24.02.1
|
Go to the documentation of this file.
47 if (output->total_size() != 0)
86 const int vec_size_x = 16 /
input->info()->element_size();
87 const int input_width_x =
input->info()->tensor_shape().x();
88 const unsigned int input_width_ceil =
ceil_to_multiple(input_width_x, vec_size_x);
89 const unsigned int input_width_tiles = input_width_ceil / vec_size_x;
90 const unsigned int offset = input_width_ceil - input_width_x;
91 const bool multi_access_x = (input_width_x / vec_size_x > 0);
115 const unsigned int size_win_x =
ceil_to_multiple(
input->info()->dimension(0), vec_size_x) * multiples[0];
119 ICLKernel::configure_internal(win);
125 for (
unsigned int i = 0; i < multiples.size(); ++i)
150 unsigned int idx = 0;
std::string to_string(T &&value)
Convert integer and float values to string.
constexpr int start() const
Return the start of the dimension.
const StringSet & options() const
Gets the current options list set.
Window calculate_max_window(const ValidRegion &valid_region, const Steps &steps, bool skip_border, BorderSize border_size)
std::string lower_string(const std::string &val)
Lower a given string.
Status validate_arguments(const ITensorInfo *src, const ITensorInfo *weights, const ITensorInfo *dst, const PadStrideInfo &conv_info)
#define ARM_COMPUTE_ERROR_ON_UNCONFIGURED_KERNEL(k)
static constexpr size_t DimX
Alias for dimension 0 also known as X dimension.
Window collapse_if_possible(const Window &full_window, size_t first, size_t last, bool *has_collapsed=nullptr) const
Collapse the dimensions between first and last if possible.
Interface for OpenCL tensor.
static Status validate(const ITensorInfo *input, const ITensorInfo *output, const Multiples &multiples)
Static function to check if given info will lead to a valid configuration of CLTileKernel.
const std::string & string_from_data_type(DataType dt)
Convert a data type identity into a string.
static CLKernelLibrary & get()
Access the KernelLibrary singleton.
#define ARM_COMPUTE_RETURN_ERROR_ON_MISMATCHING_DATA_TYPES(...)
#define ARM_COMPUTE_RETURN_ON_ERROR(status)
Checks if a status contains an error and returns it.
virtual size_t dimension(size_t index) const =0
Return the size of the requested dimension.
#define ARM_COMPUTE_ERROR_ON_NULLPTR(...)
virtual ITensorInfo * info() const =0
Interface to be implemented by the child class to return the tensor's metadata.
size_t data_size_from_type(DataType data_type)
The size in bytes of the data type.
#define ARM_COMPUTE_ERROR_THROW_ON(status)
void run(const Window &window, cl::CommandQueue &queue) override
Enqueue the OpenCL kernel to process the given window on the passed OpenCL command queue.
void add_option(std::string option)
Adds option to the existing build option list.
#define ARM_COMPUTE_RETURN_ERROR_ON(cond)
If the condition is true, an error is returned.
bool auto_init_if_empty(ITensorInfo &info, const TensorShape &shape, int num_channels, DataType data_type, QuantizationInfo quantization_info=QuantizationInfo())
Auto initialize the tensor info (shape, number of channels and data type) if the current assignment i...
cl::Kernel create_kernel(const CLCompileContext &ctx, const std::string &kernel_name, const std::set< std::string > &build_opts=std::set< std::string >())
Creates an opencl kernel using a compile context.
void add_option_if(bool cond, std::string option)
Adds option if a given condition is true;.
#define ARM_COMPUTE_ERROR_ON_INVALID_SUBWINDOW(f, s)
__global uchar * offset(const Image *img, int x, int y)
Get the pointer position of a Image.
TensorShape compute_tiled_shape(const TensorShape &input_shape, const Multiples &multiples)
Calculate the tiled shape of a tensor.
void add_4D_tensor_argument(unsigned int &idx, const ICLTensor *tensor, const Window &window)
Add the passed 4D tensor's parameters to the object's kernel's arguments starting from the index idx.
Describe one of the image's dimensions with a start, end and step.
void set(size_t dimension, const Dimension &dim)
Set the values of a given dimension.
void configure(const ICLTensor *input, ICLTensor *output, const Multiples &multiples)
Set the source, destination of the kernel.
auto ceil_to_multiple(S value, T divisor) -> decltype(((value+divisor - 1)/divisor) *divisor)
Computes the smallest number larger or equal to value that is a multiple of divisor.
const Window & window() const
The maximum window the kernel can be executed on.
std::string get_cl_unsigned_type_from_element_size(size_t element_size)
Translates the element size to an unsigned integer data type.
Describe a multidimensional execution window.
@ ELEMENTWISE
Elementwise CL kernel type.
Copyright (c) 2017-2024 Arm Limited.
#define ARM_COMPUTE_RETURN_ERROR_ON_MISMATCHING_DIMENSIONS(...)
std::vector< uint32_t > Multiples
Information to produce a tiled version of a Tensor.
static constexpr size_t DimZ
Alias for dimension 2 also known as Z dimension.
cl::NDRange lws_hint() const
Return the Local-Workgroup-Size hint.
#define ARM_COMPUTE_RETURN_ERROR_ON_NULLPTR(...)
Store the tensor's metadata.
Window first_slice_window_4D() const
First 4D slice of the window.
@ UNKNOWN
Unknown data type.
bool slide_window_slice_4D(Window &slice) const
Slide the passed 4D window slice.
constexpr const Dimension & x() const
Alias to access the first dimension of the window.
DataType
Available data types.
SimpleTensor< T > slice(const SimpleTensor< T > &src, Coordinates starts, Coordinates ends)
CLTileKernel()
Default constructor.
void enqueue(cl::CommandQueue &queue, ICLKernel &kernel, const Window &window, const cl::NDRange &lws_hint=CLKernelLibrary::get().default_ndrange(), bool use_dummy_work_items=false)
Add the kernel to the command queue with the given window.