fix(ci): use existing sherpa-onnx CUDA GPU archive name#652
Conversation
Signed-off-by: streamkit-devin <devin@streamkit.dev>
🤖 Devin AI EngineerI'll be helping with this pull request! Here's what you should know: ✅ I will automatically:
Note: I can only respond to comments from users who have write access to this repository. ⚙️ Control Options:
|
![staging-devin-ai-integration[bot]](https://avatars.githubusercontent.com/in/870844?s=60&v=4){kind=small}
| # GPU (CUDA-enabled) sherpa-onnx shared build; vendored into cuda variants so | ||
| # the execution-provider-agnostic plugin .so dispatches to the GPU at runtime. | ||
| SHERPA_ONNX_GPU_ARCHIVE: "sherpa-onnx-v1.12.17-linux-x64-gpu-shared.tar.bz2" | ||
| SHERPA_ONNX_GPU_ARCHIVE: "sherpa-onnx-v1.12.17-cuda-12.x-cudnn-9.x-linux-x64-gpu.tar.bz2" |
There was a problem hiding this comment.
🔍 New GPU archive name relies on basename==extracted-dir convention
The change updates SHERPA_ONNX_GPU_ARCHIVE to sherpa-onnx-v1.12.17-cuda-12.x-cudnn-9.x-linux-x64-gpu.tar.bz2. The download/extract step at .github/workflows/marketplace-build.yml:261-265 derives the extracted directory generically via dir="${SHERPA_ONNX_GPU_ARCHIVE%.tar.bz2}", so it adapts to the rename automatically. This is only correct if the CUDA tarball extracts to a top-level directory whose name exactly matches the archive basename (sherpa-onnx-v1.12.17-cuda-12.x-cudnn-9.x-linux-x64-gpu). If the k2-fsa release uses a different internal directory layout, the cp -r "${dir}/lib/." step would fail. Worth a quick confirmation against the actual release asset, since it can't be verified from the code alone.
Was this helpful? React with 👍 or 👎 to provide feedback.
Debug
There was a problem hiding this comment.
Verified against the actual release asset: downloaded sherpa-onnx-v1.12.17-cuda-12.x-cudnn-9.x-linux-x64-gpu.tar.bz2 and its top-level directory is exactly the archive basename, with lib/ containing libsherpa-onnx-c-api.so, libonnxruntime.so, libonnxruntime_providers_cuda.so, and libonnxruntime_providers_shared.so — everything the bundle vendoring needs.
Codecov Report✅ All modified and coverable lines are covered by tests. Additional details and impacted files@@ Coverage Diff @@
## main #652 +/- ##
==========================================
- Coverage 85.25% 85.25% -0.01%
==========================================
Files 249 249
Lines 75765 75765
Branches 2314 2314
==========================================
- Hits 64595 64594 -1
- Misses 11164 11165 +1
Partials 6 6
Flags with carried forward coverage won't be shown. Click here to find out more.
🚀 New features to boost your workflow:
|
2 participants
Summary
sherpa-onnx-v1.12.17-linux-x64-gpu-shared.tar.bz2— that asset name doesn't exist upstream. Switch tosherpa-onnx-v1.12.17-cuda-12.x-cudnn-9.x-linux-x64-gpu.tar.bz2, the same CUDA 12 / cuDNN 9 buildDockerfile.full-gpualready vendors.${ARCHIVE%.tar.bz2}(the install step relies on this), and it ships all libs the bundles need (libsherpa-onnx-c-api.so,libonnxruntime.so,libonnxruntime_providers_cuda.so,libonnxruntime_providers_shared.so).Review & Validation
gpurunner label, now fixed).Link to Devin session: https://staging.itsdev.in/sessions/a1b47be9c21e48d29ec7397d56720bf9
Requested by: @streamer45
Devin Review
6875521