leptos regression #173

compiler-errors · 2025-04-11T16:43:22Z

edit: While this looks like a regression with tachys, turns out that leptos has its own fork of tachys: https://github.com/leptos-rs/leptos/tree/main/tachys

~~Not yet minimized.~~

https://crater-reports.s3.amazonaws.com/pr-133502-1/try%23fa8e241660363f48d64b66b05eea58c93ab828fb/reg/leptos-0.7.8/log.txt

[INFO] [stdout] error[E0391]: cycle detected when computing type of `reactive_graph::<impl at /opt/rustwide/cargo-home/registry/src/index.crates.io-1949cf8c6b5b557f/tachys-0.1.8/src/reactive_graph/mod.rs:124:1: 128:23>::{synthetic#0}`
[INFO] [stdout]    --> /opt/rustwide/cargo-home/registry/src/index.crates.io-1949cf8c6b5b557f/tachys-0.1.8/src/reactive_graph/mod.rs:138:5
[INFO] [stdout]     |
[INFO] [stdout] 138 |     async fn resolve(mut self) -> Self::AsyncOutput {
[INFO] [stdout]     |     ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
[INFO] [stdout]     |
[INFO] [stdout] note: ...which requires comparing an impl and trait method signature, inferring any hidden `impl Trait` types in the process...
[INFO] [stdout]    --> /opt/rustwide/cargo-home/registry/src/index.crates.io-1949cf8c6b5b557f/tachys-0.1.8/src/reactive_graph/mod.rs:138:5
[INFO] [stdout]     |
[INFO] [stdout] 138 |     async fn resolve(mut self) -> Self::AsyncOutput {
[INFO] [stdout]     |     ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
[INFO] [stdout] note: ...which requires computing type of `reactive_graph::<impl at /opt/rustwide/cargo-home/registry/src/index.crates.io-1949cf8c6b5b557f/tachys-0.1.8/src/reactive_graph/mod.rs:124:1: 128:23>::resolve::{opaque#0}`...
[INFO] [stdout]    --> /opt/rustwide/cargo-home/registry/src/index.crates.io-1949cf8c6b5b557f/tachys-0.1.8/src/reactive_graph/mod.rs:138:5
[INFO] [stdout]     |
[INFO] [stdout] 138 |     async fn resolve(mut self) -> Self::AsyncOutput {
[INFO] [stdout]     |     ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
[INFO] [stdout] note: ...which requires computing type of opaque `reactive_graph::<impl at /opt/rustwide/cargo-home/registry/src/index.crates.io-1949cf8c6b5b557f/tachys-0.1.8/src/reactive_graph/mod.rs:124:1: 128:23>::resolve::{opaque#0}`...
[INFO] [stdout]    --> /opt/rustwide/cargo-home/registry/src/index.crates.io-1949cf8c6b5b557f/tachys-0.1.8/src/reactive_graph/mod.rs:138:5
[INFO] [stdout]     |
[INFO] [stdout] 138 |     async fn resolve(mut self) -> Self::AsyncOutput {
[INFO] [stdout]     |     ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
[INFO] [stdout] note: ...which requires borrow-checking `reactive_graph::<impl at /opt/rustwide/cargo-home/registry/src/index.crates.io-1949cf8c6b5b557f/tachys-0.1.8/src/reactive_graph/mod.rs:124:1: 128:23>::resolve`...
[INFO] [stdout]    --> /opt/rustwide/cargo-home/registry/src/index.crates.io-1949cf8c6b5b557f/tachys-0.1.8/src/reactive_graph/mod.rs:138:5
[INFO] [stdout]     |
[INFO] [stdout] 138 |     async fn resolve(mut self) -> Self::AsyncOutput {
[INFO] [stdout]     |     ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
[INFO] [stdout] note: ...which requires promoting constants in MIR for `reactive_graph::<impl at /opt/rustwide/cargo-home/registry/src/index.crates.io-1949cf8c6b5b557f/tachys-0.1.8/src/reactive_graph/mod.rs:124:1: 128:23>::resolve`...
[INFO] [stdout]    --> /opt/rustwide/cargo-home/registry/src/index.crates.io-1949cf8c6b5b557f/tachys-0.1.8/src/reactive_graph/mod.rs:138:5
