Remove CUDA build warnings
Nobody has claimed this yet.
- Dominant language
- C++
- Stars
- 196
- Forks
- 34
- Avg merge
- 4d 1h
- Merged PRs (30d)
- 11
Description
When build Axom using the host config files to configure for llvm@19.1.3_cuda (matrix and rzvector yield the same results), several compiler warnings are emitted. An earlier issue https://github.com/llnl/axom/issues/1848 describes more sources of warnings and work has been done to eliminate some of those warnings. For example,
This issue now supersedes that issue.
The remaining warnings appear to be attributed to two main causes:
- nvlink warnings due to inability to statically determine stack size
- calling a host function from host-device function where the host function is an axom::Array method. Axom's array class is not designed to be used in device code. These warnings arise in Axom core and Axom primal. Removing the warnings will require some code re-design and possibly a new storage type for some Axom internals.
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 by reproducing the build with the host config files for llvm@19.1.3_cuda, using the matrix or rzvector configuration, and inspect the warnings from nvlink and Axom core and primal. Review the changes in issues 1848 and pull requests 1931 and 1932; done means the reported CUDA builds no longer emit the targeted warnings.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- cpp
- Domain
- build-system
- Issue type
- Bug
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Quiet
- Clarity
- Mostly clear
- Newbie friendliness
- 35/100