How to use

JCublas

in

jcuda.jcublas

public static int cublasSetVectorAsync (int n, int elemSize, Pointer hostPtr, int incx, Pointer devicePtr, int incy, cudaStream_t stream)
{
  return checkResult(cublasSetVectorAsyncNative(n, elemSize, hostPtr, incx, devicePtr, incy, stream));
}
/*

/**
 * Wrapper for CUBLAS function.<br />
 * <br />
 * cublasStatus
 * cublasAlloc (int n, int elemSize, void **devicePtr)<br />
 *<br />
 * creates an object in GPU memory space capable of holding an array of
 * n elements, where each element requires elemSize bytes of storage. If
 * the function call is successful, a pointer to the object in GPU memory
 * space is placed in devicePtr. Note that this is a device pointer that
 * cannot be dereferenced in host code.<br />
 *<br />
 * Return Values<br />
 * -------------<br />
 * CUBLAS_STATUS_NOT_INITIALIZED  if CUBLAS library has not been initialized<br />
 * CUBLAS_STATUS_INVALID_VALUE    if n <= 0, or elemSize <= 0<br />
 * CUBLAS_STATUS_ALLOC_FAILED     if the object could not be allocated due to
 *                                lack of resources.<br />
 * CUBLAS_STATUS_SUCCESS          if storage was successfully allocated<br />
 */
public static int cublasAlloc(int n, int elemSize, Pointer ptr)
{
  return checkResult(cublasAllocNative(n, elemSize, ptr));
}
private static native int cublasAllocNative(int n, int elemSize, Pointer ptr);

public static int cublasSetKernelStream (cudaStream_t stream)
{
  return checkResult(cublasSetKernelStreamNative(stream));
}
private static native int cublasSetKernelStreamNative(cudaStream_t stream);

public static int cublasGetMatrixAsync (int rows, int cols, int elemSize, Pointer A, int lda,  Pointer B, int ldb, cudaStream_t stream)
{
  return checkResult(cublasGetMatrixAsyncNative(rows, cols, elemSize, A, lda, B, ldb, stream));
}

cublasCcopyNative(n, x, incx, y, incy);
checkResultBLAS();

cuComplex result = cublasCdotuNative(n, x, incx, y, incy);
checkResultBLAS();
return result;

cuComplex result = cublasCdotcNative(n, x, incx, y, incy);
checkResultBLAS();
return result;

cublasCaxpyNative(n, alpha, x, incx, y, incy);
checkResultBLAS();

cublasCgemvNative(trans, m, n, alpha, A, lda, x, incx, beta, y, incy);
checkResultBLAS();

cublasCgemmNative(transa, transb, m, n, k, alpha, A, lda, B, ldb, beta, C, ldc);
checkResultBLAS();

cublasCgbmvNative(trans, m, n, kl, ku, alpha, A, lda, x, incx, beta, y, incy);
checkResultBLAS();

public static int cublasSetMatrixAsync (int rows, int cols, int elemSize, Pointer A, int lda, Pointer B, int ldb, cudaStream_t stream)
{
  return checkResult(cublasSetMatrixAsyncNative(rows, cols, elemSize, A, lda, B, ldb, stream));
}

public static int cublasGetVectorAsync(int n, int elemSize, Pointer devicePtr, int incx, Pointer hostPtr, int incy, cudaStream_t stream)
{
  return checkResult(cublasGetVectorAsyncNative(n, elemSize, devicePtr, incx, hostPtr, incy, stream));
}

cublasZscalNative(n, alpha, x, incx);
checkResultBLAS();

float result = cublasScnrm2Native(n, x, incx);
checkResultBLAS();
return result;

int result = cublasIcamaxNative(n, x, incx);
checkResultBLAS();
return result;

cublasCsscalNative(n, alpha, x, incx);
checkResultBLAS();

cublasCscalNative(n, alpha, x, incx);
checkResultBLAS();

int result = cublasIsaminNative(n, x, incx);
checkResultBLAS();
return result;

int result = cublasIsamaxNative(n, x, incx);
checkResultBLAS();
return result;