[INFO] [stdout]     |
[INFO] [stdout] 138 |     async fn resolve(mut self) -> Self::AsyncOutput {
[INFO] [stdout]     |     ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
[INFO] [stdout] note: ...which requires checking if `reactive_graph::<impl at /opt/rustwide/cargo-home/registry/src/index.crates.io-1949cf8c6b5b557f/tachys-0.1.8/src/reactive_graph/mod.rs:124:1: 128:23>::resolve` contains FFI-unwind calls...
[INFO] [stdout]    --> /opt/rustwide/cargo-home/registry/src/index.crates.io-1949cf8c6b5b557f/tachys-0.1.8/src/reactive_graph/mod.rs:138:5
[INFO] [stdout]     |
[INFO] [stdout] 138 |     async fn resolve(mut self) -> Self::AsyncOutput {
[INFO] [stdout]     |     ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
[INFO] [stdout] note: ...which requires building MIR for `reactive_graph::<impl at /opt/rustwide/cargo-home/registry/src/index.crates.io-1949cf8c6b5b557f/tachys-0.1.8/src/reactive_graph/mod.rs:124:1: 128:23>::resolve`...
[INFO] [stdout]    --> /opt/rustwide/cargo-home/registry/src/index.crates.io-1949cf8c6b5b557f/tachys-0.1.8/src/reactive_graph/mod.rs:138:5
[INFO] [stdout]     |
[INFO] [stdout] 138 |     async fn resolve(mut self) -> Self::AsyncOutput {
[INFO] [stdout]     |     ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
[INFO] [stdout] note: ...which requires match-checking `reactive_graph::<impl at /opt/rustwide/cargo-home/registry/src/index.crates.io-1949cf8c6b5b557f/tachys-0.1.8/src/reactive_graph/mod.rs:124:1: 128:23>::resolve`...
[INFO] [stdout]    --> /opt/rustwide/cargo-home/registry/src/index.crates.io-1949cf8c6b5b557f/tachys-0.1.8/src/reactive_graph/mod.rs:138:5
[INFO] [stdout]     |
[INFO] [stdout] 138 |     async fn resolve(mut self) -> Self::AsyncOutput {
[INFO] [stdout]     |     ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
[INFO] [stdout] note: ...which requires type-checking `reactive_graph::<impl at /opt/rustwide/cargo-home/registry/src/index.crates.io-1949cf8c6b5b557f/tachys-0.1.8/src/reactive_graph/mod.rs:124:1: 128:23>::resolve`...
[INFO] [stdout]    --> /opt/rustwide/cargo-home/registry/src/index.crates.io-1949cf8c6b5b557f/tachys-0.1.8/src/reactive_graph/mod.rs:138:5
[INFO] [stdout]     |
[INFO] [stdout] 138 |     async fn resolve(mut self) -> Self::AsyncOutput {
[INFO] [stdout]     |     ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
[INFO] [stdout]     = note: ...which again requires computing type of `reactive_graph::<impl at /opt/rustwide/cargo-home/registry/src/index.crates.io-1949cf8c6b5b557f/tachys-0.1.8/src/reactive_graph/mod.rs:124:1: 128:23>::{synthetic#0}`, completing the cycle
[INFO] [stdout] note: cycle used when checking assoc item `html::<impl at /opt/rustwide/cargo-home/registry/src/index.crates.io-1949cf8c6b5b557f/tachys-0.1.8/src/html/mod.rs:53:1: 53:28>::resolve` is compatible with trait definition
[INFO] [stdout]    --> /opt/rustwide/cargo-home/registry/src/index.crates.io-1949cf8c6b5b557f/tachys-0.1.8/src/html/mod.rs:60:5
[INFO] [stdout]     |
[INFO] [stdout] 60  |     async fn resolve(self) -> Self::AsyncOutput {
[INFO] [stdout]     |     ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
[INFO] [stdout]     = note: see https://rustc-dev-guide.rust-lang.org/overview.html#queries and https://rustc-dev-guide.rust-lang.org/query.html for more information

