Currently tmLQCD for GPUs fails to compile, because the c99 complex type is not yet introduced there.
It's not quite clear to me how this is going to work. The GPU code is template based and hence C++, but C99 complex doesn't port cleanly. I do believe we can do a memory cast from _Complex double, through double, to std::complex< double >. I'd have to look where to do that, though.
yes, good that you mention this. There might be C++ issues. Just right now it doesn't compile for obvious reasons...
@florian-burger is working on this
This has been fixed, closig.