Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

ocl: GCN atomics for Mi-GPUs and other improvements #715

Merged
merged 1 commit into from
Sep 29, 2023
Merged

Conversation

hfp
Copy link
Member

@hfp hfp commented Sep 29, 2023

  • GCN: Rely on builtin atomics (Mi-GPUs); TODO: tuned params and check if supported.

  • Allow relying on platform name instead of device name (c_dbcsr_acc_opencl_device_vendor).

  • Introduced optional kernel-flags (opencl_libsmm_smm_t); needs some followup work later.

  • Improved kernel by relying on work_group_broadcast in general.

  • improved documentation for CUDA/OpenCL stand-alone drivers

  • Modernized Shell script (acc_opencl.sh).

  • Improved some runtime error messages.

  • Some more debug/developer settings.

* GCN: Rely on builtin atomics (Mi-GPUs); TODO: tuned params and feature-check can be improved.
* Allow relying on platform name instead of device name (c_dbcsr_acc_opencl_device_vendor).
* Introduced optional kernel-flags (opencl_libsmm_smm_t); needs some followup work later.
* Improved kernel by relying on work_group_broadcast in general.

* Improved documentation for CUDA/OpenCL stand-alone drivers
* Fixed warnings about unused variable (kernel code).
* Modernized Shell script (acc_opencl.sh).
* Improved some runtime error messages.
* Some more debug/developer settings.
@hfp hfp merged commit f52846a into cp2k:develop Sep 29, 2023
19 of 20 checks passed
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.

1 participant