Document: WG14 N995

Date: 2003-02-11


Defect Report #2dd

Previous Defect Report < - > Next Defect Report


Submitter: Douglas Walls (US)
Submission Date: 2003-02-11
Source: J11
Reference Document: ISO/IEC WG14 N995
Version: 1.0
Date: 2003-02-11
Subject: Does <math.h> define INT_MIN and INT_MAX?

Summary

<math.h> defines macros in terms of INT_MIN and INT_MAX. <math.h> defines functions returning the value of INT_MIN. It is unclear if inclusion of <math.h> defines INT_MIN and INT_MAX or also includes <limits.h>.

Details

The description of <math.h> in C99 (section 7.12) says that the macros FP_ILOGB0 and FP_ILOGBNAN are defined in <math.h> with values, respectively INT_MIN or -INT_MAX and INT_MAX or INT_MIN, but never says that INT_MIN and INT_MAX are defined in <math.h>.

The synopsis of 7.12.6.5 The ilogb function says:

	#include <math.h>
	int ilogb(double x);
	int ilogbf(float x);
	int ilogbl(long double x);
The description of 7.12.6.5 The ilogb functions says "if x is infinite they compute the value INT_MAX;".

Does this mean that <math.h> includes <limits.h>?

Does this mean that <math.h> defines INT_MIN and INT_MAX?

Suggested Technical Corrigendum



Previous Defect Report < - > Next Defect Report