fpclassify(3)

✨ fpclassify(3) ✨

okay bestie, so this is the library functions manual page for fpclassify(3) and friends, and honestly? it's giving "float math but make it cute". we're in the libm zone and we're about to find out what kind of vibe your floating-point number is giving 💅

---

🎀 NAME 🎀

fpclassify, isfinite, isnormal, isnan, isinf — floating-point classification macros

like, five little macros whose entire job is answering "what even is this number" 🔍

---

📚 LIBRARY 📚

Math library (libm, -lm)

girl you literally have to link it. -lm. don't forget. i'm not gonna say it again. okay i will. -lm. 🫶

---

🦋 SYNOPSIS 🦋

#include <math.h>

int fpclassify(x);
int isfinite(x);
int isnormal(x);
int isnan(x);
int isinf(x);

💫 Feature Test Macro Requirements for glibc (see feature_test_macros(7)): 💫

ok so before any of these will show up you gotta define the right macro, otherwise glibc is gonna pretend it doesn't know you 🙈 here's the whole list, no skips:

• fpclassify(), isfinite(), isnormal():
  • _ISOC99_SOURCE || _POSIX_C_SOURCE >= 200112L
• isnan():
  • _ISOC99_SOURCE || _POSIX_C_SOURCE >= 200112L
  • || _XOPEN_SOURCE
  • || /* Since glibc 2.19: */ _DEFAULT_SOURCE
  • || /* glibc <= 2.19: */ _BSD_SOURCE || _SVID_SOURCE
• isinf():
  • _ISOC99_SOURCE || _POSIX_C_SOURCE >= 200112L
  • || /* Since glibc 2.19: */ _DEFAULT_SOURCE
  • || /* glibc <= 2.19: */ _BSD_SOURCE || _SVID_SOURCE

---

🌸 DESCRIPTION 🌸

Floating point numbers can have special values, such as infinite or NaN. yes bestie, numbers can be weird. not all floats are chill normal floats — some of them are out here living their own life 💀

With the macro fpclassify(x) you can find out what type x is. The macro takes any floating-point expression as argument. The result is one of the following values:

• FP_NAN — x is "Not a Number". (yes it's literally called that. iconic naming honestly.)
• FP_INFINITE — x is either positive infinity or negative infinity.
• FP_ZERO — x is zero. just vibing at the origin ✨
• FP_SUBNORMAL — x is too small to be represented in normalized format. tiny little guy 🥺
• FP_NORMAL — if nothing of the above is correct then it must be a normal floating-point number. the main character of the float world 👑

The other macros provide a short answer to some standard questions. like a rapid-fire Q&A, group chat style 📱

• isfinite(x) — returns a nonzero value if (fpclassify(x) != FP_NAN && fpclassify(x) != FP_INFINITE)
• isnormal(x) — returns a nonzero value if (fpclassify(x) == FP_NORMAL)
• isnan(x) — returns a nonzero value if (fpclassify(x) == FP_NAN)
• isinf(x) — returns 1 if x is positive infinity, and -1 if x is negative infinity.

---

💅 ATTRIBUTES 💅

For an explanation of the terms used in this section, see attributes(7).

ok so the table got eaten by the vibe check, but here's what it said, i promise i'm not gatekeeping:

• Interface: fpclassify(), isfinite(), isnormal(), isnan(), isinf()
• Attribute: Thread safety
• Value: MT-Safe

like all five of them, safe to use from multiple threads, no cap 🧵✨

---

⭐ STANDARDS ⭐

C11, POSIX.1-2008.

so yes it's officially official. certified. periodt. 👑

---

🌈 HISTORY 🌈

POSIX.1-2001, C99.

In glibc 2.01 and earlier, isinf() returns a nonzero value (actually: 1) if x is positive infinity or negative infinity. (This is all that C99 requires.)

so basically old glibc wasn't out here distinguishing +inf from -inf, it just went "yep that's infinite alright" 🫠 which, technically, is all C99 asked for. the bar was on the floor bestie.

---

🍓 NOTES 🍓

For isinf(), the standards merely say that the return value is nonzero if and only if the argument has an infinite value.

that's it. that's the whole requirement. no direction, no signs, just "is it infinite or nah" 🙈

---

🔥 SEE ALSO 🔥

• finite(3)
• INFINITY(3)
• isgreater(3)
• signbit(3)

---

and that's the whole page bestie 💖 this was Linux man-pages 6.20, dated 2026-02-08 — go forth and classify your floats, you absolute icon, and remember to link -lm 🎀🦋✨