c++ - Can you use cublasDdot() to use blas operations in non-GPU memory? -


so have code performs matrix multiplicaiton, problem returns zeroes when use library -lcublas , compiler nvcc; however, code runs great few tweaks function names when use compiler, g++ library -lblas.

can use -lcublas library perform matrix multiplication memory not on gpu?

here's code returns 0's:

extern "c" //external reference function code compiles {     double cublasddot(int *n, double *a, int *inca, double *b, int *incb); }  //stuff happens      cout << "calculating/printing contents of matrix c ddot...\n";             c[i][t]=cublasddot(&n, parta, &inca, partb, &incb); //this thing isn't working reason (although compiles fine) 

i compile using command: nvcc program -lcublas

this work however:

extern "c" //external reference function code compiles {     double ddot_(int *n, double *a, int *inca, double *b, int *incb); }  //stuff happens  c[i][t]=ddot_(&n, parta, &inca, partb, &incb); 

compiled g++ program -lblas

cublas requires functioning cuda gpu.

probably doing no error checking. read on how error checking in cublas manual. , @ error checking sample code.

the ordinary usage of cublas requires data transferred gpu , results transferred back.


Comments

Popular posts from this blog

SPSS keyboard combination alters encoding -

Add new record to the table by click on the button in Microsoft Access -

javascript - jQuery .height() return 0 when visible but non-0 when hidden -