> > forking and monkey-patching it for ourselves is a solution for us, > the people who don't know better would just give up. > Would you be willing to share a proposed patch? I think everyone on the list is a reasonable adult, and if it's indeed as simple and costless as you suggest I'm sure pragmatism would prevail. On the other hand public scrutiny might expose some limitation in the general case that is not triggered in your specific case (which would bring us back to "there are unavoidable tradeoffs and easy fixes"). Given the history of LuaJIT development and discussion on this list, my firm belief is that actually-superior solutions are quickly adopted. But finding an actually-superior solution for a general purpose JIT compiler is a lot harder than one may think. Demetri