in the affected release/26.10 or main nightly workflow; done means both smoke tests pass on GB300 runners.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- cpp, github-actions
- Domain
- ci-cd, testing-qa
- Issue type
- Bug
- Difficulty
- 3/5
- Estimated time
- 1-2 days
- Activity status
- Active
- Clarity
- Mostly clear
- Newbie friendliness
- 55/100