super_relate_consts ty field equality assert doesnt support <T, const N: T>
#107898
Assignees
Labels
A-const-generics
Area: const generics (parameters and arguments)
C-bug
Category: This is a bug.
F-generic_const_exprs
`#![feature(generic_const_exprs)]`
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
michael-swan
added
C-bug
|
I determined it was due to the above linked-from PR. |
|
Prior to this PR, the above code would emit the following warning instead of generating the ICE: |
BoxyUwU
added
A-const-generics
BoxyUwU
changed the title
super_relate_consts ty field equality assert doesnt support <T, const N: T>
|
This is caused by the fact that |
3 tasks
tautschnig
added a commit
to tautschnig/kani
that referenced
this issue
Apr 15, 2023
Skip over all the nightlies from 2023-02-06 until 2023-02-15 as those
ICE when trying to build kani with:
```
error: internal compiler error: cannot relate constants (Const { ty: fn() -> usize {std::mem::size_of::<[T; N]>}, kind: Value(Branch([])) }, Const { ty: fn() -> usize {std::mem::size_of::<[T; _]>}, kind: Value(Branch([])) }) of different types: fn() -> usize {std::mem::size_of::<[T; N]>} != fn() -> usize {std::mem::size_of::<[T; _]>}
```
This issue was reported upstream as
rust-lang/rust#107898, and fixed in
rust-lang/rust#107940, which isn't part of any
of the above nightlies.
Doing this multi-day update also requires addressing:
Remove some superfluous type parameters from layout.rs rust-lang/rust#107163
Introduce -Zterminal-urls to use OSC8 for error codes rust-lang/rust#107838
s/eval_usize/eval_target_usize/ for clarity rust-lang/rust#108029
Co-authored-by: Qinheping Hu <qinhh@amazon.com>
tautschnig
added a commit
to tautschnig/kani
that referenced
this issue
Apr 15, 2023
Skip over all the nightlies from 2023-02-06 until 2023-02-15 as those
ICE when trying to build kani with:
```
error: internal compiler error: cannot relate constants (Const { ty: fn() -> usize {std::mem::size_of::<[T; N]>}, kind: Value(Branch([])) }, Const { ty: fn() -> usize {std::mem::size_of::<[T; _]>}, kind: Value(Branch([])) }) of different types: fn() -> usize {std::mem::size_of::<[T; N]>} != fn() -> usize {std::mem::size_of::<[T; _]>}
```
This issue was reported upstream as
rust-lang/rust#107898, and fixed in
rust-lang/rust#107940, which isn't part of any
of the above nightlies.
Doing this multi-day update also requires addressing:
Remove some superfluous type parameters from layout.rs rust-lang/rust#107163
Introduce -Zterminal-urls to use OSC8 for error codes rust-lang/rust#107838
s/eval_usize/eval_target_usize/ for clarity rust-lang/rust#108029
Co-authored-by: Qinheping Hu <qinhh@amazon.com>
4 tasks
tautschnig
added a commit
to tautschnig/kani
that referenced
this issue
Apr 15, 2023
Skip over all the nightlies from 2023-02-06 until 2023-02-15 as those
ICE when trying to build kani with:
```
error: internal compiler error: cannot relate constants (Const { ty: fn() -> usize {std::mem::size_of::<[T; N]>}, kind: Value(Branch([])) }, Const { ty: fn() -> usize {std::mem::size_of::<[T; _]>}, kind: Value(Branch([])) }) of different types: fn() -> usize {std::mem::size_of::<[T; N]>} != fn() -> usize {std::mem::size_of::<[T; _]>}
```
This issue was reported upstream as
rust-lang/rust#107898, and fixed in
rust-lang/rust#107940, which isn't part of any
of the above nightlies.
Doing this multi-day update also requires addressing:
Remove some superfluous type parameters from layout.rs rust-lang/rust#107163
Introduce -Zterminal-urls to use OSC8 for error codes rust-lang/rust#107838
s/eval_usize/eval_target_usize/ for clarity rust-lang/rust#108029
Co-authored-by: Qinheping Hu <qinhh@amazon.com>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Code
Meta
rustc --version --verbose:Error output
Backtrace
This occurred in code different from the above but this is a minimal example that causes this failure. I believe I am disobeying some rules in the manner in which
Tis not bound in the underlying type of the trait. This bug appeared in my own code that worked with a previous nightly build. That code does bindTinto the underlying type and still fails. If I bind it in this example with the following changes:the error goes away. I will need to create a minimal example that produces an ICE closer to the failure in my own code but I was struggling to do this. This is at least one ICE of interest and perhaps the unbound
Tshould itself be prevented. I'm not entirely sure. Worthy of note, the following also compiles fine:which omits all of the
generic_const_exprstuff.The text was updated successfully, but these errors were encountered: