There is no output when I declare a matrix inside the kernel

Accelerated Computing CUDA CUDA Programming and Performance
1

The error due to internal declaration.,for eg, the kernel is as follows.

extern „C“
global void ComputationdClustersInternelOnGPU(int numTokenSrc,int numWordSrc,int srcLength, char *src,int *srctokensSFIndices,int *srctokensLength,int *srcIndices, int *srcStartIndices,int totalLengthDistinct, char *patternRemoved,int numTokenPattern,int numWordPattern,int patternLength,char *pattern,int *patterntokensSFIndices,int *patterntokensLength,int *patternIndices,int patternStartIndices,float WordsFinal)
{
int ix = blockIdx.x * blockDim.x + threadIdx.x;
int dX = (int) malloc(srcLength * totalLengthDistinct * sizeof(int));
if(ix<totalLengthDistinct)
{
for (int i = 0; i < srcLength; i++) {
if (src[i] == ‚,‘)
dX[ix * srcLength + i] = 0;
else
{
if (src[i] == patternRemoved[ix])
dX[ix * srcLength + i] = srcIndices[i];
else if (src[i] != patternRemoved[ix])
dX[ix * srcLength + i] = dX[ix * srcLength + i-1];
}
}

    } 
    __syncthreads(); 

    for(int i =0;i<srcLength*totalLengthDistinct;i++){

      printf("Elements of an array");
      printf("%d\n",dX[i]);

    }

}



When I run the kernel, it gives all zeros as an output for the matrix dX. While, when I declare dX inside the host and send it to the kernel, it gives the right output for the matrix. What is wrong in the declaration in the device, it is only an int matrix of size srcLength * totalLengthDistinct .
the output : 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

Can anyone help me,please?
2

I doubt this is the code you are actually running. This code won’t compile.

but in any event, I suggest reading this section on of the programming guide, carefully.