oneapi-src / oneapi-src/SYCLomatic
Use `sycl::bfloat16` class and functions instead of float casts.
Open
@tangjj11 is already working on this.
Since Sep 23, 2024.
enhancement
- Dominant language
- LLVM
- Stars
- 291
- Forks
- 99
- Avg merge
- 5d 5h
- Merged PRs (30d)
- 1
Description
The bfloat16 class has been non-experimental for a while now, supporting all backends: https://github.com/oneapi-src/SYCLomatic/pull/1286
However SYCLomatic appears to be not be using this, and instead just always casting to float, see e.g.https://github.com/oneapi-src/SYCLomatic/pull/1286.
This seems to be a lost opportunity. For example there are native cuda bfloat16 implementations of bfloat16 math functions in DPC++ that make bfloat16 math much faster than using casts to float.
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.
Assessment
This issue has not been assessed yet.