[−][src]Crate onednn_sys
Structs
| __fsid_t | |
| dnnl_batch_normalization_desc_t | A descriptor of a Batch Normalization operation. |
| dnnl_binary_desc_t | A descriptor of a binary operation. |
| dnnl_blocking_desc_t | Generic description of blocked data layout for most memory formats. |
| dnnl_convolution_desc_t | A descriptor of a convolution operation. |
| dnnl_eltwise_desc_t | A descriptor of a element-wise operation. |
| dnnl_engine | @struct dnnl_engine @brief An opaque structure to describe an engine. |
| dnnl_exec_arg_t | A structure that contains an index and a memory object, and is used to pass arguments to dnnl_primitive_execute(). |
| dnnl_inner_product_desc_t | A descriptor of an inner product operation. |
| dnnl_layer_normalization_desc_t | A descriptor of a Layer Normalization operation. |
| dnnl_lrn_desc_t | A descriptor of a Local Response Normalization (LRN) operation. |
| dnnl_matmul_desc_t | A descriptor of a matrix multiplication operation. |
| dnnl_memory | @struct dnnl_memory An opaque structure to describe a memory. |
| dnnl_memory_desc_t | Memory descriptor. The description is based on a number of dimensions, dimensions themselves, plus information about elements type and memory format. Additionally, contains format-specific descriptions of the data layout. |
| dnnl_memory_extra_desc_t | Description of extra information stored in memory |
| dnnl_pooling_desc_t | A descriptor of a pooling operation. |
| dnnl_post_ops | @struct dnnl_post_ops @brief An opaque structure for a chain of post operations. |
| dnnl_primitive | @struct dnnl_primitive An opaque structure to describe a primitive. |
| dnnl_primitive_attr | @struct dnnl_primitive_attr @brief An opaque structure for primitive descriptor attributes. |
| dnnl_primitive_desc | @struct dnnl_primitive_desc @brief An opaque structure to describe a primitive descriptor. |
| dnnl_primitive_desc_iterator | @struct dnnl_primitive_desc_iterator @brief An opaque structure to describe a primitive descriptor iterator. |
| dnnl_resampling_desc_t | A descriptor of resampling operation. |
| dnnl_rnn_desc_t | A descriptor for an RNN operation. |
| dnnl_rnn_packed_desc_t | Description of tensor of packed weights for rnn. |
| dnnl_shuffle_desc_t | A descriptor of a shuffle operation. |
| dnnl_softmax_desc_t | A descriptor of a Softmax operation. |
| dnnl_stream | @struct dnnl_stream An opaque structure to describe an execution stream. |
| dnnl_stream_attr | An opaque structure to describe execution stream attrbutes. |
| dnnl_version_t | Structure containing version information as per Semantic Versioning |
| dnnl_wino_desc_t | Description of tensor of weights for winograd 2x3 convolution. |
| max_align_t |
Enums
| dnnl_alg_kind_t | Kinds of algorithms. |
| dnnl_cpu_isa_t | CPU instruction set flags |
| dnnl_data_type_t | Data type specification |
| dnnl_engine_kind_t | @brief Kinds of engines. |
| dnnl_format_kind_t | Memory format kind |
| dnnl_format_tag_t | Memory format tag specification. |
| dnnl_memory_extra_flags_t | Flags for memory special features |
| dnnl_normalization_flags_t | Flags for normalization primitives. |
| dnnl_primitive_kind_t | Kinds of primitives. Used to implement a way to extend the library with new primitives without changing the ABI. |
| dnnl_prop_kind_t | Kinds of propagation. |
| dnnl_query_t | Primitive descriptor query specification |
| dnnl_rnn_direction_t | A direction of RNN primitive execution. |
| dnnl_rnn_flags_t | Flags for RNN cell. |
| dnnl_rnn_packed_memory_format_t | |
| dnnl_scratchpad_mode_t | Scratchpad mode |
| dnnl_status_t | Status values returned by the library functions. |
| dnnl_stream_flags_t | @brief Stream flags. |
| dnnl_wino_memory_format_t | Winograd-specific formats |
Constants
Statics
