Stabilize RFC3324 dyn upcasting coercion

[Urgau]

This PR stabilize the trait_upcasting feature, aka rust-lang/rfcs#3324.

The FCP was completed here: #65991 (comment).

And also remove the deref_into_dyn_supertrait lint which is now handled by dyn upcasting coercion.

Heavily inspired by #101718
Fixes #65991

[rustbot]

r? @b-naber

(rustbot has picked a reviewer for you, use r? to override)

[rustbot]

The Miri subtree was changed

cc @rust-lang/miri

[lcnr]

Should we keep the deref_into_dyn_supertrait lint around, maybe changing its wording? These impls probably don't do what the user expects them to, given that they can't be used implicitly. I feel like pointing out "hey, this impl is now implicit via a coercion, consider removing it" is preferable to completely removing the lint

[Urgau]

Sure, if those impls don't do what the user expects we should tell them.

What about this diagnostic change?

- warning: `dyn Foo<'_>` implements `Deref` with supertrait `Bar<'_>` as target
+ warning: `dyn Foo<'_>` implements `Deref` with supertrait `Bar<'_>` as target is shadowed by coercion
   --> $DIR/migrate-lint-deny-regions.rs:8:1
    |
 LL | impl<'a> Deref for dyn Foo<'a> {
    | ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
 ...
 LL |     type Target = dyn Bar<'a>;
    |     -------------------------- target type is set here
    |
-   = warning: this will change its meaning in a future release!
-   = note: for more information, see issue #89460 <https://github.com/rust-lang/rust/issues/89460>
+   = help: consider removing this implementation in favour for the implicit coercion

[lcnr]

maybe something like

- warning: `dyn Foo<'_>` implements `Deref` with supertrait `Bar<'_>` as target
+ warning: this `Deref` implementation is covered by the implicit super-trait coercion
   --> $DIR/migrate-lint-deny-regions.rs:8:1
    |
 LL | impl<'a> Deref for dyn Foo<'a> {
    | ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
 ...
 LL |     type Target = dyn Bar<'a>;
    |     -------------------------- target type is set here
    |
-   = warning: this will change its meaning in a future release!
-   = note: for more information, see issue #89460 <https://github.com/rust-lang/rust/issues/89460>
+   = help: consider removing this implementation or replacing it with a method instead

something like this maybe? 🤔

[Urgau]

Done. I've restored the lint, removed the future incompatibility mark and changed the diagnostic output as discussed.

[WaffleLapkin]

r? WaffleLapkin
r=me with the nit fixed

---

Also, could you open a reference PR editing this page?

[WaffleLapkin]

@bors delegate+

[bors]

✌️ @Urgau, you can now approve this pull request!

If @WaffleLapkin told you to "r=me" after making some further change, please make that change, then do @bors r=@WaffleLapkin

[Urgau]

Also, could you open a reference PR editing this page?

@crlf0710 already has one opened rust-lang/reference#1259 (from his previous attempt as stabilizing this feature).

[Urgau]

@bors r=@WaffleLapkin

[bors]

📌 Commit 4c2d6de has been approved by WaffleLapkin

It is now in the queue for this repository.

[bors]

⌛ Testing commit 4c2d6de with merge 6d2b84b...

[rust-log-analyzer]

A job failed! Check out the build log: (web) (plain)

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

[bors]

☀️ Test successful - checks-actions
Approved by: WaffleLapkin
Pushing 6d2b84b to master...

[Urgau]

@rustbot label +relnotes

[rust-timer]

Finished benchmarking commit (6d2b84b): comparison URL.

Overall result: ❌ regressions - no action needed

@rustbot label: -perf-regression

Instruction count

This is a highly reliable metric that was used to determine the overall result at the top of this comment.

	mean	range	count
Regressions ❌ (primary)	-	-	0
Regressions ❌ (secondary)	0.2%	[0.2%, 0.2%]	1
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-	-	0
All ❌✅ (primary)	-	-	0

Max RSS (memory usage)

Results

This is a less reliable metric that may be of interest but was not used to determine the overall result at the top of this comment.

	mean	range	count
Regressions ❌ (primary)	-	-	0
Regressions ❌ (secondary)	0.6%	[0.4%, 0.7%]	4
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-2.1%	[-2.1%, -2.1%]	1
All ❌✅ (primary)	-	-	0

Cycles

Results

This is a less reliable metric that may be of interest but was not used to determine the overall result at the top of this comment.

	mean	range	count
Regressions ❌ (primary)	-	-	0
Regressions ❌ (secondary)	-	-	0
Improvements ✅ (primary)	-2.1%	[-2.4%, -1.9%]	3
Improvements ✅ (secondary)	-	-	0
All ❌✅ (primary)	-2.1%	[-2.4%, -1.9%]	3

Binary size

This benchmark run did not return any relevant results for this metric.

Bootstrap: 675.017s -> 675.132s (0.02%)
Artifact size: 313.79 MiB -> 313.79 MiB (0.00%)

[gabrielecastellano]

Hi, thanks for this!
I was wondering why this was not included in release 1.75.0.

[Urgau]

I was wondering why this was not included in release 1.75.0.

Because when this PR landed on the 22 of November we were already in track for 1.76.
In general you can look at the milestone of each PR to see in which version it will be included.

[gabrielecastellano]

Thank you for the explanation!