{lang}[llvm-compilers/20.1.8] AdaptiveCpp v25.10.0 w/ CUDA 12.9.1#24894
Conversation
….9.1.eb, AdaptiveCpp-25.10.0-llvm-compilers-20.1.8.eb
|
Test report by @Thyre |
|
@boegelbot please test @ jsc-zen3 |
|
@Thyre: Request for testing this PR well received on jsczen3l1.int.jsc-zen3.fz-juelich.de PR test command '
Test results coming soon (I hope)... Details- notification for comment with ID 3661222025 processed Message to humans: this is just bookkeeping information for me, |
|
Test report by @Thyre |
|
Test report by @boegelbot |
|
Forgot Test report by @Crivella |
|
Test report by @Crivella |
|
Test report by @SebastianAchilles |
|
Going in, thanks @Thyre! |
(created using
eb --new-pr)Formerly hipSYCL and OpenSYCL, this project implements two accelerator paradigms to "easily" run code in parallel on the host or on GPUs. This particular PR adds the host-only variant via
libomp.soand the CUDA accelerated variant (also capable of using OpenMP). Theoretically, support for ROCm & Level Zero / OpenCL could be added, but we're missing a few things in EasyBuild for that.Chose LLVM as the toolchain, as this is really the only sensible toolchain to put this in, based on their documentation. Some features, like accelerated OpenMP support, require LLVM, as does basically any other backend.
Running tests is unfortunately a bit more complicated, since that requires a whole separate compilation step.