public EllpackDenseVectorBuilder(CUDA cu,CUdeviceptr vector, CUdeviceptr vals,CUdeviceptr cols,CUdeviceptr length,int rows,int dim)
        {
            cuda = cu;
            vecPtr = vector;
            valsPtr = vals;
            idxPtr = cols;
            vecLengthPtr = length;
            nrRows = (uint)rows;
            vecDim = (uint)dim;

            blocksPerGrid = (int) Math.Ceiling( (vecDim + 0.0) / threadsPerBlock);

            var blocksPerGrid1 = (vecDim + threadsPerBlock - 1) / threadsPerBlock;

            Debug.Assert(blocksPerGrid == blocksPerGrid1);

        }

 public static extern void cublasSsyrk(char uplo, char trans, int n, int k, float alpha, CUdeviceptr A, int lda, float beta, CUdeviceptr C, int ldc);

 public static extern void cublasDger(int m, int n, double alpha, CUdeviceptr x, int incx, CUdeviceptr y, int incy, CUdeviceptr A, int lda);

 public static extern void cublasDgemm(char transa, char transb, int m, int n, int k, double alpha, CUdeviceptr A, int lda, CUdeviceptr B, int ldb, double beta, CUdeviceptr C, int ldc);

 public static extern void cublasDcopy(int n, CUdeviceptr x, int incx, CUdeviceptr y, int incy);

 public static extern void cublasCtrsv(char uplo, char trans, char diag, int n, CUdeviceptr A, int lda, CUdeviceptr x, int incx);

 public static extern void cublasZtrmm(char side, char uplo, char transa, char diag, int m, int n, cuDoubleComplex alpha, CUdeviceptr A, int lda, CUdeviceptr B, int ldb);

 public static extern void cublasZsyrk(char uplo, char trans, int n, int k, cuDoubleComplex alpha, CUdeviceptr A, int lda, cuDoubleComplex beta, CUdeviceptr C, int ldc);

 public static extern void cublasZgemv(char trans, int m, int n, cuDoubleComplex alpha, CUdeviceptr A, int lda, CUdeviceptr x, int incx, cuDoubleComplex beta, CUdeviceptr y, int incy);

 public static extern void cublasZgemm(char transa, char transb, int m, int n, int k, cuDoubleComplex alpha, CUdeviceptr A, int lda, CUdeviceptr B, int ldb, cuDoubleComplex beta, CUdeviceptr C, int ldc);

 public static extern void cublasZdscal(int n, double alpha, CUdeviceptr x, int incx);

 public static extern void cublasZdrot(int n, CUdeviceptr x, int incx, CUdeviceptr y, int incy, double c, double s);

 public static extern cuDoubleComplex cublasZdotu(int n, CUdeviceptr x, int incx, CUdeviceptr y, int incy);

 public static extern void cublasZaxpy(int n, cuDoubleComplex alpha, CUdeviceptr x, int incx, CUdeviceptr y, int incy);

 public static extern void cublasStrsm(char side, char uplo, char transa, char diag, int m, int n, float alpha, CUdeviceptr A, int lda, CUdeviceptr B, int ldb);

 public static extern void cublasZswap(int n, CUdeviceptr x, int incx, CUdeviceptr y, int incy);

 public static extern void cublasZsymm(char side, char uplo, int m, int n, cuDoubleComplex alpha, CUdeviceptr A, int lda, CUdeviceptr B, int ldb, cuDoubleComplex beta, CUdeviceptr C, int ldc);

 public static extern void cublasZgeru(int m, int n, cuDoubleComplex alpha, CUdeviceptr x, int incx, CUdeviceptr y, int incy, CUdeviceptr A, int lda);

 public static extern void cublasZtpsv(char uplo, char trans, char diag, int n, CUdeviceptr AP, CUdeviceptr x, int incx);

 public static extern void cublasZherk(char uplo, char trans, int n, int k, double alpha, CUdeviceptr A, int lda, double beta, CUdeviceptr C, int ldc);