Page History

Integer types in C have both a size and a width precision. The size indicates the number of bytes used by an object, and can be retrieved for any object or type using the sizeof operator.  The width precision of an integer type is the number of bits it uses to represent values, excluding any optional sign and padding bits.

  Padding bits contribute to the integer's size, but not to its widthprecision. Consequently, inferring the width precision of an integer type from its size may result in a width precision that is larger than it actually is which can then lead to incorrect assumptions about the range of these types.  Make sure you use correct integer widths precisions in your code, and in particular, do not use the sizeof operator to compute the width precision of an integer type on architectures that use padding bits or in strictly conforming (that is, portable) programs.

Noncompliant Code Example

This noncompliant code example illustrates a function that produces 2 raised to the power of the function argument. To prevent undefined behavior, in compliance with INT34-C. Do not shift a negative number of bits or more bits than exist in the operand, the function ensures that the argument is less than the number of bits used to store an unsigned int.

#include <limits.h>
 
unsigned int pow2(unsigned int exp) {
  if (exp >= sizeof(unsigned int) * CHAR_BIT) {
    /* Handle error */
  }
  return 1 << exp;
}

Compliant Solution

This compliant solution uses a popcount() function, which counts the number of bits set on any unsigned integer. This allows this code to count determine the number precision of width bits used by any unsigned int typeany integer type, signed or unsigned.

#include <stddef.h>
#include <stdint.h>
#include <limits.h>
#include <inttypes.h>
 
/* Returns the number of set bits */
size_t popcount(uintmax_t num) {
  size_t widthprecision = 0;
  while (num != 0) {
    if (num % 2 == 1) {
      widthprecision++;
    }
    num >>= 1;
  }
  return widthprecision;
}
#define UWIDTHPRECISION(umax_value) popcount(umax_value) 

#include <stddef.h>
#include <stdint.h>
#include <limits.h>
#include <inttypes.h>

extern size_t popcount(uintmax_t);
#define UWIDTHPRECISION(umax_value) popcount(umax_value)  

unsigned int pow2(unsigned int exp) {
  if (exp >= UWIDTHPRECISION(UINT_MAX)) {
    /* handle error */
  }
  return 1 << exp;
}

#define UWIDTHPRECISION(umax_value) _popcnt(umax_value)

Exceptions

INT19-EX1: Platforms in which integer types have no padding may use the sizeof operator to compute integer widths precision including, for example, 32-bit and 64-bit x86 platforms.

#define UWIDTHPRECISION(type) (sizeof(type) * CHAR_BIT)

Mistaking an integer size for its width precision can permit invalid width precision arguments to operations such as bitwise shifts, resulting in undefined behavior.

Bibliography