The text was updated successfully, but these errors were encountered:

compiler-errors · 2025-04-11T16:43:47Z

Smells like #158 though.

edit: possibly not. IDK.

compiler-errors · 2025-04-11T17:51:28Z

Edit: THIS IS ON CRATES.IO'S VERSION OF TACHYS, NOT THE FORK!

Interestingly enough, there are more/different regressions when building tachys on nightly with the new solver:

error[E0275]: overflow evaluating the requirement `<InertElement as AddAnyAttr>::Output<NewAttr>: RenderHtml`
   --> src/html/mod.rs:141:32
    |
141 |         Self::Output<NewAttr>: RenderHtml,
    |                                ^^^^^^^^^^
    |
note: the requirement `<InertElement as AddAnyAttr>::Output<NewAttr>: RenderHtml` appears on the `impl`'s method `add_any_attr` but not on the corresponding trait's method
   --> src/view/add_attr.rs:16:8
    |
11  | pub trait AddAnyAttr {
    |           ---------- in this trait
...
16  |     fn add_any_attr<NewAttr: Attribute>(
    |        ^^^^^^^^^^^^ this trait's method doesn't have the requirement `<InertElement as AddAnyAttr>::Output<NewAttr>: RenderHtml`

error[E0275]: overflow evaluating the requirement `<InertElement as AddAnyAttr>::Output<NewAttr>: RenderHtml`
   --> src/html/mod.rs:141:32
    |
141 |         Self::Output<NewAttr>: RenderHtml,
    |                                ^^^^^^^^^^

error[E0275]: overflow evaluating the requirement `<InertElement as AddAnyAttr>::Output<NewAttr>: Sized`
   --> src/html/mod.rs:139:10
    |
139 |     ) -> Self::Output<NewAttr>
    |          ^^^^^^^^^^^^^^^^^^^^^
    |
    = note: the return type of a function must have a statically known size

error[E0275]: overflow evaluating the requirement `<InertElement as AddAnyAttr>::Output<NewAttr> well-formed`
   --> src/html/mod.rs:139:10
    |
139 |     ) -> Self::Output<NewAttr>
    |          ^^^^^^^^^^^^^^^^^^^^^

error[E0275]: overflow evaluating the requirement `<InertElement as AddAnyAttr>::Output<NewAttr> well-formed`
   --> src/html/mod.rs:141:32
    |
141 |         Self::Output<NewAttr>: RenderHtml,
    |                                ^^^^^^^^^^

error[E0275]: overflow evaluating the requirement `<AnyView as AddAnyAttr>::Output<NewAttr>: RenderHtml`
   --> src/view/any_view.rs:346:32
    |
346 |         Self::Output<NewAttr>: RenderHtml,
    |                                ^^^^^^^^^^
    |
note: the requirement `<AnyView as AddAnyAttr>::Output<NewAttr>: RenderHtml` appears on the `impl`'s method `add_any_attr` but not on the corresponding trait's method
   --> src/view/add_attr.rs:16:8
    |
11  | pub trait AddAnyAttr {
    |           ---------- in this trait
...
16  |     fn add_any_attr<NewAttr: Attribute>(
    |        ^^^^^^^^^^^^ this trait's method doesn't have the requirement `<AnyView as AddAnyAttr>::Output<NewAttr>: RenderHtml`

error[E0275]: overflow evaluating the requirement `<AnyView as AddAnyAttr>::Output<NewAttr>: RenderHtml`
   --> src/view/any_view.rs:346:32
    |
346 |         Self::Output<NewAttr>: RenderHtml,
    |                                ^^^^^^^^^^

error[E0275]: overflow evaluating the requirement `<AnyView as AddAnyAttr>::Output<NewAttr>: Sized`
   --> src/view/any_view.rs:344:10
    |
344 |     ) -> Self::Output<NewAttr>
    |          ^^^^^^^^^^^^^^^^^^^^^
    |
    = note: the return type of a function must have a statically known size

error[E0275]: overflow evaluating the requirement `<AnyView as AddAnyAttr>::Output<NewAttr> well-formed`
   --> src/view/any_view.rs:344:10
    |
344 |     ) -> Self::Output<NewAttr>
    |          ^^^^^^^^^^^^^^^^^^^^^

error[E0275]: overflow evaluating the requirement `<AnyView as AddAnyAttr>::Output<NewAttr> well-formed`
   --> src/view/any_view.rs:346:32
    |
346 |         Self::Output<NewAttr>: RenderHtml,

compiler-errors · 2025-04-11T18:05:42Z

Minimal-ish:

//@ check-pass
//@ edition: 2018
//@ revisions: current next
//@[next] compile-flags: -Znext-solver
//@ ignore-compare-mode-next-solver (explicit revisions)

// Regression test for leptos when building it with the new solver. Unlike
// avoid-query-cycle-via-item-bound.rs we never actually use the auto trait bound
// of the opaque here, as that one is only necessary when trying to normalize the
// `<V as AttributeValue>::{synthetic#0}` RPITIT via the impl. Given the explicit
// `V: AttributeValue` where-bound, this associated item is rigid and we don't
// care whether the impl may apply.

use std::future::Future;

pub trait ReactiveFunction: Send {
    type Output;

    fn invoke(self) -> Self::Output;
}

trait AttributeValue {
    fn resolve(self) -> impl Future<Output = ()> + Send;
}

impl<F, V> AttributeValue for F
where
    F: ReactiveFunction<Output = V>,
    V: AttributeValue,
{
    async fn resolve(self) {
        // this require `Self::Output: `
        self.invoke().resolve().await
    }
}

fn main() {}

compiler-errors · 2025-04-11T18:32:51Z

So, it seems that we're doing "unnecessary work" when normalizing the <V as AttributeValue>::{synthetic#0} RPITIT. Although the V: AttributeValue where clause in the impl should keep that associated type as rigid, we're still assembling the impl candidate which inevitably leads to a cycle error.

