I wrote some shitty code here: https://play.rust-lang.org/?version=nightly&mode=debug&edition=2018&gist=d44eb1015f294e937d86b157339d0978
And it broke the compiler. Apparently by reaching supposedly unreachable code of a parsing function.
Some background can be found at https://users.rust-lang.org/t/the-squeeze-challenge/25762/10
As pointed by @cuviper the problem can be reduced to
trait Howness {}
impl Howness for (){
fn how_are_you(&self->Empty
{
Empty
}
}
I-ICE regression-from-stable-to-beta
Here's the backtrace from the reduced case:
thread 'rustc' panicked at 'internal error: entered unreachable code', src/libsyntax/parse/parser.rs:648:22
stack backtrace:
0: std::sys::unix::backtrace::tracing::imp::unwind_backtrace
at src/libstd/sys/unix/backtrace/tracing/gcc_s.rs:39
1: std::sys_common::backtrace::_print
at src/libstd/sys_common/backtrace.rs:71
2: std::panicking::default_hook::{{closure}}
at src/libstd/sys_common/backtrace.rs:59
at src/libstd/panicking.rs:200
3: std::panicking::default_hook
at src/libstd/panicking.rs:215
4: std::panicking::rust_panic_with_hook
at src/libstd/panicking.rs:478
5: std::panicking::begin_panic
6: syntax::parse::parser::Parser::parse_fn_decl_with_self
7: syntax::parse::parser::Parser::parse_impl_item
8: syntax::parse::parser::Parser::parse_item_implementation
9: syntax::parse::parser::Parser::parse_item_
10: syntax::parse::parser::Parser::parse_item
11: syntax::parse::parser::Parser::parse_mod_items
12: syntax::parse::parser::Parser::parse_crate_mod
13: syntax::parse::parse_crate_from_file
14: rustc_driver::driver::phase_1_parse_input::{{closure}}
15: rustc::util::common::time
16: rustc_driver::driver::phase_1_parse_input
17: rustc_driver::driver::compile_input
18: <scoped_tls::ScopedKey<T>>::set
19: rustc_driver::run_compiler
20: <scoped_tls::ScopedKey<T>>::set
21: syntax::with_globals
22: __rust_maybe_catch_panic
at src/libpanic_unwind/lib.rs:87
23: <F as alloc::boxed::FnBox<A>>::call_box
24: std::sys::unix::thread::Thread::new::thread_start
at /rustc/350674b7180a41c8e508d93c6ab8e203b69d3df7/src/liballoc/boxed.rs:759
at src/libstd/sys_common/thread.rs:14
at src/libstd/sys/unix/thread.rs:80
25: start_thread
26: __clone
Further reduced test case:
impl a {
fn b(self>
}
Woah, how is the compiler even supposed to be able to deal with such malformed syntax?
@OptimisticPeach it can throw an error (as it does on stable), it just shouldn't ICE.
Different possibly related bug:
struct A<B: !C
Backtrace:
thread 'rustc' panicked at 'called `Option::unwrap()` on a `None` value', src/libcore/option.rs:345:21
stack backtrace:
0: std::sys::unix::backtrace::tracing::imp::unwind_backtrace
at src/libstd/sys/unix/backtrace/tracing/gcc_s.rs:39
1: std::sys_common::backtrace::_print
at src/libstd/sys_common/backtrace.rs:71
2: std::panicking::default_hook::{{closure}}
at src/libstd/sys_common/backtrace.rs:59
at src/libstd/panicking.rs:200
3: std::panicking::default_hook
at src/libstd/panicking.rs:215
4: std::panicking::rust_panic_with_hook
at src/libstd/panicking.rs:478
5: std::panicking::continue_panic_fmt
at src/libstd/panicking.rs:385
6: rust_begin_unwind
at src/libstd/panicking.rs:312
7: core::panicking::panic_fmt
at src/libcore/panicking.rs:85
8: core::panicking::panic
at src/libcore/panicking.rs:49
9: syntax::parse::parser::Parser::parse_generic_bounds_common
10: syntax::parse::parser::Parser::parse_generic_params
11: syntax::parse::parser::Parser::parse_generics
12: syntax::parse::parser::Parser::parse_item_implementation
13: syntax::parse::parser::Parser::parse_item_
14: syntax::parse::parser::Parser::parse_item
15: syntax::parse::parser::Parser::parse_mod_items
16: syntax::parse::parser::Parser::parse_crate_mod
17: syntax::parse::parse_crate_from_file
18: rustc_driver::driver::phase_1_parse_input::{{closure}}
19: rustc::util::common::time
20: rustc_driver::driver::phase_1_parse_input
21: rustc_driver::driver::compile_input
22: <scoped_tls::ScopedKey<T>>::set
23: rustc_driver::run_compiler
24: <scoped_tls::ScopedKey<T>>::set
25: syntax::with_globals
26: __rust_maybe_catch_panic
at src/libpanic_unwind/lib.rs:87
27: <F as alloc::boxed::FnBox<A>>::call_box
28: std::sys::unix::thread::Thread::new::thread_start
at /rustc/350674b7180a41c8e508d93c6ab8e203b69d3df7/src/liballoc/boxed.rs:759
at src/libstd/sys_common/thread.rs:14
at src/libstd/sys/unix/thread.rs:80
29: start_thread
30: clone
Thanks to @OptimisticPeach for finding it.
I think that's #58857
@estebank maybe 7451cd8dc0b?
@jethrogb looks likely.
I believe I can have a fix shortly.
Edit:
I believe it is caused when calling unexpected
,
which _expects_ to fail,
but the recovery mechanism makes it succeed
@OptimisticPeach gracefully :)
triage: P-high. estebank is on it so assigning them.
Most helpful comment
Further reduced test case: