General purpose GPU compute framework built on Vulkan to support 1000s of cross vendor graphics cards (AMD, Qualcomm, NVIDIA & friends). Blazing fast, mobile-enabled, asynchronous and optimized for advanced GPU data processing usecases. Backed by the Linux Foundation.
I noticed
Kompute.hpp
contains just a copy of all other headers. Is there any reason for not doing something like this:Since then the preprocessor will take over the task of expanding all
#include
macros for us. An example of that can be found here: https://github.com/libcpr/cpr/blob/master/include/cpr/cpr.hThen we also get rid of all the logic for compiling either Kompute as a single header via
KOMPUTE_OPT_BUILD_SINGLE_HEADER
.