We could perhaps rework candidate assembly for NormalizesTo goals to only assemble param-env/alias-bound candidates when the trait goal is TraitGoalProvenVia::ParamEnv/AliasBound. 🤷

…=compiler-errors do not unnecessarily leak auto traits in item bounds fixes rust-lang/trait-system-refactor-initiative#158 Not a fix for rust-lang/trait-system-refactor-initiative#173 as you may have realized/tried yourself, cc rust-lang#139788. However, fixing this feels desirable regardless and I don't see any reason not to. r? `@compiler-errors`

…=compiler-errors do not unnecessarily leak auto traits in item bounds fixes rust-lang/trait-system-refactor-initiative#158 Not a fix for rust-lang/trait-system-refactor-initiative#173 as you may have realized/tried yourself, cc rust-lang#139788. However, fixing this feels desirable regardless and I don't see any reason not to. r? ``@compiler-errors``

…=compiler-errors do not unnecessarily leak auto traits in item bounds fixes rust-lang/trait-system-refactor-initiative#158 Not a fix for rust-lang/trait-system-refactor-initiative#173 as you may have realized/tried yourself, cc rust-lang#139788. However, fixing this feels desirable regardless and I don't see any reason not to. r? ```@compiler-errors```

Rollup merge of rust-lang#139789 - lcnr:opaques-auto-trait-leakage, r=compiler-errors do not unnecessarily leak auto traits in item bounds fixes rust-lang/trait-system-refactor-initiative#158 Not a fix for rust-lang/trait-system-refactor-initiative#173 as you may have realized/tried yourself, cc rust-lang#139788. However, fixing this feels desirable regardless and I don't see any reason not to. r? ```@compiler-errors```

…-errors do not unnecessarily leak auto traits in item bounds fixes rust-lang/trait-system-refactor-initiative#158 Not a fix for rust-lang/trait-system-refactor-initiative#173 as you may have realized/tried yourself, cc #139788. However, fixing this feels desirable regardless and I don't see any reason not to. r? ```@compiler-errors```

Don't assemble non-env/bound candidates if projection is rigid Putting this up for an initial review, it's still missing comments, clean-up, and possibly a tweak to deal with ambiguities in the `BestObligation` folder. This PR fixes rust-lang/trait-system-refactor-initiative#173. Specifically, we're creating an unnecessary query cycle in normalization by assembling an *impl candidate* even if we know later on during `merge_candidates` that we'll be filtering out that impl candidate. This PR adjusts the `merge_candidates` to assemble *only* env/bound candidates if we have `TraitGoalProvenVia::ParamEnv | TraitGoalProvenVia::AliasBound`. I'll leave some thoughts/comments in the code. r? lcnr

compiler-errors changed the title ~~tachys regression~~ leptos ~~tachys~~ regression Apr 11, 2025

compiler-errors changed the title ~~leptos ~~tachys~~ regression~~ leptos regression Apr 11, 2025

compiler-errors linked a pull request Apr 13, 2025 that will close this issue

Don't assemble non-env/bound candidates if projection is rigid rust-lang/rust#139762

Open

lcnr added this to -Znext-solver=globally Apr 14, 2025

github-project-automation bot moved this to unknown in -Znext-solver=globally Apr 14, 2025

lcnr added the from-crater A regression found via a crater run, not part of our test suite label Apr 14, 2025

lcnr mentioned this issue Apr 14, 2025

opaque type leakage and RPITIT normalization rust-lang/rust#139788

Open

lcnr moved this from unknown to todo in -Znext-solver=globally Apr 14, 2025

lcnr mentioned this issue Apr 14, 2025

do not unnecessarily leak auto traits in item bounds rust-lang/rust#139789

Merged

compiler-errors self-assigned this Apr 14, 2025

lcnr moved this from todo to in progress in -Znext-solver=globally Apr 16, 2025

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

leptos regression #173

leptos regression #173

compiler-errors commented Apr 11, 2025 •

edited

Loading

compiler-errors commented Apr 11, 2025 •

edited

Loading

compiler-errors commented Apr 11, 2025 •

edited

Loading

compiler-errors commented Apr 11, 2025 •

edited by lcnr

Loading

compiler-errors commented Apr 11, 2025

leptos regression #173

leptos regression #173

Comments

compiler-errors commented Apr 11, 2025 • edited Loading

compiler-errors commented Apr 11, 2025 • edited Loading

compiler-errors commented Apr 11, 2025 • edited Loading

compiler-errors commented Apr 11, 2025 • edited by lcnr Loading

compiler-errors commented Apr 11, 2025

compiler-errors commented Apr 11, 2025 •

edited

Loading

compiler-errors commented Apr 11, 2025 •

edited

Loading

compiler-errors commented Apr 11, 2025 •

edited

Loading

compiler-errors commented Apr 11, 2025 •

edited by lcnr

Loading