Add Bitwise Consistent Attribute
Nobody has claimed this yet.
- Dominant language
- Jupyter Notebook
- Stars
- 135
- Forks
- 55
- Avg merge
- 4d 16h
- Merged PRs (30d)
- 5
Description
Add an attribute to kernels for if the answer/checksum for each tuning is bitwise the same run-to-run or not. Note that one tuning may differ bitwise with other tunings of the same kernel but be consistent with itself. This should be true for most kernels like DAXPY and HALOEXCHANGE but not true for kernels with atomics like REDUCE_SUM and NODAL_ACCUMULATION_3D.
Contributor guide
No contributing guide indexed for this repository
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 locating how kernel tuning answers and checksums are recorded, then compare the behavior of DAXPY and HALOEXCHANGE with REDUCE_SUM and NODAL_ACCUMULATION_3D. The attribute is done when it reports whether each tuning is bitwise consistent with itself across runs, without requiring different tunings to match.
Written by the indexing model from the issue text.
Assessment
- Domain
- performance
- Issue type
- Feature
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Stale
- Clarity
- Mostly clear
- Newbie friendliness
- 45/100