1 /* 2 * Copyright (c) 2016-2022 Arm Limited. 3 * 4 * SPDX-License-Identifier: MIT 5 * 6 * Permission is hereby granted, free of charge, to any person obtaining a copy 7 * of this software and associated documentation files (the "Software"), to 8 * deal in the Software without restriction, including without limitation the 9 * rights to use, copy, modify, merge, publish, distribute, sublicense, and/or 10 * sell copies of the Software, and to permit persons to whom the Software is 11 * furnished to do so, subject to the following conditions: 12 * 13 * The above copyright notice and this permission notice shall be included in all 14 * copies or substantial portions of the Software. 15 * 16 * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR 17 * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, 18 * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE 19 * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER 20 * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, 21 * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE 22 * SOFTWARE. 23 */ 24 #ifndef ARM_COMPUTE_CL_CAST_KERNEL_H 25 #define ARM_COMPUTE_CL_CAST_KERNEL_H 26 27 #include "src/core/common/Macros.h" 28 #include "src/gpu/cl/ClCompileContext.h" 29 #include "src/gpu/cl/IClKernel.h" 30 31 namespace arm_compute 32 { 33 namespace opencl 34 { 35 namespace kernels 36 { 37 /** Casts a given tensor to a new type 38 * 39 * @note When casting between quantized types the scale and zeroPoint are ignored 40 */ 41 class ClCastKernel : public IClKernel 42 { 43 public: 44 ClCastKernel(); 45 ARM_COMPUTE_DISALLOW_COPY_ALLOW_MOVE(ClCastKernel); 46 /** Set the src and dst of the kernel. 47 * 48 * Valid conversions src -> dst : 49 * 50 * - QSYMM8_PER_CHANNEL -> QASYMM8 (ATTENTION: it is the user's responsibility to keep track of the quantization info in the TensorInfo meta-data) 51 * - U8 -> S8, U16, S16, U32, S32, F16, F32 52 * - S8 -> U8, U16, S16, U32, S32, F16, F32 53 * - U16 -> U8, S8, S16, U32, S32, F16, F32 54 * - S16 -> U8, S8, U16, U32, S32, F16, F32 55 * - U32 -> U8, S8, U16, S16, S32, F16, F32 56 * - S32 -> U8, S8, U16, S16, U32, F16, F32 57 * - F16 -> U8, S8, U16, S16, U32, F32 58 * - F32 -> U8, S8, U16, S16, U32, F16 59 * 60 * @param[in] compile_context The compile context to be used. 61 * @param[in] src The source tensor to convert. Data types supported: U8/S8/QSYMM8_PER_CHANNEL/U16/S16/U32/S32/F16/F32. 62 * @param[out] dst The destination tensor. Data types supported: U8/S8/QASYMM8/U16/S16/U32/S32/F16/F32. 63 * @param[in] policy Conversion policy 64 */ 65 void configure(const CLCompileContext &compile_context, const ITensorInfo *src, ITensorInfo *dst, ConvertPolicy policy); 66 /** Static function to check if given info will lead to a valid configuration 67 * 68 * Similar to @ref ClCastKernel::configure() 69 * 70 * @return a status 71 */ 72 static Status validate(const ITensorInfo *src, const ITensorInfo *dst, ConvertPolicy policy); 73 74 // Inherited methods overridden: 75 void run_op(ITensorPack &tensors, const Window &window, ::cl::CommandQueue &queue) override; 76 }; 77 } // namespace kernels 78 } // namespace opencl 79 } // namespace arm_compute 80 #endif /* ARM_COMPUTE_CL_CAST_KERNEL_H */ 81