Hacktoberfest 2026: the issues maintainers tagged for October, open and beginner-friendly. Browse Hacktoberfest issues

No test catches missing synchronization between work-groups

Open
  • Is something wrong?
#547 0 comments 0 reactions 0 assignees View on GitHub

Maintainers usually reply within 1 day

Nobody has claimed this yet.

Assessment

Difficulty
4/5
Estimated time
3-5 days
Newbie friendliness
48/100
Issue type
Feature
Clarity
Mostly clear
Activity status
Active
Tech stack
julia
Domain
backend, testing

Research direction

Start by reading the existing KernelAbstractions “fences” test and OpenCL.jl’s atomics tests; the issue does not name specific test files or entry points. Add a repeated message-passing test across many work-groups that covers the listed atomic paths, explicit fences, and both backends. Done means the test exercises these cases on hardware and catches weakened device-scope synchronization.

Written by the indexing model from the issue text.

Description

ci kernels

JuliaGPU/spirv2clc#7 was a translator bug that weakened device-scope atomics and fences to work-group scope. On NVIDIA, a device-scope release/acquire fence compiled to membar.cta. No test in OpenCL.jl, nor the KernelAbstractions suite it runs, caught it:

  • KernelAbstractions' "fences" test uses a single flag, which almost never shows the race.
  • The atomics tests here only check results of read-modify-write operations, and the disassembly's Scope operand.

A message-passing test across many work-groups would catch this class of bug on hardware:

  • one work-item per work-group writes a payload, then publishes a flag with a release store or fence;
  • work-items in other work-groups spin on the flag with acquire loads, then read the payload;
  • run it at sizes large enough to keep many work-groups resident, and repeat it.

It should cover:

  • the Atomix/UnsafeAtomics paths (:release/:acquire, device and default system scope);
  • explicit fences;
  • both the IL backend and the spirv2clc backend.
Dominant language
Julia
Stars
313
Forks
50
Avg merge
1d 10h
Merged PRs (30d)
41

Getting set up

This project ships no dev container, Dockerfile or contributing guide, so setting up is up to you: start from its README, and see our first-contribution guide for the general steps.

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

More from JuliaGPU/OpenCL.jl

All issues in JuliaGPU/OpenCL.jl

Similar issues

More Julia issues

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.