CUDA bindings
ponce via Digitalmars-d-announce
digitalmars-d-announce at puremagic.com
Sun Oct 26 03:46:21 PDT 2014
On Sunday, 26 October 2014 at 08:18:11 UTC, Tofu Ninja wrote:
> On Sunday, 26 October 2014 at 05:31:52 UTC, Dmitri Nesteruk
> wrote:
>> This is great! I know that C++ uses <<< and >>> to enclose
>> kernel calls and thus create the link between CPU and GPU code
>> when NVCC rips things apart. How is this done in your bindings?
>
> It's just the driver api, its not CUDA code in D.
>
> Also I think you are mistaking where the <<< >>> are actually
> used. The <<< >>> are used in CUDA code, not in C++ code. While
> CUDA is a variation on C++, it is still not C++ and has to pass
> through a special parser that splits out the host code and the
> gpu code to be compiled.
Yeah even in C++ it isn't that desirable, apart from prototyping.
More information about the Digitalmars-d-announce
mailing list