Add ignore-rustc-debug-assertions to tests/ui/associated-consts/issue-93775.rs

[DianQK]

Closes #132111. Closes #133432.

I think this test case is flaky because the recursive calls happen to hit the upper limit of the call stack.

IMO, this may not be an issue, as it's reasonable for overly complex code to require additional build configurations (such as increasing the call stack size).

After set rust.debug-assertions is true, the test case requires a larger call stack, so disable it on rust.debug-assertions=true.

r? jieyouxu

try-job: x86_64-msvc
try-job: i686-msvc

[jieyouxu]

I think moving to crashes is reasonable. Looking at this version, it should hit the limits across platforms much more reliably?

I'll run two msvc try jobs (the try jobs succeeding doesn't guarantee they won't be flaky, but just as a basic sanity check). If they come back green, then please r=me.

@bors try

[bors]

⌛ Trying commit 091613d with merge f4e6d38...

[bors]

☀️ Try build successful - checks-actions
Build commit: f4e6d38 (f4e6d38f09ebf535987fd60b27210194a432cca1)

[DianQK]

🟩
@bors r=jieyouxu

[bors]

📌 Commit 091613d has been approved by jieyouxu

It is now in the queue for this repository.

[matthiaskrgr]

@bors r-

#134613 (comment)

[DianQK]

I will add RUST_MIN_STACK later.

@rustbot author

[DianQK]

When enable rust.debug-assertions, I can notice it requires larger call stacks to handle it, however if I disable it, rustc can even compile the following code:

#![recursion_limit = "8193"]

use std::marker::PhantomData;

struct Z;
struct S<T>(PhantomData<T>);

type Nested4<T> = S<S<S<S<T>>>>;
type Nested16<T> = Nested4<Nested4<Nested4<Nested4<T>>>>;
type Nested64<T> = Nested16<Nested16<Nested16<Nested16<T>>>>;
type Nested256<T> = Nested64<Nested64<Nested64<Nested64<T>>>>;
type Nested1024<T> = Nested256<Nested256<Nested256<Nested256<T>>>>;
type Nested4096<T> = Nested1024<Nested1024<Nested1024<Nested1024<T>>>>;
type Nested8192<T> = Nested4096<Nested4096<T>>;

type Nested = Nested8192<Z>;

trait AsNum {
    const NUM: u32;
}

impl AsNum for Z {
    const NUM: u32 = 0;
}

impl<T: AsNum> AsNum for S<T> {
    const NUM: u32 = T::NUM + 1;
}

fn main() {
    let _ = Nested::NUM;
}

$ time rustc +nightly 93775.rs --out-dir /tmp/93775
rustc +nightly 93775.rs --out-dir /tmp/93775  17.96s user 79.23s system 99% cpu 1:37.74 total

So I disable the test case when disable rust.debug-assertions.

[jieyouxu]

Huh...

[jieyouxu]

I guess we'll have to see in full CI... r=me after PR CI is green.

@bors rollup=iffy

[DianQK]

I guess we'll have to see in full CI... r=me after PR CI is green.

IIUC, only tasks that have DEPLOY environment or running on apple platform disable debug-assertions.
Free to try them if you'd like.

[DianQK]

BTW, I can reproduce the crash at 091613d if I enable debug-assertions on my local Mac mini.

[DianQK]

🟩
@bors r=jieyouxu

[bors]

📌 Commit 4dca485 has been approved by jieyouxu

It is now in the queue for this repository.

[jieyouxu]

IIUC, only tasks that have DEPLOY environment or running on apple platform disable debug-assertions. Free to try them if you'd like.

I marked this PR as iffy, if it fails full CI, oh well :3

[bors]

⌛ Testing commit 4dca485 with merge 66bb586...

[bors]

☀️ Test successful - checks-actions
Approved by: jieyouxu
Pushing 66bb586 to master...

[rust-timer]

Finished benchmarking commit (66bb586): comparison URL.

Overall result: no relevant changes - no action needed

@rustbot label: -perf-regression

Instruction count

This benchmark run did not return any relevant results for this metric.

Max RSS (memory usage)

Results (secondary 3.4%)

This is a less reliable metric that may be of interest but was not used to determine the overall result at the top of this comment.

	mean	range	count
Regressions ❌ (primary)	-	-	0
Regressions ❌ (secondary)	3.4%	[3.4%, 3.4%]	1
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-	-	0
All ❌✅ (primary)	-	-	0

Cycles

Results (secondary -3.3%)

This is a less reliable metric that may be of interest but was not used to determine the overall result at the top of this comment.

	mean	range	count
Regressions ❌ (primary)	-	-	0
Regressions ❌ (secondary)	-	-	0
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-3.3%	[-3.8%, -2.1%]	11
All ❌✅ (primary)	-	-	0

Binary size

This benchmark run did not return any relevant results for this metric.

Bootstrap: 763.42s -> 762.559s (-0.11%)
Artifact size: 330.55 MiB -> 330.61 MiB (0.02%)