Function onednn_sys::dnnl_pooling_backward_desc_init

pub unsafe extern "C" fn dnnl_pooling_backward_desc_init(
    pool_desc: *mut dnnl_pooling_desc_t, 
    alg_kind: dnnl_alg_kind_t, 
    diff_src_desc: *const dnnl_memory_desc_t, 
    diff_dst_desc: *const dnnl_memory_desc_t, 
    strides: *mut dnnl_dim_t, 
    kernel: *mut dnnl_dim_t, 
    padding_l: *mut dnnl_dim_t, 
    padding_r: *mut dnnl_dim_t
) -> dnnl_status_t

Initializes a descriptor for pooling backward propagation primitive.

Arrays @p strides, @p kernel, @p padding_l, and @p padding_r contain values for spatial dimensions only and hence must have the same number of elements as there are spatial dimensions. The order of values is the same as in the tensor: depth (for 3D tensors), height (for 3D and 2D tensors), and width.

@param pool_desc Output descriptor for a pooling primitive. @param alg_kind Pooling algorithm kind: either #dnnl_pooling_max, #dnnl_pooling_avg_include_padding, or #dnnl_pooling_avg (same as #dnnl_pooling_avg_exclude_padding). @param diff_src_desc Diff source memory descriptor. @param diff_dst_desc Diff destination memory descriptor. @param strides Array of strides for spatial dimension. @param kernel Array of kernel spatial dimensions. @param padding_l Array of padding values for low indices for each spatial dimension ([[front,] top,] left). @param padding_r Array of padding values for high indices for each spatial dimension ([[back,] bottom,] right). Can be NULL in which case padding is considered to be symmetrical. @returns #dnnl_success on success and a status describing the error otherwise.