| DNNL_RUNTIME_F32_VAL_REP |
Functions
| dnnl_batch_normalization_backward_desc_init⚠ | Initializes a descriptor for a batch normalization backward propagation primitive. |
| dnnl_batch_normalization_forward_desc_init⚠ | Initializes a descriptor for a batch normalization forward propagation primitive. |
| dnnl_binary_desc_init⚠ | Initializes a descriptor for a binary primitive. |
| dnnl_concat_primitive_desc_create⚠ | Creates a primitive descriptor for an out-of-place concatenation primitive. |
| dnnl_convolution_backward_data_desc_init⚠ | Initializes a descriptor for a convolution backward propagation primitive. |
| dnnl_convolution_backward_weights_desc_init⚠ | Initializes a descriptor for a convolution weights gradient primitive. |
| dnnl_convolution_forward_desc_init⚠ | Initializes a descriptor for a convolution forward propagation primitive. |
| dnnl_deconvolution_backward_data_desc_init⚠ | Initializes a descriptor for a deconvolution backward propagation primitive. |
| dnnl_deconvolution_backward_weights_desc_init⚠ | Initializes a descriptor for a deconvolution weights gradient primitive. |
| dnnl_deconvolution_forward_desc_init⚠ | Initializes a descriptor for a deconvolution forward propagation primitive. |
| dnnl_dilated_convolution_backward_data_desc_init⚠ | Initializes a descriptor for a dilated convolution backward propagation primitive. |
| dnnl_dilated_convolution_backward_weights_desc_init⚠ | Initializes a descriptor for a dilated convolution weights gradient primitive. |
| dnnl_dilated_convolution_forward_desc_init⚠ | Initializes a descriptor for a dilated convolution forward propagation primitive. |
| dnnl_dilated_deconvolution_backward_data_desc_init⚠ | Initializes a descriptor for a dilated deconvolution backward propagation primitive. |
| dnnl_dilated_deconvolution_backward_weights_desc_init⚠ | Initializes a descriptor for a dilated deconvolution weights gradient primitive. |
| dnnl_dilated_deconvolution_forward_desc_init⚠ | Initializes a descriptor for a dilated deconvolution forward propagation primitive. |
| dnnl_eltwise_backward_desc_init⚠ | Initializes a descriptor for eltwise backward propagation primitive. |
| dnnl_eltwise_forward_desc_init⚠ | Initializes a descriptor for eltwise forward propagation primitive. |
| dnnl_engine_create⚠ | Creates an engine. |
| dnnl_engine_destroy⚠ | Destroys an engine. |
| dnnl_engine_get_count⚠ | Returns the number of engines of a particular kind. |
| dnnl_engine_get_kind⚠ | Returns the kind of an engine. |
| dnnl_gemm_s8s8s32⚠ | Performs integer matrix-matrix multiply on 8-bit signed matrix A, 8-bit signed matrix B, and 32-bit signed resulting matrix C. |
| dnnl_gemm_u8s8s32⚠ | Performs integer matrix-matrix multiply on 8-bit unsigned matrix A, 8-bit signed matrix B, and 32-bit signed resulting matrix C. |
| dnnl_get_effective_cpu_isa⚠ | Gets the maximal ISA the library can dispatch to on the CPU. See #dnnl_cpu_isa_t and #dnnl::cpu_isa for the list of the values returned by the C and C++ API functions respectively. |
| dnnl_get_primitive_cache_capacity⚠ | Returns the number of primitives that can be held in the primitive cache at the same time. |
| dnnl_gru_backward_desc_init⚠ | Initializes a descriptor for GRU backward propagation primitive. |
| dnnl_gru_forward_desc_init⚠ | Initializes a descriptor for GRU forward propagation primitive. |
| dnnl_inner_product_backward_data_desc_init⚠ | Initializes descriptor for inner product backward propagation. |
| dnnl_inner_product_backward_weights_desc_init⚠ | Initializes descriptor for inner product weights gradient primitive. |
| dnnl_inner_product_forward_desc_init⚠ | Initializes descriptor for inner product forward propagation. |
| dnnl_layer_normalization_backward_desc_init⚠ | Initializes a descriptor for a layer normalization backward propagation primitive. |
