JuliaORNL / JACC.jl

CPU/GPU parallel performance portable layer in Julia via functions as arguments
MIT License
21 stars 13 forks source link

Synchronization not working when using CUDA backend in JACC 0.0.5. #117

Open alexishuante opened 3 weeks ago

alexishuante commented 3 weeks ago

On JACC 0.0.5, I had an issue where the synchronization when using the CUDA backend was not performed correctly. I had to manually add CUDA.synchronize() on the code, which defeats the purpose of performance portability.

However, updating to the current branch ((Package)> add JACC#main) solved this issue. It is important to mention that this occurred on the miniBUDE proxy app.