Yes indeed, we'll do it next (unless somebody beats me to it). First I am focusing on compiling fastGPT with LFortran, we can do it, but have a few workarounds

Yes indeed, we'll do it next (unless somebody beats me to it). First I am focusing on compiling fastGPT with LFortran, we can do it, but have a few workarounds that I want to fix. Then we'll do llama2.