StackOverflow2013

Note that there are some explanatory texts on larger screens.

plurals

POCUDA 2D array nvidia
text
Body
copied!<p>I am using <code>cudaMallocPitch</code> and <code>cudaMemcpy2D</code> for 2D array. I am not sure that I have coded correct even though I could not get the output correctly. Can any one help please? Can any one debug my error? Thanks in advance. </p> <pre><code>#include<stdio.h> #include<cuda.h> #define siz 4*sizeof(int) __global__ void addmatrix(int *m1,int *m2,size_t pitch) { int r=threadIdx.x; int *r1=m1+r*pitch; int *r2=m2+r*pitch; int c; for(c=1;c<=4;c++) { r1[c]+=r2[c]; } } int main() { int i,j; int **m1_c,**m2_c; int *m1_d,*m2_d; size_t pitch; cudaError_t err; m1_c=(int **)malloc(4*sizeof(int *)); for(i=1;i<=4;i++) { m1_c[i]=(int *)malloc(siz); } m2_c=(int **)malloc(4*sizeof(int *)); for(i=1;i<=4;i++) { m2_c[i]=(int *)malloc(siz); } for(i=1;i<=4;i++) { for(j=1;j<=4;j++) { m1_c[i][j]=rand()%10; m2_c[i][j]=rand()%10; } } for(i=1;i<=4;i++) { for(j=1;j<=4;j++) { printf("%d\t",m1_c[i][j]); } printf("\n"); } printf("\n\n"); for(i=1;i<=4;i++) { for(j=1;j<=4;j++) { printf("%d\t",m2_c[i][j]); } printf("\n"); } err=cudaMallocPitch((void **)&m1_d,&pitch,siz,siz); err=cudaMallocPitch((void **)&m2_d,&pitch,siz,siz); err=cudaMemcpy2D(m1_d,pitch,m1_c,siz,siz,4,cudaMemcpyHostToDevice); err=cudaMemcpy2D(m2_d,pitch,m2_c,siz,siz,4,cudaMemcpyHostToDevice); dim3 grid(1); dim3 block(16); addmatrix<<<grid,block>>>(m1_d,m2_d,siz); cudaMemcpy2D(m1_c,siz,m1_d,pitch,siz,4,cudaMemcpyDeviceToHost); for(i=1;i<=4;i++) { for(j=1;j<=4;j++) { printf("%d\t",m1_c[i][j]); } printf("\n"); } err=cudaFree(m1_d); err=cudaFree(m2_d); err=cudaDeviceReset(); } </code></pre>

Querying!

Guidance

An individual column

Larger individual text columns get their own page to allow for proper reading.

SQuiL has stopped working due to an internal error.

If you are curious you may find further information in the browser console, which is accessible through the devtools (F12).

Reload