how size of a structure varies with different data types

Question

I am using Linux 32 bit os, and GCC compiler.

I tried with three different type of structure. in the first structure i have defined only one char variable. size of this structure is 1 that is correct.

in the second structure i have defined only one int variable. here size of the structure is showing 4 that is also correct.

but in the third structure when i defined one char and one int that means total size should be 5, but the output it is showing 8. Can anyone please explain how a structure is assigned?

typedef struct struct_size_tag
{
    char c;
    //int i;
}struct_size;

int main()
{
        printf("Size of structure:%d\n",sizeof(struct_size));
        return 0;
}

Output: Size of structure:1

typedef struct struct_size_tag
{
    //char c;
    int i;
}struct_size;

int main()
{
        printf("Size of structure:%d\n",sizeof(struct_size));
        return 0;
}

Output: Size of structure:4

typedef struct struct_size_tag
{
    char c;
    int i;
}struct_size;

int main()
{
        printf("Size of structure:%d\n",sizeof(struct_size));
        return 0;
}

Output:

Size of structure:8

The size varies because of padding. When char c is allocated 1 byte, then int i comes which requires 4 bytes. So 3 bytes which is left after allocating char is left off, increasing the size to 8. — Raghuram
– Raghuram, Commented Dec 9, 2013 at 10:55
Before your structure definition add this line: #pragma pack(1) — Fiddling Bits
– Fiddling Bits, Commented Dec 9, 2013 at 10:57

Tom Zych · Accepted Answer · 2014-06-30 02:09:53Z

3

The difference in size is due to alignment. The compiler is free to choose padding bytes, which make the total size of a structure not necessarily the sum of its individual elements.

If the padding of a structure is undesired, because it might have to interface with some hardware requirement (or other reasons), compilers usually support packing structures, so the padding is disabled.

edited Jun 30, 2014 at 2:09

Tom Zych

13.7k9 gold badges38 silver badges55 bronze badges

answered Dec 9, 2013 at 10:54

Devolus

22.2k15 gold badges70 silver badges122 bronze badges

Sign up to request clarification or add additional context in comments.

Comments

amdixon · Accepted Answer · 2013-12-09 11:01:27Z

2

You definitely get Data Structure Alignment

"Data alignment means putting the data at a memory offset equal to some multiple of the word size, which increases the system's performance due to the way the CPU handles memory. To align the data, it may be necessary to insert some meaningless bytes between the end of the last data structure and the start of the next, which is data structure padding."

For more, take a look at this, Data Structure Alignment

edited Dec 9, 2013 at 11:01

amdixon

3,8438 gold badges27 silver badges35 bronze badges

answered Dec 9, 2013 at 10:54

Daniel

2,1831 gold badge14 silver badges14 bronze badges

Comments

Andrew W. Phillips · Accepted Answer · 2013-12-10 03:39:08Z

2

The C standard allows a compiler to add padding bytes to structs after any field to allow the following field to be aligned according to any specific requirements of the compiler (or the user of the compiler). The standard does not specify, but typically a compiler will provide a command line argument to specify the (default) alignment. Good compilers also invariably support the de facto standard of #pragma pack, including push and pop options.

Padding bytes provide improved performance by reducing the amount of memory accesses required by suitably aligned data types. For example, on a 32-bit processor (more specifically a system which uses memory with 32 data lines) accessing a 32-bit integer will require two memory accesses when reading and writing the value rather than just one if it crosses a 4-byte boundary (ie, unless the bottom two bits of the address of the integer are zero).

See My Blog Post for more details (better than Wikipedia article).

answered Dec 10, 2013 at 3:39

Andrew W. Phillips

3,6001 gold badge24 silver badges27 bronze badges

2 Comments

rajenpandit Over a year ago

Thanks Andrew, I have gone through your blog ... its nice.

Tony Delroy Over a year ago

"...will require two memory accesses..." - even more crucially, some CPUs don't support operations on misaligned data (e.g. UltraSparc generates a SIGBUS), so the compiler would need to generate additional instructions with bit shifting, anding and oring to orchestrate such an operation.

Nicola Coretti · Accepted Answer · 2013-12-09 10:54:35Z

1

The magic word is padding/memory alignment @see data structure alignment.

answered Dec 9, 2013 at 10:54

Nicola Coretti

2,6853 gold badges21 silver badges23 bronze badges

Collectives™ on Stack Overflow

how size of a structure varies with different data types

Output: Size of structure:1

Output: Size of structure:4

Size of structure:8

4 Answers 4

Comments

Comments

2 Comments

Comments

Your Answer

Linked

Hot Network Questions

Collectives™ on Stack Overflow

Output: Size of structure:1

Output: Size of structure:4

Size of structure:8

4 Answers 4

Comments

Comments

2 Comments

Comments

Your Answer

Sign up or log in

Post as a guest

Linked

Related