Bit Fields in C Structures

A bit field is a member of a structure that occupies a specified number of bits rather than full bytes. Bit fields are used to save memory when storing small integer values.

---

Defining Bit Fields

struct Flags {
    unsigned int isActive  : 1;  /* 1 bit — 0 or 1 */
    unsigned int isAdmin   : 1;  /* 1 bit */
    unsigned int priority  : 3;  /* 3 bits — 0 to 7 */
    unsigned int reserved  : 3;  /* 3 bits padding */
};

struct Flags f;
f.isActive = 1;
f.isAdmin  = 0;
f.priority = 5;

printf("Active: %d, Priority: %d\n", f.isActive, f.priority);

---

Bit Field Rules

• Type must be int, unsigned int, or signed int (C99 also allows _Bool)
• Width must be between 0 and the width of the type (0 = padding to next boundary)
• Cannot take address of a bit field (& not allowed)
• Cannot create arrays of bit fields

---

Memory Savings Example

struct WithoutBitField {
    int isActive;  /* 4 bytes */
    int isAdmin;   /* 4 bytes */
    int priority;  /* 4 bytes */
};  /* total: 12 bytes */

struct WithBitField {
    unsigned int isActive : 1;
    unsigned int isAdmin  : 1;
    unsigned int priority : 3;
};  /* total: 4 bytes (fits in one int) */

---

Anonymous Bit Fields (Padding)

struct Register {
    unsigned int bit0   : 1;
    unsigned int        : 3;   /* skip 3 bits */
    unsigned int bit4   : 1;
};

---

Practical Use: Hardware Register Mapping

Bit fields are widely used in embedded systems to map hardware registers:

struct StatusRegister {
    unsigned int carry    : 1;
    unsigned int zero     : 1;
    unsigned int negative : 1;
    unsigned int overflow : 1;
    unsigned int          : 4;   /* reserved */
};

volatile struct StatusRegister *psr = (struct StatusRegister *)0x4000;
if (psr->zero) {
    /* result was zero */
}

Note: The exact layout of bit fields is implementation-defined and may differ between compilers. Avoid relying on bit field layout for portable code.