Error in a simple for loop of a SIMD code

Hi all,

the code is like this:

void testfun()
{
	float *buf1 = new float[dataLen];
	float *buf2 = new float[dataLen];

	for(int i=0; i<dataLen; i++)
	{
		buf1[i] = 1;
		buf2[i] = 1;
	}


	__m256 *pp1;
	__m256 *pp2;
	__m256 _rresult = _mm256_set1_ps(0.0f);


	for(int j=0;j<1000000; j++)
	{	
		pp1 = (__m256 *)buf1;
		pp2 = (__m256 *)buf2;

		_rresult = _mm256_sub_ps(_mm256_set_ps(j,0,0,0,0,0,0,0), _rresult);
		
		for(int i=0; i<dataLen/8; i++)
		{			
			_rresult = _mm256_add_ps(_mm256_mul_ps(*pp1, *pp2), _rresult);// There will be an error
			pp1++;
			pp2++;
			cout<<"i"<<i<<endl;
		}
	}

}

In debug mode there will no error, but when in release mode it will be an error every time!

Wish someone can help me , Thanks!