type parameter `E/#1` (E/1) out of range when substituting

[tvallotton]

Code

// mod structs;
// mod table;

use std::env;

type Result<T, E = impl std::error::Error> = std::result::Result<T, E>;

fn main() -> Result<()> {
    let args: Vec<String> = env::args().collect();
    

    println!("{:?}", args.get(0));
    Ok(())
}

Meta

rustc --version --verbose:

rustc 1.53.0-nightly (673d0db5e 2021-03-23)
binary: rustc
commit-hash: 673d0db5e393e9c64897005b470bfeb6d5aec61b
commit-date: 2021-03-23
host: x86_64-apple-darwin
release: 1.53.0-nightly
LLVM version: 12.0.0

Error output

error: internal compiler error: compiler/rustc_middle/src/ty/subst.rs:529:17: type parameter `E/#1` (E/1) out of range when substituting, substs=[()]
 --> src/main.rs:8:14
  |
8 | fn main() -> Result<()> {
  |              ^^^^^^^^^^

thread 'rustc' panicked at 'Box<Any>', /rustc/673d0db5e393e9c64897005b470bfeb6d5aec61b/library/std/src/panic.rs:59:5
note: run with `RUST_BACKTRACE=1` environment variable to display a backtrace

note: the compiler unexpectedly panicked. this is a bug.

note: we would appreciate a bug report: https://github.com/rust-lang/rust/issues/new?labels=C-bug%2C+I-ICE%2C+T-compiler&template=ice.md

note: rustc 1.53.0-nightly (673d0db5e 2021-03-23) running on x86_64-apple-darwin

note: compiler flags: -C embed-bitcode=no -C debuginfo=2 -C incremental --crate-type bin

note: some of the compiler flags provided by cargo are hidden

query stack during panic:
#0 [fn_sig] computing function signature of `main`
#1 [collect_mod_item_types] collecting item types in top-level module
end of query stack
error: aborting due to previous error

error: could not compile `project`

Backtrace

<backtrace>

[tvallotton]

I tried on stable cargo 1.51.0 (43b129a20 2021-03-16), and it didn't work either.

[jyn514]

Slightly smaller:

type Result<T, E = impl std::error::Error> = std::result::Result<T, E>;

fn main() -> Result<()> {
    Ok(())
}

[SNCPlay42]

Another version:

struct Foo<T = impl Copy>(T);

fn x() -> Foo {
    Foo(0)
}

impl Trait should probably not be allowed in generic param defaults, right?

[tvallotton]

Mmm, I intended to write Box<dyn Error> instead of impl Error. Idk if it is still a bug though.