Reading from the return place is fine
Const eval thinks that reading from local `_0` is UB, but it isn't. `_0` is just a normal local like any other, and codegen handles it that way too. The only special thing is that the `Return` terminator will read from it.
I've hit these errors while working on an NRVO pass that can merge other locals with `_0` in https://github.com/rust-lang/rust/pull/71003.
r? @oli-obk
proc_macro::is_available()
This PR adds `proc_macro::is_available() -> bool` to determine whether proc_macro has been made accessible to the currently running program.
The proc_macro crate is only intended for use inside the implementation of procedural macros. All the functions in the crate panic if invoked from outside of a procedural macro, such as from a build script or unit test or ordinary Rust binary.
Unfortunately those panics made it impossible for libraries that are designed to support both macro and non-macro use cases (e.g. Syn) to be used from binaries that are compiled with panic=abort. In panic=unwind mode we're able to attempt a proc macro call inside catch_unwind and use libproc_macro's result if it succeeds, otherwise fall back to a non-macro alternative implementation. But in panic=abort there was no way to determine which implementation needs to be used.
r? @eddyb
attn: @petrochenkov @adetaylor
ref: https://github.com/dtolnay/cxx/issues/130
allow wasm32 compilation of librustc_data_structures/profiling.rs
I'm trying to use rustfmt from a wasm app. I ran into this compilation problem https://github.com/rust-lang/rustfmt/issues/4132 and after investigating, it looked like just adjusting a few cfg's. I based it on how measureme added support in https://github.com/rust-lang/measureme/pull/43.
My testing on my macbook was just that librustc_data_structures builds now with both:
- cargo build
- cargo build --target wasm32-unknown-unknown
Lint must_use on mem::replace
This adds a hint on `mem::replace`, "if you don't need the old value,
you can just assign the new value directly". This is in similar spirit
to the `must_use` on `ManuallyDrop::take`.
Remove `BodyAndCache`
...returning to the original approach using interior mutability within `Body`. This simplifies the API at the cost of some uncontended mutex locks when the parallel compiler is enabled.
The current API requires you to either have a mutable reference to `Body` (`&mut BodyAndCache`), or to compute the predecessor graph ahead of time by creating a `ReadOnlyBodyAndCache`. This is not a good fit for, e.g., the dataflow framework, which
1. does not mutate the MIR
2. only sometimes needs the predecessor graph (for backward dataflow problems)
Rollup of 4 pull requests
Successful merges:
- #71280 (Miri: mplace_access_checked: offer option to force different alignment on place)
- #71336 (Exhaustively match on `{Statement,Terminator}Kind` during const checking)
- #71370 (Added detailed error code explanation for issue E0696 in Rust compiler.)
- #71401 (visit_place_base is just visit_local)
Failed merges:
r? @ghost
