[BUG] [CPU] Some ops fallback to fp32 implementation even fro fp64
Open
Nobody has claimed this yet.
bug
low priority
- Dominant language
- C++
- Stars
- 28.5k
- Forks
- 2.3k
- Avg merge
- 3d 8h
- Merged PRs (30d)
- 62
Description
Some ops (exp, sin, cos, erf, erfinv) use the fp32 approximation for double and are hence not so accurate for double precision. Would be good to have specializations of these for double as well.
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 locating the CPU implementations for exp, sin, cos, erf, and erfinv, then trace where double inputs select the fp32 approximations. Compare the available fp64 math paths and existing CPU tests, if present. Done means each listed operation uses a double-precision specialization and its accuracy is covered by validation.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- cpp
- Domain
- backend, performance
- Issue type
- Bug
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Quiet
- Clarity
- Mostly clear
- Newbie friendliness
- 48/100