| dnnl_layer_normalization_forward_desc_init⚠ | Initializes a descriptor for layer normalization forward propagation primitive. |
| dnnl_lbr_gru_backward_desc_init⚠ | Initializes a descriptor for LBR GRU backward propagation primitive. |
| dnnl_lbr_gru_forward_desc_init⚠ | Initializes a descriptor for LBR GRU forward propagation primitive. |
| dnnl_logsoftmax_backward_desc_init⚠ | Initializes a descriptor for logsoftmax backward propagation primitive. |
| dnnl_logsoftmax_forward_desc_init⚠ | Initializes a descriptor for logsoftmax forward propagation primitive. |
| dnnl_lrn_backward_desc_init⚠ | Initializes a descriptor for LRN backward propagation primitive. |
| dnnl_lrn_forward_desc_init⚠ | Initializes a descriptor for LRN forward propagation primitive. |
| dnnl_lstm_backward_desc_init⚠ | Initializes a descriptor for an LSTM backward propagation primitive. |
| dnnl_lstm_backward_desc_init_v2⚠ | Initializes a descriptor for an LSTM (with or without peephole) backward propagation primitive. |
| dnnl_lstm_backward_desc_init_v3⚠ | Initializes a descriptor for an LSTM (with or without peephole and with or with out recurrent projection layer) backward propagation primitive. |
| dnnl_lstm_forward_desc_init⚠ | Initializes a descriptor for LSTM forward propagation primitive. |
| dnnl_lstm_forward_desc_init_v2⚠ | Initializes a descriptor for an LSTM (with or without peephole) forward propagation primitive. |
| dnnl_lstm_forward_desc_init_v3⚠ | Initializes a descriptor for an LSTM (with or without peephole and with or without recurrent projection layer) forward propagation primitive. |
| dnnl_matmul_desc_init⚠ | Initializes a matrix multiplication descriptor. |
| dnnl_memory_create⚠ | Creates a memory object. |
| dnnl_memory_desc_equal⚠ | Compares two memory descriptors. |
| dnnl_memory_desc_get_size⚠ | Returns the size of a memory descriptor. |
| dnnl_memory_desc_init_by_strides⚠ | Initializes a memory descriptor using dimensions and strides. |
| dnnl_memory_desc_init_by_tag⚠ | Initializes a memory descriptor using dimensions and memory format tag. |
| dnnl_memory_desc_init_submemory⚠ | @param memory_desc Output memory descriptor. @param parent_memory_desc An existing memory descriptor. @param dims Sizes of the region. @param offsets Offsets to the region from the encompassing memory object in each dimension @returns #dnnl_success on success and a status describing the error otherwise. |
| dnnl_memory_desc_permute_axes⚠ | Initializes a memory descriptor by permuting axes in an existing one. |
| dnnl_memory_desc_reshape⚠ | Initializes a memory descriptor by reshaping an existing one. The new memory descriptor inherits the data type. This operation is valid only for memory descriptors that have format_kind set to #dnnl_blocked or #dnnl_format_kind_any. |
| dnnl_memory_destroy⚠ | Destroys a memory object. |
| dnnl_memory_get_data_handle⚠ | Returns memory object's data handle. |
| dnnl_memory_get_engine⚠ | Returns the engine of a memory object. |
| dnnl_memory_get_memory_desc⚠ | Returns the memory descriptor for a memory object. |
| dnnl_memory_map_data⚠ | Maps a memory object and returns a host-side pointer to a memory buffer with a copy of its contents. |
| dnnl_memory_set_data_handle⚠ | Sets the underlying memory buffer. |
| dnnl_memory_set_data_handle_v2⚠ | Sets the underlying memory buffer. |
| dnnl_memory_unmap_data⚠ | Unmaps a memory object and writes back any changes made to the previously mapped memory buffer. The pointer to the mapped buffer must be obtained via the dnnl_memory_map_data() call. |
| dnnl_pooling_backward_desc_init⚠ | Initializes a descriptor for pooling backward propagation primitive. |
