The following code,according to me should run successfully,but fails at runtime.I don\'t get the reason:
void main()
{
int arr[5][3]={1,2,3,4,5,6,7,8,9,10,1
int **p=&m;
p points to address, where m is placed:
... | m | sth | ... | p | ...
^ V
|_________________|
Now, increment it:
... | m | sth | ... | p | ...
^ V
|___________|
So, now p points to sth. What is sth? Nobody knows. But you're trying to get access to the address sth contains. This is undefined behavior.
With the line
int **p=&m
you create a pointer to a pointer to an integer.
Then, you add one to the address - one memory address, that is, not one times the number of bytes to point to the next integer.
Then you deference it twice:
Here int **p=&m;. p points to m. Then when p = p + 1; p will point to the address next to m (integer). That address may not be accessible.
An array of arrays and a pointer to a pointer is quite different, and can't be used interchangeably.
For example, if you look at your array arr it looks like this in memory
+-----------+-----------+-----------+-----------+-----+-----------+ | arr[0][0] | arr[0][1] | arr[0][2] | arr[1][0] | ... | arr[4][2] | +-----------+-----------+-----------+-----------+-----+-----------+
When you have the pointer-to-pointer p the program don't really knows that it points to an array of arrays, instead it's treated as an array of pointers, which looks like this in memory:
+------+------+------+-----+ | p[0] | p[1] | p[2] | ... | +------+------+------+-----+ | | | | | v | | something | v | something v something
So when you do p + 1 you get to p[1] which is clearly not the same as arr[1].