llvm/llvm-project

[libc][math] Add missing isnan variants for float16, bfloat16, and float128

Open

#195400 opened on May 2, 2026

View on GitHub
 (16 comments) (0 reactions) (3 assignees)C++ (26,378 stars) (10,782 forks)batch import
good first issuelibc

Description

The current implementations of isnan are limited to float, double, and long double. We should add support for the remaining types (float16, bfloat16, and float128).

Additionally, the existing implementations currently rely on __builtin_isnan. I think refactoring them to use fputil would be better, especially since issignaling already uses fputil.

Contributor guide

Tech stack
cpp
Domain
api
Issue type
feature
DifficultyEstimated implementation difficulty for a new contributor, from 1 for very small changes to 5 for expert-level work.
4
Estimated timeA rough time range for an experienced contributor to investigate, implement, test, and prepare a pull request.
3-5 days
Activity statusHow available the issue appears right now: fresh, active, stale, blocked, or waiting on maintainer input.
active
ClarityHow clearly the issue explains the expected change, acceptance criteria, and next step.
clear
Prerequisites
Familiarity with floating point types (float16, bfloat16, float128)Understanding of C/C++ math function implementationKnowledge of LLVM libc structureExperience with compiler built ins like builtin isnan
Newbie friendlinessA 1-100 score estimating how approachable this issue is for first-time contributors.
35
Research direction
Examine existing isnan implementations for float, double, and long double to understand the pattern. Look at the issignaling implementation to see how it uses fputil. Determine which files need modification (likely in libc/src/math/) for the new types. Check the comments in this issue for any discussion about specific files or approaches. The refactoring to fputil may require understanding of the fputil library.