| dnnl_pooling_forward_desc_init⚠ | Initializes a descriptor for pooling forward propagation primitive. |
| dnnl_post_ops_append_dw_k3s1p1⚠ | Appends a depthwise post-op convolution with stride 1. |
| dnnl_post_ops_append_dw_k3s2p1⚠ | Appends a depthwise post-op convolution with stride 2. |
| dnnl_post_ops_append_eltwise⚠ | Appends an elementwise post-op. |
| dnnl_post_ops_append_sum⚠ | Appends an accumulation (sum) to post-ops. Prior to accumulating the result, the previous value is multiplied by a scale. |
| dnnl_post_ops_append_sum_v2⚠ | Appends an accumulation v2 (sum) to post-ops. Prior to accumulating the result, the previous value is multiplied by a scale. |
| dnnl_post_ops_create⚠ | Creates empty post-ops sequence. |
| dnnl_post_ops_destroy⚠ | Destroys post-ops. |
| dnnl_post_ops_get_kind⚠ | Returns the kind of a post-op entry. |
| dnnl_post_ops_get_params_dw_k3s1p1⚠ | Returns the parameters of an depthwise post-op with stride 1. |
| dnnl_post_ops_get_params_dw_k3s2p1⚠ | Returns the parameters of an depthwise post-op with stride 2. |
| dnnl_post_ops_get_params_eltwise⚠ | Returns the parameters of an elementwise post-up. |
| dnnl_post_ops_get_params_sum⚠ | Returns the parameters of an accumulation (sum) post-op. |
| dnnl_post_ops_get_params_sum_v2⚠ | Returns the parameters of an accumulation (sum) post-op with a data type parameter. |
| dnnl_post_ops_len⚠ | Returns the length of post-ops. |
| dnnl_primitive_attr_clone⚠ | Clones primitive attributes. |
| dnnl_primitive_attr_create⚠ | Creates an empty (default) primitive attributes with all the parameters set to their default values. |
| dnnl_primitive_attr_destroy⚠ | Destroys primitive attributes. |
| dnnl_primitive_attr_get_output_scales⚠ | Returns primitive attributes output scaling factors correspondence mask and values. |
| dnnl_primitive_attr_get_post_ops⚠ | Returns primitive attributes post-ops. |
| dnnl_primitive_attr_get_scales⚠ | Returns primitive attributes scaling factors correspondence mask and values for a given memory argument. |
| dnnl_primitive_attr_get_scratchpad_mode⚠ | Returns the primitive attributes scratchpad mode. |
| dnnl_primitive_attr_get_zero_points⚠ | Returns @p count, correspondence zero point @p mask, and a pointer to a constant int32_t array of @p zero_points for given @p attr and memory argument (index), previously set by dnnl_primitive_attr_set_zero_points. |
| dnnl_primitive_attr_set_output_scales⚠ | Sets output scaling factors correspondence mask and values. |
| dnnl_primitive_attr_set_post_ops⚠ | Sets primitive attributes post-ops. |
| dnnl_primitive_attr_set_rnn_data_qparams⚠ | Set quantization scale and shift parameters for RNN data tensors. |
| dnnl_primitive_attr_set_rnn_weights_qparams⚠ | Sets quantization scaling factors for RNN weights tensors. The low-precision configuration of the RNN primitives expects input weights to use the signed 8-bit integer data type. The scaling factors are used to quantize floating-point data to signed integer and must be passed to RNN primitives using attributes. |
| dnnl_primitive_attr_set_scales⚠ | Sets primitive attributes scaling factors for primitive operations for a given memory argument. |
| dnnl_primitive_attr_set_scratchpad_mode⚠ | Sets primitive attributes scratchpad mode. |
| dnnl_primitive_attr_set_zero_points⚠ | Sets primitive attributes zero points for primitive operations for a given memory argument. |
| dnnl_primitive_create⚠ | Creates a primitive. |
| dnnl_primitive_desc_clone⚠ | Clones a primitive descriptor. The resulting primitive descriptor must be destroyed separately. |
| dnnl_primitive_desc_create⚠ | Creates a primitive descriptor. This function is equivalent to a sequence of #dnnl_primitive_desc_iterator_create() and #dnnl_primitive_desc_iterator_fetch(). In other words, the library will pick the first suitable implementation. |
