DO-NOT-MERGE: libgccjit experiment (work-in-progress) #109
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Inspired by Vladimir's talk at LPC, this is a proof-of-concept of populating a libgccjit context from a MIR module, and then compiling it with libgccjit.
This could allow for libgccjit to be used as a tier 2 JIT compiler in conjunction with MIR for tier 1.
Additionally, Vladimir's talk identified a weakness of libgccjit: the difficult of creating the environment through the API. With this approach, C could be compiled to MIR, and then injected into libgccjit.
Right now it's a very rough proof-of-concept that only implements a few instructions and modes. I'm posting it now in order to get early feedback. In particular it doesn't yet support forward references. But it does support compiling trivial binary mir files via libgccjit.
Given that you don't want external deps, I'm not sure where this should ultimately live. Also, I used C++ to simplify the implementation, but could recode it in C if that's a deal-breaker.