Is GT 420M supported by Nsight?

Accelerated Computing CUDA CUDA Programming and Performance
1

Hi, I am a beginner in CUDA programming and want to practice it on my notebook.

My video card is GT 420M, with i5 CPU.

I installed Nsight successfully into VS2010 and followed the instruction below to create a test project.

How to create a new CUDA project from scratch in VS2010 that works with Nsight 1.5 and CUDA 3.2 toolkit

My test code is shown as below

#include <iostream>

#include <cuda_runtime.h>

#include <device_launch_parameters.h>

using namespace std;

__global__ void VectorAddKernel(float* a,float *b,float *c){

	int nIndex=blockIdx.x*blockDim.x+threadIdx.x;

	c[nIndex]=a[nIndex]+b[nIndex];

	return ;

}

int main (){

	float a[6],b[6],c[6];

	for (int i=0;i<6;i++){

		a[i]=b[i]=c[i]=i;

	}

	float *ad,*bd,*cd;

	int nSize=6;

	cudaMalloc((void**)&ad,nSize);

	cudaMemcpy(ad,a,nSize,cudaMemcpyHostToDevice);

	cudaMalloc((void**)&bd,6);

	cudaMemcpy(bd,b,nSize,cudaMemcpyHostToDevice);

	cudaMalloc((void**)&cd,6);

	//////////////////	

	dim3 DimGrid(3,1,1);

	dim3 DimBlock(2,1,1);

	VectorAddKernel <<< DimGrid,DimBlock >>>(ad,bd,cd);

	cudaMemcpy(c,cd,nSize,cudaMemcpyDeviceToHost);

	////////////////

	for (int i=0;i<6;i++){

		cout<<c[i]<<" ";

	}

	cout<<endl;

	cin>>nSize;

	///////////////////

	cudaFree(ad);cudaFree(bd);cudaFree(cd);

	return 1;

}

However, from debuging, it seems that cudaMalloc is not performed successfully.

My notebook has both an individual Video card and a HD integrated card, and is using the optimus technology to switch.

I am wondering whether the code is not running with Nvidia card or that GT420M is not supported by Nsight.

Thanks for your concerning.