Skip to content
Snippets Groups Projects

Extend Support for CUDA and HIP kernel invocations

Merged Extend Support for CUDA and HIP kernel invocations
fhennig/cuda-invoke into master
Merged Frederik Hennig requested to merge fhennig/cuda-invoke into master

This MR extends the support for CUDA and HIP kernel invocations and API reflection in pystencils-sfg, reflecting the concepts introduced in pystencils!449 (merged).

  • Move GPU-related stuff into new composer mix-in SfgGpuComposer
  • Replace cuda_invoke() by gpu_invoke()
  • Implement three separate signatures of gpu_invoke(), depending on the kernel indexing scheme selected by the user
  • Replace extensions.gpu by lang.gpu
  • Add explanations on GPU kernels to user manual
  • Fix some errors and clarify some interface descriptions in the user manual

This MR will already be using Target.HIP and therefore will have to wait for pystencils!458 (merged).

Edited by Frederik Hennig

Merge request reports

Pipeline #76963 passed

Stage:
    Stage:
      Stage:
        Stage:

          Pipeline passed for dc1a3935 on fhennig/cuda-invoke

          Test coverage 88.00% (1.00%) from 2 jobs
          Approval is optional

          Merged by Christoph AltChristoph Alt 2 weeks ago (Mar 21, 2025 11:25am UTC)

          Loading

          Pipeline #76971 passed

          Stage:
            Stage:
              Stage:
                Stage:

                  Pipeline passed for 020347e7 on master

                  Test coverage 88.00% (1.00%) from 2 jobs

                  Activity

                  Filter activity
                  • Approvals
                  • Assignees & reviewers
                  • Comments (from bots)
                  • Comments (from users)
                  • Commits & branches
                  • Edits
                  • Labels
                  • Lock status
                  • Mentions
                  • Merge request status
                  • Tracking
                  Please register or sign in to reply