[llvm-dev] [RFC] Make Lanai backend non-experimental
Pete Cooper via llvm-dev
llvm-dev at lists.llvm.org
Tue Jul 19 12:57:04 PDT 2016
> On Jul 19, 2016, at 9:42 AM, Renato Golin via llvm-dev <llvm-dev at lists.llvm.org> wrote:
> A few basic rules to get accepted are if:
> * the target exists and can be easily purchased / emulated for
> investigating problems,
> * there are official documents / specs published by the project /
> company that maintains the targets,
> * there is a reasonable community maintaining the rest of the system
> (firmware, OS, other tools, etc),
> * enough people commit themselves to maintain the LLVM back-end to
> avoid bit-rot,
> * the back-end is free of contentious features that would mean
> breaking every other target.
This is an excellent list. We should probably have something like it in the docs if we don’t already.
I have no problem with Lanai in LLVM (non-experimentally) if it meets most of these criteria to some degree. Obviously that doesn’t mean every target in LLVM needs to have detailed microarchitecture documents, but a basic ISA should be publicly available. Personally I also think the ability to at least simulate the code is very important too.
I tried googling it but mostly found news reports about the existence of the backend, not documentation or anything else.
I think the Lanai community have done an excellent job meeting the last 2 points. However, I’d really appreciate more information on the first 3. I think meeting each of them (even minimally) is important to leave experimental status.
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the llvm-dev