Struct Padding

Question

Problem: Struct Padding. Solve this hands-on electronics task on EWskills.

Accepted Answer

Padding is extra unused memory automatically inserted by the compiler between structure members to ensure data alignment as per the CPU’s architecture requirements.

Why It Matters:

In embedded systems (e.g., 32-bit ARM Cortex-M controllers), as it has 32bit address and data buse, memory is usually word-aligned, which means it can fetch 32 bits (4-bytes) as a time.

It helps to fetch Intiger variable in one-signle cycle, increasing read/ write & execution speed.

— meaning a 4-byte int should be stored at a memory address divisible by 4.

Real Impact in Firmware

Let’s say you define this structure:

struct SensorData {
    char status;   // 1 byte
    int reading;   // 4 bytes
};

On a 32-bit microcontroller:

char status is at offset 0
Compiler inserts 3 bytes of padding
int reading starts at offset 4 (aligned)

Total size = 8 bytes

If padding is not added (e.g., using packed structs), the int may fall at an unaligned address (e.g., 1, 2, or 3), which causes:

Unaligned access
Extra memory fetch cycles for int variable.

Above code manual inspection of memory layout, which is a common low-level debugging or optimization practice.

Solution Logic:

Cast addresses to char* for byte-level pointer arithmetic
Subtract field address from base address to get offset
Print total size with sizeof()

Alternate Solution (using offsetof)

#include 
#include 

typedef struct {
    char a;
    int b;
    short c;
} MyStruct;

void print_offsets() {
    printf("Offset of a: %lu
", offsetof(MyStruct, a));
    printf("Offset of b: %lu
", offsetof(MyStruct, b));
    printf("Offset of c: %lu
", offsetof(MyStruct, c));
    printf("Size: %lu", sizeof(MyStruct));
}

int main() {
    print_offsets();
    return 0;
}

78. Struct Padding