I finished writing up an initial draft for weave on the kernel. I finally compiled it and it obviously contained a host of errors.
To prevent rebuilding after every small change, I started editing the rt.cl in share/opencl instead of the one in src.
I also wrote code for weaving 0 thickness segments. I’m very unsure about my implementation though, so I’ll need to spend some time running and fixing it. For now, it gives a segmentation fault if I run a loop to try to read all the generated partitions. I’ll try to figure out what’s wrong tomorrow.