sizeof(structure)

[juice]

sizeof(int) 4 + sizeof(float) 4 + sizeof(char) 1 = 9 bytes, then why does this code report the size of structure p as 12 bytes????

#include<stdio.h>

int main(void)
{
	struct
	{
		int a;
		float b;
		char c;
	}p;
	
	printf("Size of float = %d, Size of int = %d, size of char = %d\n",sizeof(float),sizeof(int),sizeof(char));
    printf("Size of structure = %d\n",sizeof(p));     
}

[laserlight]

sizeof(int) 4 + sizeof(float) 4 + sizeof(char) 1 = 9 bytes, then why does this code report the size of structure p as 12 bytes????

Read: Why does sizeof report a larger size than I expect for a structure type, as if there were padding at the end?

[ushacy]

It is becuase of structure padding.. just reffer to structure padding concept in net... you will get the solution

[memcpy]

If you want to remove the padding, try the __packed__ flag:

	struct
	{
		int a;
		float b;
		char c;
	} __attribute__((__packed__)) p;

You can do this over the entire struct, or just on specific variables if the need arises.

[mnd22]

Read: Why does sizeof report a larger size than I expect for a structure type, as if there were padding at the end?

Wow this is a great site laserlight! Thank you for the link!

[iMalc]

If you want to remove the padding, try ...

That's unresponsible to straight out show how to work around the padding without first saying why it is there and why it is probably better that it stays there.
Not only that, but the way you've shown is highly compiler dependent. Visual Studio for one, does it a different way.

ushacy, the padding is there for performance reasons, or in the case of certain CPUs, it's required for alignment when an array of that type is used. Taking away the padding will most likely at least result in slower code and possibly a crash.