Performance regression using `^`
Nobody has claimed this yet.
- Dominant language
- Julia
- Stars
- 344
- Forks
- 79
- Avg merge
- 2d 23h
- Merged PRs (30d)
- 27
Description
As reported in https://github.com/luraess/JuliaGPUPerf/issues/2 and https://github.com/luraess/JuliaGPUPerf/issues/3, there is an issue significantly affecting performance when doing ^ operation within GPU kernels.
The Int32 on Int32 case (https://github.com/luraess/JuliaGPUPerf/issues/2) may have been fixed as upon suggestion from @vchuravy by using
my_pow(x, p) = ccall("llvm.powi.f32.i32", llvmcall, Float32, (Float32, Int32), x, p)
#[...]
A[ix,iy] = B[ix,iy] + s*my_pow(C[ix,iy], pow_int)
But the Float32 and Float64 cases are still lacking behind.
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 with the linked JuliaGPUPerf reports for the reproducible Float32 and Float64 kernel cases, then trace how ^ is lowered in AMDGPU GPU kernels. Compare those cases with the documented Int32 workaround and benchmark the affected operations; done means the Float32 and Float64 regressions are characterized and performance is improved without breaking kernel execution.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- julia
- Domain
- performance
- Issue type
- Bug
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 30/100