Cristian-Vasile Achim

Thank you for your prompt reply. I have been playing with the spack installations for a few days without knowing if what I did was ok or not. The available...

Hello, I managed to advance a little. I took the existing spack installing file edited to compile llvm with nvtpx and hipsycl + cuda. Now I can select A100 as...

This is the code I compiled: ``` syclcc -O2 --hipsycl-targets="omp;cuda:sm_80" hello.cpp #include #include using namespace sycl; int main() { const std::string secret { "Ifmmp-!xpsme\"\012J(n!tpssz-!Ebwf/!" "J(n!bgsbje!J!dbo(u!ep!uibu/!.!IBM\01" }; const auto sz =...

After more trials it seems that it works with the development branch. I changed the spack script to install install the llvm14 with nptx and cuda 11.5. I am using...

Thanks for the help.

> I am not, although others have asked. Have a look at Discussions over at FINUFFT GitHub. There are others that may want to help. Best, Alex > […](#) >...

Thank you for reply. I can infer that there are no special cuda features used. I can not dig now the amount of shared memory available per CU (SMP). This...

Thank you for your replies. I will let you know how it goes. Cristian

> Hello, @csccva, I see that you are trying to build Rocrand backend with Intel oneAPI compiler (icpx), but it works only with open source compiler [oneAPI DPC++ Compiler](https://github.com/intel/llvm) or...

> Though it isn't strictly supported by oneMKL Interface Library, the oneAPI base toolkit with Codeplay's plugins for AMD seems to be able to build the rocRAND backend fine for...