NaN > NaN returns true

The following tests do not pass for me, but I think they should, as comparisons with NaN should always simply evaluate to false:

#[test]
fn test_zero_zero() {
    assert_eq!(0.0/0.0 < 0.0/0.0, false);
    assert_eq!(0.0/0.0 > 0.0/0.0, false);
}

#[test]
fn test_nan_nan() {
    assert_eq!(std::f64::NAN < std::f64::NAN, false);
    assert_eq!(std::f64::NAN > std::f64::NAN, false);
}

fn main() {
    println!("Hello, world!");
}

I get this result on both debug and release builds. I'm on an x86_64 running Linux, rustc 1.27.0-nightly (f0fdaba 2018-05-15).

Playground link for this test.

https://play.rust-lang.org/?gist=4d6f7f54553b065615caac65b3cb6222&version=nightly&mode=debug

Affects both debug and release, Nightly and Stable.

Playground:

fn main() {
    let operator = std::f64::NAN > std::f64::NAN;
    let method = std::f64::NAN.gt(&std::f64::NAN);
    println!("{} {}", operator, method); // false false
    assert_eq!(operator, method); // doesn't fail
    assert_eq!(std::f64::NAN > std::f64::NAN, std::f64::NAN.gt(&std::f64::NAN)); // fails
}

~~I think this is an issue with assert_eq! rather than the floating-point comparisons, but I haven't dug into it yet.~~

This seems to be a regression introduced by 1.26.

@oli-obk

Reduced, ~~related to rvalue promotion I guess~~.

extern { fn abort() -> !; }

fn main() {
    let f = &(std::f64::NAN > std::f64::NAN);
    if *f {
        unsafe { abort(); }
    }
}

Edit: Or maybe a bug in Miri (cc @oli-obk)

extern { fn abort() -> !; }

static B: bool = std::f64::NAN > std::f64::NAN;

fn main() {
    if B {
        unsafe { abort(); }
    }
}

Bisection between c83fa5d...097efa9 gives #46882 as the regression PR, as expected.

@eddyb

I guess this logic, which I blindly copied, was never correct to begin with:

https://github.com/rust-lang/rust/blob/master/src/librustc_mir/interpret/operator.rs#L188

I think this is how old const eval did it, and we didn't want to regress that. cc @eddyb

40b118c#diff-580fb7d0b0ede57fb2ce24a5dd33910cL50

Has the comment about existing bad behaviour.

@oli-obk

@oli-obk The old logic used unwrap, meaning NaN > NaN would just ICE (?). I think we can safely replace them directly as PrimVal::from_bool(l > r) etc.

Edit: 🤔 The unwrap is in 40b118c#diff-e847a6da94440f7bf6783206efe4275bL138

The incorrect behavior mentioned in #50811 (comment) was this:

let x: [u8; (std::f64::NAN > std::f64::NAN) as usize] = [1];

If we fix the float comparison behavior, this code will no longer compile. But I think this is clearly a bug fix and breaking code like this is acceptable.

Fixed in #50812

Auto merge of #50812 - kennytm:fix-50756-miri-bad-float-behavior, r=o…

Fix issue rust-lang#50811 (`NaN > NaN` was true).

Auto merge of #51037 - pietroalbini:beta-backports, r=oli-obk

Auto merge of #51037 - pietroalbini:beta-backports, r=pietroalbini

Rollup merge of rust-lang#156317 - lokirithm:reorg-tests-04, r=Kivooeo

Auto merge of #156317 - lokirithm:reorg-tests-04, r=Kivooeo

Unrolled build for #156317

Navigation Menu

Repository navigation

NaN > NaN returns true #50811

Metadata

Assignees

Labels

Type

Projects

Milestone

Relationships

Development

NaN > NaN returns true #50811

Description

Activity

jorendorff commented on May 16, 2018

varkor commented on May 16, 2018

kennytm commented on May 16, 2018

kennytm commented on May 16, 2018

kennytm commented on May 16, 2018

oli-obk commented on May 16, 2018

oli-obk commented on May 16, 2018

kennytm commented on May 16, 2018

kennytm commented on May 16, 2018

nikomatsakis commented on May 17, 2018

Metadata

Metadata

Assignees

Labels

Type

Projects

Milestone

Relationships

Development

Issue actions