Skip to content

AcceleratedKernels accumulate and sort - #423

Open
michel2323 wants to merge 3 commits into
JuliaGPU:mainfrom
michel2323:accel-kernels
Open

AcceleratedKernels accumulate and sort#423
michel2323 wants to merge 3 commits into
JuliaGPU:mainfrom
michel2323:accel-kernels

Conversation

@michel2323

Copy link
Copy Markdown
Member

This adds accumulate and sort through AcceleratedKernels.jl to be on par with CUDA.jl, AMDGPU.jl, and oneAPI.jl.

@github-actions

github-actions Bot commented Mar 4, 2026

Copy link
Copy Markdown
Contributor

Your PR no longer requires formatting changes. Thank you for your contribution!

@codecov

codecov Bot commented Mar 4, 2026

Copy link
Copy Markdown

Codecov Report

❌ Patch coverage is 0% with 7 lines in your changes missing coverage. Please review.
✅ Project coverage is 80.57%. Comparing base (024304e) to head (fab4f2c).

Files with missing lines Patch % Lines
src/accumulate.jl 0.00% 4 Missing ⚠️
src/sorting.jl 0.00% 3 Missing ⚠️
Additional details and impacted files
@@            Coverage Diff             @@
##             main     #423      +/-   ##
==========================================
- Coverage   81.17%   80.57%   -0.61%     
==========================================
  Files          13       15       +2     
  Lines         935      942       +7     
==========================================
  Hits          759      759              
- Misses        176      183       +7     

☔ View full report in Codecov by Harness.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:
  • ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

Comment thread src/accumulate.jl
Comment on lines +1 to +3
# Use a smaller block size to work around a scan correctness issue
# with the Blelloch parallel prefix sum at larger block sizes (>=128).
const _ACCUMULATE_BLOCK_SIZE = 64

Copy link
Copy Markdown
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Can you expand on this? Is this a general AK issue or specific to OpenCL? Does this not depend on the workgroup/subgroup size?

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants