[RalfJung]

This basically means a nicer error for #121610, see the new test in tests/ui/consts/refs-to-cell-in-final.rs.

Currently we have a mismatch between the dynamic semantics of & (things might be mutable when the pointee type is !Freeze, no matter the pointee value) and const-qualif (doing value-based reasoning). This removes that mismatch. I don't think this can break anything that builds on current nightly; any such code would already later run into the error that the interner complains about the mutable pointer (that's exactly #121610). So a crater run makes little sense.

Overall this peanuts compared to the actual problem with value-based reasoning for interior mutability in promotion (rust-lang/unsafe-code-guidelines#493). I have no idea how to resolve that. But at least now the use of the problematic qualif is down to 1...

The first commit is independent, I just got worried about promotion of unions and was happy to notice that at least there we don't do value-based reasoning.

r? @oli-obk

[RalfJung]

Lol, so much for this not being a breaking change. It breaks bootstraping.^^

error[E0492]: constants cannot refer to interior mutable data
  --> /home/r/.cargo/registry/src/index.crates.io-6f17d22bba15001f/litemap-0.7.2/src/store/slice_impl.rs:15:33
   |
15 |     const EMPTY: &'a [(K, V)] = &[];
   |                                 ^^^ this borrow of an interior mutable value may end up in the final value

Why does that not get promoted, though...?

[RalfJung]

Surrounding code can be seen here.

This does not get promoted because the type of the slice is not known to be 'static. Interesting, I didn't realize such lifetime constraints factor into promotion decisions...

[RalfJung]

Very strange, this breaks the following code:

const NONE: &'static Option<AtomicUsize> = &None;

This gets promoted, but then there are reborrows in the const itself. But reborrows should be ignored by const-checking...

[RalfJung]

Oh wait is const-checking happening before promotion?

[RalfJung]

Hm yes that may be the case. In that case I am very surprised that this does not lead to more errors... now I wonder what crater would say.
@bors try

[bors]

⌛ Trying commit ac1ed22 with merge ca7d067...

[RalfJung]

Ah wait this doesn't entirely remove value-based reasoning from promotion, it only makes that not work inside const any more. Never mind.

It is probably futile to use different checks in promotion and const-checking here...

When it comes to promotion, even rustc itself relies on value-based reasoning:

rust/compiler/rustc_pattern_analysis/src/pat.rs

Lines 229 to 234 in be29cd1

	pub(crate) fn ctor(self) -> &'p Constructor<Cx> {
	match self {
	PatOrWild::Wild => &Wildcard,
	PatOrWild::Pat(pat) => pat.ctor(),
	}
	}

Wildcard here is of type Constructor<Cx> which involves some associated types of Cx which we do not know to be Freeze at this point.

[rustbot]

Some changes occurred to MIR optimizations

cc @rust-lang/wg-mir-opt

[rust-log-analyzer]

The job x86_64-gnu-llvm-16 failed! Check out the build log: (web) (plain)

Click to see the possible cause of the failure (guessed by this bot)

#12 writing image sha256:297e8dfe892dd1c08dbbf1d4b9bc5c25bff8b8b29a8e84fa658ba844013243ef done
#12 naming to docker.io/library/rust-ci done
#12 DONE 10.2s
##[endgroup]
Setting extra environment values for docker:  --env ENABLE_GCC_CODEGEN=1 --env GCC_EXEC_PREFIX=/usr/lib/gcc/
[CI_JOB_NAME=x86_64-gnu-llvm-16]
##[group]Clock drift check
  local time: Sat Mar  2 09:50:48 UTC 2024
  network time: Sat, 02 Mar 2024 09:50:48 GMT
  network time: Sat, 02 Mar 2024 09:50:48 GMT
##[endgroup]
sccache: Starting the server...
##[group]Configure the build
configure: processing command line
configure: 
configure: build.configure-args := ['--build=x86_64-unknown-linux-gnu', '--llvm-root=/usr/lib/llvm-16', '--enable-llvm-link-shared', '--set', 'rust.thin-lto-import-instr-limit=10', '--set', 'change-id=99999999', '--enable-verbose-configure', '--enable-sccache', '--disable-manage-submodules', '--enable-locked-deps', '--enable-cargo-native-static', '--set', 'rust.codegen-units-std=1', '--set', 'dist.compression-profile=balanced', '--dist-compression-formats=xz', '--set', 'build.optimized-compiler-builtins', '--disable-dist-src', '--release-channel=nightly', '--enable-debug-assertions', '--enable-overflow-checks', '--enable-llvm-assertions', '--set', 'rust.verify-llvm-ir', '--set', 'rust.codegen-backends=llvm,cranelift,gcc', '--set', 'llvm.static-libstdcpp', '--enable-new-symbol-mangling']
configure: target.x86_64-unknown-linux-gnu.llvm-config := /usr/lib/llvm-16/bin/llvm-config
configure: llvm.link-shared     := True
configure: rust.thin-lto-import-instr-limit := 10
configure: change-id            := 99999999
---
error[E0515]: cannot return value referencing temporary value
   --> compiler/rustc_pattern_analysis/src/pat.rs:229:9
    |
229 | /         match self {
230 | |             PatOrWild::Wild => &Wildcard,
    | |                                 -------- temporary value created here
231 | |             PatOrWild::Pat(pat) => pat.ctor(),
232 | |         }
    | |_________^ returns a value referencing data owned by the current function
For more information about this error, try `rustc --explain E0515`.
error: could not compile `rustc_pattern_analysis` (lib) due to 1 previous error
warning: build failed, waiting for other jobs to finish...
Build completed unsuccessfully in 0:07:39

[RalfJung]

I don't think there's anything we can do here, unless and until we decide to properly tackle value-based reasoning for interior mutability...

[RalfJung]

Oh wait is const-checking happening before promotion?

For the record -- yes it does: mir_promoted first calls mir_const_qualif, which does const-checking, and then it runs the promote_consts pass.