24efdf4f57841d5d73fda3745b96a1b102ca484f,skcuda/linalg.py,,diag,#Any#,1118

Before Change

    elif len(v_gpu.shape) >= 3:
        raise ValueError("input array cannot have greater than 2-dimensions")

    use_double = int(v_gpu.dtype in [np.float64, np.complex128])
    use_complex = int(v_gpu.dtype in [np.complex64, np.complex128])

    // Initialize output matrix:
    d_gpu = misc.zeros((v_gpu.size, v_gpu.size), v_gpu.dtype, allocator=alloc)

    // Get block/grid sizes:
    dev = misc.get_current_device()
    block_dim, grid_dim = misc.select_block_grid_sizes(dev, d_gpu.shape)

    diag = _get_diag_kernel(use_double, use_complex)
    diag(v_gpu, d_gpu, np.uint32(v_gpu.size),
         block=block_dim,
         grid=grid_dim)

After Change

    d_gpu = misc.zeros((N, N), v_gpu.dtype, allocator=alloc)

    func = _get_diag_kernel(v_gpu.dtype)
    func(d_gpu, v_gpu, N, slice=slice(0, N))
    return d_gpu

@context_dependent_memoize

Instances