Joachim Meyer

The library-only OpenMP CPU backend should still be supported with those options. `-DWITH_ACCELERATED_CPU=OFF` just deactivates dedicated compiler support for `nd_range` parallel for, which should improve the performance of that paradigm...

Btw... this actually even improves the performance (at least for all kernels of the performance tests), since until now, all other blocks had to wait for the first block to...

Hey :) what are the plans regarding moving to `gsl_lite` primarily? Just tried to use the `gsl_lite` namespace in a library with VC14, which failed somewhat badly, as VC14 doesn't...

I see, that is a legitimate concern. As a tester I mostly adapted the simple example from the readme and added some span and string_span usage, but only the owner...

The main issue with supporting Open SYCL on Windows is not really the backend support (sure, AMD GPUs were not supported due to lack of HIP on Windows), but rather...

Hi, I didn't have much time to look into it thus far, but I was able to reproduce the issue. The problem seems to be that the part after the...

I'll just add two cents: what you're doing with the if def at runtime is basically kernel specialization. You can achieve the same thing more elegantly in SYCL: use multiple...

In library-only flows, there's not really a translation, but instead as the name suggests, OpenSYCL acts as a library that internally uses the constructs by the programming models. For OpenMP...

Yes. The `_OPENMP` is usually set by the compilers whenever the corresponding `-fopenmp` (or similar flag to activate compilation with OpenMP) is provided. When using the OpenMP backend as you...

This image is a few days old, but should provide a Cuda environment where you can test your kernel.. https://hub.docker.com/r/fodinabor/opensycl