Match monty's floating point modulo behavior to cpython#498
Open
marcbrooker wants to merge 3 commits into
Open
Match monty's floating point modulo behavior to cpython#498marcbrooker wants to merge 3 commits into
marcbrooker wants to merge 3 commits into
Conversation
…ncated fmod Rust's `%` on f64 is C-style fmod: the remainder takes the sign of the dividend. Python's `%` uses floored division: the remainder takes the sign of the divisor. For example, `-7.0 % 3.0` should be `2.0` (Python) not `-1.0` (C fmod). The difference is always exactly one multiple of the divisor, which is why the fuzzer saw offsets like 9.0, 10.0, 5.0, etc. Fix: add py_float_mod(a, b) that adjusts the C remainder when its sign disagrees with the divisor, applied to all three float-mod arms (float%float, float%int, int%float). Integer modulo already had this correction.
When the floored modulo result is exactly zero, IEEE 754 can leave a stale sign bit from the dividend (e.g. `-1.0 % 1.0` → `-0.0`). CPython normalizes this: the zero result always carries the divisor's sign. Fix: use f64::copysign(r, b) in the zero branch of py_float_mod so the sign bit matches the divisor, consistent with CPython's float_rem.
Merging this PR will not alter performance
Comparing 
Footnotes
|
![cubic-dev-ai[bot]](https://avatars.githubusercontent.com/in/1082092?s=60&v=4){kind=small}
Contributor
There was a problem hiding this comment.
2 issues found across 2 files
Reply with feedback, questions, or to request a fix.
Re-trigger cubic
Codecov Report❌ Patch coverage is
📢 Thoughts on this report? Let us know! |
…n tests py_mod_eq (the COMPARE_MOD_EQ optimization) still used raw Rust `%` for float cases, so the fast path disagreed with py_mod for negative floats. Switch to py_float_mod for consistency. Also strengthen the zero-result test assertions to verify the sign bit via math.copysign, since `== 0.0` cannot distinguish +0.0 from -0.0.
Contributor
There was a problem hiding this comment.
1 issue found across 2 files (changes from recent commits).
Prompt for AI agents (unresolved issues)
Check if these issues are valid — if so, understand the root cause of each and fix them. If appropriate, use sub-agents to investigate and fix each issue separately.
<file name="crates/monty/src/value.rs">
<violation number="1" location="crates/monty/src/value.rs:555">
P1: `py_mod_eq` float branches call `py_float_mod` without zero-division guards, diverging from runtime `%` semantics</violation>
</file>
Reply with feedback, questions, or to request a fix.
Re-trigger cubic
| (Self::Float(v1), Self::Float(v2)) => Some(v1 % v2 == right_value as f64), | ||
| (Self::Float(v1), Self::Int(v2)) => Some(v1 % (*v2 as f64) == right_value as f64), | ||
| (Self::Int(v1), Self::Float(v2)) => Some((*v1 as f64) % v2 == right_value as f64), | ||
| (Self::Float(v1), Self::Float(v2)) => Some(py_float_mod(*v1, *v2) == right_value as f64), |
Contributor
There was a problem hiding this comment.
P1: py_mod_eq float branches call py_float_mod without zero-division guards, diverging from runtime % semantics
Prompt for AI agents
Check if this issue is valid — if so, understand the root cause and fix it. At crates/monty/src/value.rs, line 555:
<comment>`py_mod_eq` float branches call `py_float_mod` without zero-division guards, diverging from runtime `%` semantics</comment>
<file context>
@@ -552,9 +552,9 @@ impl PyTrait<'_> for Value {
- (Self::Float(v1), Self::Float(v2)) => Some(v1 % v2 == right_value as f64),
- (Self::Float(v1), Self::Int(v2)) => Some(v1 % (*v2 as f64) == right_value as f64),
- (Self::Int(v1), Self::Float(v2)) => Some((*v1 as f64) % v2 == right_value as f64),
+ (Self::Float(v1), Self::Float(v2)) => Some(py_float_mod(*v1, *v2) == right_value as f64),
+ (Self::Float(v1), Self::Int(v2)) => Some(py_float_mod(*v1, *v2 as f64) == right_value as f64),
+ (Self::Int(v1), Self::Float(v2)) => Some(py_float_mod(*v1 as f64, *v2) == right_value as f64),
</file context>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary by cubic
Match float modulo to CPython: use floored semantics, normalize zero sign, and align
py_mod_eqfast path to fix wrong results with negative operands and -0.0/+0.0 anomalies.py_float_modfor float%float, float%int, and int%float (remainder takes the divisor’s sign).py_mod_eqfast path to usepy_float_modfor float cases to keep compare-mod-equals consistent.f64::copysignso 0.0 matches the divisor’s sign; strengthened zero-sign tests withmath.copysign.Written for commit 75d71ef. Summary will update on new commits.