ICE during async closure experiments #127331
Labels
C-bug
Category: This is a bug.
I-ICE
Issue: The compiler panicked, giving an Internal Compilation Error (ICE) ❄️
T-compiler
Relevant to the compiler team, which will review and decide on the PR/issue.
Comments
chadaustin
added
C-bug
rustbot
added
the
needs-triage
|
Oh, here's the error message: |
Yep, I can't fix this with no code. It's not clear what the issue is from the ICE backtrace alone. |
|
you could probably fix #125099 and it would fix this one as well? |
|
Clone the batch-channel project and check out the async-closure-autobatch: Then cargo check the throughput benchmark: |
saethlin
added
A-async-closures
|
Self-contained reproduction (probably can be minimized further): use std::future::Future;
trait AsyncCallback<A>: FnOnce(A) -> Self::Fut {
type Fut: Future<Output = ()>;
}
impl<A, Out: Future<Output = ()>, F: FnOnce(A) -> Out> AsyncCallback<A> for F {
type Fut = Out;
}
struct BatchSender<T>(T);
trait ChannelSender<T>: Clone + Send {
type BatchSender;
fn autobatch<F>(self, f: F) -> impl Future<Output = ()> + Send
where
F: for<'a> AsyncCallback<&'a mut Self::BatchSender>;
}
struct Sender<T>(T);
impl<T: Send> ChannelSender<T> for Sender<T> {
type BatchSender = BatchSender<T>;
fn autobatch<F>(self, f: F) -> impl Future<Output = ()> + Send
where
F: for<'a> AsyncCallback<&'a mut Self::BatchSender>,
{
async {}
}
}Error outputNotably doesn't require the async_closure feature. @rustbot label -F-async_closure |
|
Thanks!! |
|
Minimized (I can't get it smaller than this) and got rid of dependence on std: trait MyFn<T> {
type Output;
}
trait MyFnAlias<A>: MyFn<A, Output = Self::OutputAlias> {
type OutputAlias;
}
impl<A, F: MyFn<A>> MyFnAlias<A> for F {
type OutputAlias = F::Output;
}
struct Thing;
trait Trait {}
impl Trait for Thing {}
trait ChannelSender {
type Arg;
fn autobatch<F>(self) -> impl Trait
where
F: MyFnAlias<Self::Arg>;
}
struct Sender;
impl ChannelSender for Sender {
type Arg = i32;
fn autobatch<F>(self) -> impl Trait
where
F: MyFnAlias<Self::Arg>,
{
Thing
}
}I believe this code should compile. |
|
Oh yeah, this has nothing to do with async closures at all. This is a shortcoming of the old trait solver having to do with -- and pardon for my technical jargon -- not being able to normalize param envs correctly. I think there should be other issues like this, probably labeled fixed-by-next-solver... I'll look into the ICE though, since I think we crash the compiler when trying to report the error lol. |
bors
added a commit
to rust-lang-ci/rust
that referenced
this issue
Jul 7, 2024
…lse-positive, r=oli-obk Don't try to label `ObligationCauseCode::CompareImplItem` for an RPITIT, since it has no name The old (current) trait solver has a limitation that when a where clause in param-env must be normalized using the same where clause, then we get spurious errors in `normalize_param_env_or_error`. I don't think there's an issue tracking it, but it's the root cause for many of the "fixed-by-next-solver" labeled issues. Specifically, these errors may occur when checking predicate entailment of the GAT that comes out of desugaring RPITITs. Since we use `ObligationCauseCode::CompareImplItem` for these predicates, we try calling `item_name` on an RPITIT which fails, since the RPITIT has no name. We simply suppress this logic when we're reporting a predicate entailment error for an RPITIT. RPITITs should never have predicate entailment errors, *by construction*, but they may due to this bug in the old solver. Addresses the ICE in rust-lang#127331, though doesn't fix the underlying issue (which is fundamental to the old solver). r? types
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
rustc-ice-2024-07-04T18_10_25-10188.txt
Code
Minimizing the code is nontrivial. Let me know if it is required to make a branch that reproduces the issue.
Meta
Error output
See attached rust-ice.
The text was updated successfully, but these errors were encountered: