Place to discuss the expert map
we are building a map between different areas of the compiler and adding people that know about them
@nikomatsakis feel free to add yourself to the incremental compilation line
You're too kind ;)
could use "CTFE/Miri" with oli and eddy, presumably
I added various lines
i.e., I feel like... I could probably fix or explain the codegen unit partitioner
I'm not sure I'd say I'm an expert in it :P
whoever added "Libs", what is it meant to be? :)
I am debating what it means to be an "expert" now :)
I was thinking .. maybe we should include "directories" in this listing?
I do think @Santiago Pastorino that "MIR" should probably be divided into
librustc/mir-- MIR definition, not sure what to do with this
librustc_mir/build-- MIR construction
librustc_mir/borrow_check-- MIR borrow check
librustc_mir/transform-- various optimizations
in general, there are some areas of the compiler (e.g.,
librustc/ty) that are like "type definitions" -- I would call e.g. @eddyb or @varkor an expert on those, but also many other folks, maybe we should list them, maybe not
I'm also not sure if listing paths is right but I definitely think it will be useful in some sense:)
how specific do we intend to get with the map?
for example, I have a good understanding of the exhaustiveness/pattern matching code, but I'm not sure if that's broad enough an area to include here
I added myself to a bunch of things where I think I can be of help. I'd also like to split up a few things into dirs/files
we should probably have a line for the query system...
OK, I added one. I put @nikomatsakis, @Zoxc, @eddyb, and me to the list
feel free to add yourself to/remove your self from it, as you feel appropriate
I put myself and @eddyb for "debuginfo". @Manish Goregaokar would you say you know about debuginfo in the compiler too? Anyone else?
I know a bit, not a ton. I know more about the gdb implementation. Feel free to put me down as backup though.
My areas of expertise are: lints, features, editions, diagnostics, parsing, lowering, resolution. I'm willing to mentor stuff in these areas if pinged.
(I'm also decently okay with the overall codebase and can figure stuff out reasonably well, but i'm sure that's true for everyone else on this expert map too)
I threw on some rows; feel free to delete entries.
I am tempted to add "WG-nll" next to "MIR borrow check" because I feel like a lot of the knowledge there has been pretty well dispersed amongst the group
does anyone feel like an expert on
specifically regarding the confidence to review https://github.com/rust-lang/rust/pull/59008
Also, does anybody wants to review https://github.com/rust-lang/rust/pull/58349 (resolve: Simplify import resolution for mixed 2015/2018 edition mode)?
I usually reassign anything assigned to @pnkfelix, but in this case I'm not sure who'd be the proper candidate.
Ugh I’m sorry to have overlooked that. I can look at it today and then delegate if necessary
(Should we make a #t-compiler topic dedicated to general review requests?)
created topic for review requests over here
does anyone feel like an expert on
@varkor I am
@nikomatsakis: I cc'd you in the pull request, but I know you're very busy at the moment, so I was wondering whether anyone else was familiar enough, to avoid piling more reviews on your plate
@nikomatsakis @davidtwco added a PR for this as we agreed https://github.com/rust-lang/compiler-team/pull/40
Regarding your comment about embedding it directly into the repository, I think the intent was that we would do that eventually, once we had felt it settled down and wouldn't get changed as frequently. relevant message from meeting
yes, I was thinking if it's not sufficiently stable
but anyway, I'm fine with both things :)
also done this https://github.com/rust-lang/rustc-guide/pull/286
sorry for my very limited english vocabulary
it's like written by a 4 year old kid :P
@Santiago Pastorino it's perfectly fine English :slight_smile:
writing complicated sentences is not better formal writing, simpler is often better ^^
@centril thanks :)
@Santiago Pastorino I left a review with a few suggestions
will check now
@nikomatsakis both things are done!
@nikomatsakis @davidtwco I've opened a thread about this on
saw that that was one of the things that happened during last
unsure what else we can do, other than what I've done which is asking the rest of compiler people if they can fill the thing
maybe we can mention this during next steering meeting?
after that I guess we can just move the thing to github and start using PRs as a way to modify it accordingly
@Santiago Pastorino , yup, thanks. already got all info