Extend Support for CUDA and HIP kernel invocations

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).

changed milestone to %Version 0.1

added 1 commit

• e93ca92c - fix include flags in genscript test suite

Compare with previous version

added 1 commit

• 6c120a84 - Add CUDA and HIP API provider protocols. Factor out GPU stuff into separate Gpu Composer.

Compare with previous version

added 1 commit

• 635d3693 - first working HIP test case

Compare with previous version

added 2 commits

• b1b71c32 - extend HIP test case. Do not alter block sizes in composer.
• 4bc5ff44 - add cuda tests; add CUDA requirement to CI

Compare with previous version

added 1 commit

• 39e83b12 - fix GPU CI task to use cuda image

Compare with previous version

added 1 commit

• d81a23d9 - Update .gitlab-ci.yml file

Compare with previous version

added 1 commit

• 455b455e - fix header suffix

Compare with previous version

added 1 commit

• 8fea4416 - test automatic block size in linear3d

Compare with previous version

added 1 commit

• e9fdbbe1 - fix cuda runtime header path

Compare with previous version

added 1 commit

• 3ff729e9 - fix CUDA kernels test

Compare with previous version

added 2 commits

• 21c0ba7f - documentation on GPU invocation
• f2ad7231 - more explanation in composer guide

Compare with previous version

added 1 commit

• 377bab57 - remove extensions.gpu

Compare with previous version

marked this merge request as ready

changed title from Draft: Extended Support for CUDA kernel invocations to Extended Support for CUDA and HIP kernel invocations

changed the description

requested review from @ob28imeq

assigned to @da15siwa

changed title from Extended Support for CUDA and HIP kernel invocations to Extend Support for CUDA and HIP kernel invocations

added 1 commit

• e4c257df - add lang.gpu to api docs

Compare with previous version