FWIW, this is hypothetical, as in I don't actually have a particular GCN compute shader binary to run yet. I am working on something related to this, but I'm realizing that I have a lot ideas in my head that are hazy and uncertain, so I'm going back to basics and trying to fix any misconceptions I might have.
So assuming I did have such a binary that I wanted my GPU to run (let's say I want to do this in C#), what steps would I need to go through/are there particularly instructive/relevant examples out there in github?
Second question; let's say that I have a directX binary that I'm successfully running as a compute shader on my AMD GPU... is there some way I can get the GCN binary that the directX binary is being compiled into?