Logo Questions Linux Laravel Mysql Ubuntu Git Menu
 

Please explain the difference

i have a program about 2-D arrays

base adress is 8678

#include<stdio.h>
#include<conio.h>
main()
{
 int arr[3][3]={
                 {83,8,43},
                 {73,45,6},
                 {34,67,9}
                 };
printf("%d ",&arr+1);  //points to 8696
printf("%d ",arr+1);   //points to 8684
return 0;
}  

what is the difference between arr+1 and &arr+1?

like image 449
Amol Singh Avatar asked Aug 08 '12 04:08

Amol Singh


1 Answers

Well, they're different things. arr decays in most contexts to a pointer to the first element of your array - that means a pointer to the first 3-element row in your 2D array: type int (*)[3]. arr + 1, then, points to the second row in the array.

&arr is the address of the array itself (type int (*)[3][3]), so &arr + 1 points to memory just past the end of the entirety of your 2D array.

You can confirm this behaviour easily by printing differently. Specifically, printing the offsets to the new pointers rather than the values themselves will help clear things up. The output from your program from these print statements:

printf("%ld\n",(intptr_t)(&arr+1) - (intptr_t)arr);
printf("%ld\n",(intptr_t)(arr+1) - (intptr_t)arr);

Will be the decimal offsets to &arr+1 and arr+1 respectively. Here's the output from a test run I just made:

36
12

36 matches up: 3 rows × 3 columns × 4 bytes per entry = 36 bytes. So does the 12: 1 row × 3 columns × 4 bytes per entry = 12 bytes.

Note - you're also printing pointers using %d, which is wrong. You should probably be using %p for that.

like image 185
Carl Norum Avatar answered Oct 02 '22 17:10

Carl Norum