Array elements get overridden?

Hello,

first the code that works:

Code:

int main(int argc, char *argv[])

{

	int r;

	int rc;

	int c;

	float *m1;

	float *m2;

	float *m3;

	int i;

	float scan;



	printf("Please enter rows of matrix 1: ");

	scanf("%i", &r);

	printf("Please enter columns/rows of matrix 1/2: ");

	scanf("%i", &rc);

	printf("Please enter columns of matrix 2: ");

	scanf("%i", &c);



	m1 = malloc(r*rc);

	m2 = malloc(rc*c);

	m3 = malloc(r*c);



	printf("Enter elements of matrix 1:\n");



	for(i = 0; i < r*rc; i++)

	{

		scanf("%f", &scan);

		m1[i] = scan;

	}



	printf("Enter elements of matrix 2:\n");



	for(i = 0; i < rc*c; i++)

	{

		scanf("%f", &scan);

		m2[i] = scan;

	}




	for(i = 0; i < rc*c; i++)

	{

		printf("%f\n", m2[i]);

	}



	printf("Thank you, come again\n");

	

	return 0;

}

When I enter 9 1s (3*3 matrix) in the 2nd matrix here, I get 9 1s printed out.

Now the broken code:

Code:

int main(int argc, char *argv[])

{

	int r;

	int rc;

	int c;

	float *m1;

	float *m2;

	float *m3;

	int i;

	float scan;



	printf("Please enter rows of matrix 1: ");

	scanf("%i", &r);

	printf("Please enter columns/rows of matrix 1/2: ");

	scanf("%i", &rc);

	printf("Please enter columns of matrix 2: ");

	scanf("%i", &c);



	m1 = malloc(r*rc);

	m2 = malloc(rc*c);

	m3 = malloc(r*c);



	printf("Enter elements of matrix 1:\n");



	for(i = 0; i < r*rc; i++)

	{

		scanf("%f", &scan);

		m1[i] = scan;

	}



	printf("Enter elements of matrix 2:\n");



	for(i = 0; i < rc*c; i++)

	{

		scanf("%f", &scan);

		m2[i] = scan;

	}

	for(i = 0; i < r*c; i++)

	{

		m3[i] = 0.f;

	}


	for(i = 0; i < rc*c; i++)

	{

		printf("%f\n", m2[i]);

	}



	printf("Thank you, come again\n");

	

	return 0;

}

Here I get 1.0 1.0 1.0 1.0 0.0 0.0 ...

As an output.

What's wrong? Get the elements of my 2nd matrix overriden? If so, why? Or is it something else?

Double-spaced code is unnecessarily hard to read.

I think it's because you're not malloc'ing enough space for a 2-d array of ints. You need to do malloc(r*c*sizeof(int)) or calloc(r*c, sizeof(int)) for all your malloc calls, since an int is more than 1 byte (4 on a 32-bit machine). malloc takes the number of bytes, not elements. It doesn't know what type of data you're mallocing, and an int on your system is likely 4 bytes (32-bit machine). You may also be having the same issue with m1, but since you don't print it, you don't notice it.

I figured out that in the 2nd example some of my m3 elements override some m2 elements.

Why?

/edit: nvm

You need to multiply the amount of bytes to malloc by sizeof(float) in each calculation.

D'oh! Nice catch nonoob.

As a matter of fact, this could all be avoided if we did this the smart way. Multiply by sizeof(*m1), etc. That way, if you change m1 from float to double, you don't need to change your code.

@ThatDudeMan: Try this on for size:

Code:

m1 = malloc(r*rc * sizeof(*m1));
m2 = malloc(rc*c * sizeof(*m2));
m3 = malloc(r*c * sizeof(*m3));