---

9.28 math.h ¶

POSIX specification:
https://pubs.opengroup.org/onlinepubs/9799919799/basedefs/math.h.html

Gnulib module: math-h

Portability problems fixed by Gnulib:

• The conversion from int to long double in incorrect on some platforms: glibc 2.7 on Linux/SPARC64.
• The macro NAN is not defined on some platforms: OpenBSD 4.0, AIX 5.1.
• The macro NAN is not exposed outside of C99 compilation on some platforms: glibc.
• The macros NAN and HUGE_VAL expand to a function address rather than a floating point constant on some platforms: Solaris 10.
• The macros HUGE_VALF and HUGE_VALL are not defined on some platforms: glibc/HPPA, glibc/SPARC, AIX 5.1, Solaris 9, MSVC 9.
• The macros FP_ILOGB0 and FP_ILOGBNAN are not defined on some platforms: NetBSD 5.1, AIX 5.1, Solaris 9, MSVC 9.
• HUGE_VAL, HUGE_VALF, HUGE_VALL are not compile time constants on some platforms: FreeBSD 5.2, AIX 7.1.
• INFINITY and NAN are not compile time constants on some platforms: FreeBSD 5.2, AIX 7.1.
• INFINITY and NAN are of type double instead of float on some platforms: FreeBSD 7.1, mingw 5.0.
• The macros NAN, HUGE_VALL, and INFINITY are not defined on some platforms: OpenVMS.

Portability problems not fixed by Gnulib:

• NAN is not a compile time constant with some compilers: OpenVMS.
• The macro or variable math_errhandling is not defined on some platforms: glibc 2.11, OpenBSD 4.9, NetBSD 5.1, UP-UX 11, Cygwin 1.7.9, mingw, MSVC 9.
• The macros FP_ILOGB0 and FP_ILOGBNAN have wrong values on some platforms: Haiku before 2023-04-08.