| dnnl_primitive_desc_destroy⚠ | Destroys a primitive descriptor. |
| dnnl_primitive_desc_get_attr⚠ | Returns a constant reference to the attributes of a primitive descriptor. |
| dnnl_primitive_desc_iterator_create⚠ | Creates a primitive descriptor iterator. |
| dnnl_primitive_desc_iterator_destroy⚠ | Destroys a primitive descriptor iterator. |
| dnnl_primitive_desc_iterator_fetch⚠ | Fetches the current primitive descriptor from a primitive descriptor iterator. |
| dnnl_primitive_desc_iterator_next⚠ | Advances the primitive descriptor iterator to point to the next available implementation. |
| dnnl_primitive_desc_query⚠ | Queries a primitive descriptor for various pieces of information. |
| dnnl_primitive_desc_query_md⚠ | Queries primitive descriptor for a memory descriptor. |
| dnnl_primitive_desc_query_s32⚠ | Queries primitive descriptor for a signed 32bit int. |
| dnnl_primitive_destroy⚠ | Destroys a primitive. |
| dnnl_primitive_execute⚠ | Executes a primitive. |
| dnnl_primitive_get_primitive_desc⚠ | Retrieves a constant reference to the primitive descriptor of a given primitive. |
| dnnl_reorder_primitive_desc_create⚠ | Creates a primitive descriptor for a reorder primitive. |
| dnnl_resampling_backward_desc_init⚠ | Initializes a descriptor for resampling backward propagation primitive. |
| dnnl_resampling_forward_desc_init⚠ | Initializes a descriptor for a resampling forward propagation primitive. |
| dnnl_set_jit_dump⚠ | Configures dumping of JIT-generated code. |
| dnnl_set_jit_profiling_flags⚠ | Sets library profiling flags. The flags define which profilers are supported. |
| dnnl_set_jit_profiling_jitdumpdir⚠ | Sets JIT dump output path. Only applicable to Linux and is only used when profiling flags have DNNL_JIT_PROFILE_LINUX_PERF bit set. |
| dnnl_set_max_cpu_isa⚠ | Sets the maximal ISA the library can dispatch to on the CPU. See #dnnl_cpu_isa_t and #dnnl::cpu_isa for the list of the values accepted by the C and C++ API functions respectively. |
| dnnl_set_primitive_cache_capacity⚠ | Sets a number of primitives that can be held in the primitive cache at a time. |
| dnnl_set_verbose⚠ | Configures verbose output to stdout. |
| dnnl_sgemm⚠ | Performs single-precision matrix-matrix multiply. |
| dnnl_shuffle_backward_desc_init⚠ | Initializes a descriptor for shuffle backward propagation primitive. |
| dnnl_shuffle_forward_desc_init⚠ | Initializes a descriptor for shuffle forward propagation primitive. |
| dnnl_softmax_backward_desc_init⚠ | Initializes a descriptor for softmax backward propagation primitive. |
| dnnl_softmax_forward_desc_init⚠ | Initializes a descriptor for softmax forward propagation primitive. |
| dnnl_stream_attr_create⚠ | Creates execution stream attributes for a stream that runs on an engine of a particular kind. |
| dnnl_stream_attr_destroy⚠ | Destroys execution stream attributes. |
| dnnl_stream_create⚠ | Creates an execution stream. |
| dnnl_stream_create_v2⚠ | Creates an execution stream. |
| dnnl_stream_destroy⚠ | Destroys an execution stream. |
| dnnl_stream_wait⚠ | Waits for all primitives in the execution stream to finish computations. |
| dnnl_sum_primitive_desc_create⚠ | Creates a primitive descriptor for an (out-of-place) sum primitive. |
| dnnl_vanilla_rnn_backward_desc_init⚠ | Initializes a descriptor for vanilla RNN backward propagation primitive. |
| dnnl_vanilla_rnn_forward_desc_init⚠ | Initializes a descriptor for vanilla RNN forward propagation primitive. |
| dnnl_version⚠ | Returns library version information. @returns Pointer to a constant structure containing |
Type Definitions
Unions
| _bindgen_ty_1 | @cond DO_NOT_DOCUMENT_THIS Hex representation for a special quiet NAN (!= NAN from math.h) |
| dnnl_memory_desc_t__bindgen_ty_1 |