dotnet / dotnet/machinelearning
CpuMath Enhancement: Optimize codegen of inlining in hardware intrinsics APIs
Open
area-Native
Priority:2
- Dominant language
- C#
- Stars
- 9.4k
- Forks
- 2k
- Avg merge
- 2d 20h
- Merged PRs (30d)
- 11
Description
Style changes needed to solve part of https://github.com/dotnet/machinelearning/issues/823
## Details
- (from Intel partners) In the original code of `src\Microsoft.ML.CpuMath\SseIntrinsics.cs`, even though `VectorSum` is inlined, the codegen is not optimized due to register spill and reload. It seems JIT has optimization opportunity over there. Do you mind opening an issue to discuss about it on CoreCLR github repo?
cc: @tannergooding since you might already have addressed this issue.
Contributor guide
Assessment
This issue has not been assessed yet.