Fixes a race condition in the kernel batch_dotpv.
Description
This is related to a change in the CUDA programming model introduced with Volta cards.
The link is added showing how to solve the problem.
cuda-memcheck
is now fine for the tested input file.`
The debug information are now also added at the compile time of the kernels to ease tracking errors.
News snippet
Bugfix for GPUs.
Checklist
-
I have checked that my code follows the Octopus coding standards -
I have added tests for all the new features added in this request.