CompilerDevTools: add proof of concept for caching runtime calls (#57193)
This PR leverages new capabilities that have been implemented in
https://github.com/JuliaLang/julia/pull/56660 to also support caching
methods for runtime calls under a custom `AbstractInterpreter`.
In my understanding, the idea is that when executing code compiled with
a custom `AbstractInterpreter`, only methods that can be compiled ahead
of execution will be cached, because dynamic calls requiring runtime
compilation will not have enough context to know which interpreter and
cache to use. They will instead use the native interpreter and cache.
This sample package demonstrates a way to ensure that runtime calls go
through an entry point (`with_new_compiler`) that provides the context
required to use the right cache and interpreter for subsequent
compilation.
I'd be happy to get feedback on the logic used to redirect runtime calls
to `with_new_compiler`. Overloading `optimize(::SplitCacheInterp)`
seemed the most convenient (right after that, the `IRCode` gets
converted to a `CodeInfo`), but perhaps there might be a better place
for it.
---------
Co-authored-by: Cédric Belmant <cedric.belmant@juliahub.com>
julia
edf1cd0c - CompilerDevTools: add proof of concept for caching runtime calls (#57193)
