diff options
author | SiCong Li <sicong.li@arm.com> | 2023-05-16 16:46:34 +0100 |
---|---|---|
committer | SiCong Li <sicong.li@arm.com> | 2023-07-04 09:06:41 +0000 |
commit | 19844f605f5e5b71d05164711dee13f8652adafe (patch) | |
tree | 99aa1242a37aefda01e4284bc48c63f93f625258 /src/dynamic_fusion/runtime | |
parent | 2f0ef00cc092b5eff901debee0c3f5a38114f40a (diff) | |
download | ComputeLibrary-19844f605f5e5b71d05164711dee13f8652adafe.tar.gz |
Add Kernel Writer driver code to dynamic fusion
* Partially port ElementwiseBinary component to ckw (broadcast not
supported yet)
* Port Store component to ckw
* Move KernelArgumentsHelpers to ckw_driver/ as it's only used by the
driver
ckw_driver is a middle layer between dynamic fusion and Compute Kernel
Writer (CKW). It consumes the fused kernel component stream produced by
Dynamic Fusion and uses CKW to write the kernel code complete with all
meta info needed by the runtime to enqueue the kernel.
It consists of two parts:
* Kernel writing: This resides in dynamic_fusion/sketch
* Runtime utilities: This resides in dynamic_fusion/runtime
The integration (separation between DF and CKW) occurs in two places:
* Inside GpuCKWDriver
global driver that coordinates how the
final fused kernel code is assembled together alongwith other meta
info needed by runtime.
* Inside each instantiated IGpuCKWComponentDriver
component driver that drives CKW to write component-specific code
or do component-specific configurations
Partially resolves: COMPMID-5792 COMPMID-6282 COMPMID-6260 COMPMID-6266
Signed-off-by: SiCong Li <sicong.li@arm.com>
Change-Id: Ib57a080a65fe8cfee1a8df1529fe572005a6d2f2
Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/9847
Reviewed-by: Gunes Bayir <gunes.bayir@arm.com>
Tested-by: Arm Jenkins <bsgcomp@arm.com>
Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
Benchmark: Arm Jenkins <bsgcomp@arm.com>
Diffstat (limited to 'src/dynamic_fusion/runtime')
3 files changed, 171 insertions, 0 deletions
diff --git a/src/dynamic_fusion/runtime/gpu/cl/ClKernelRuntime.cpp b/src/dynamic_fusion/runtime/gpu/cl/ClKernelRuntime.cpp index b3ec39362c..6a57482bb2 100644 --- a/src/dynamic_fusion/runtime/gpu/cl/ClKernelRuntime.cpp +++ b/src/dynamic_fusion/runtime/gpu/cl/ClKernelRuntime.cpp @@ -59,6 +59,8 @@ void ClKernelRuntime::configure(const ClCompileContext &compile_ctx, const GpuKe inline void ClKernelRuntime::add_tensor_argument(unsigned int &idx, const GpuKernelArgumentInfo &arg, const ICLTensor *tensor, const Window &arg_slice, std::vector<cl::Image2D> &cl_images) { + ARM_COMPUTE_ERROR_ON_NULLPTR(tensor); + switch(arg.type) { case GpuKernelArgumentInfo::Type::Scalar: @@ -142,6 +144,18 @@ inline void ClKernelRuntime::add_tensor_argument(unsigned int &idx, const GpuKer add_4d_tensor_nhwc_argument(idx, tensor); break; } + case GpuKernelArgumentInfo::Type::Tensor_Special_0: + { + const ITensorInfo *info = tensor->info(); + const Strides &strides = info->strides_in_bytes(); + + _kernel.setArg(idx++, tensor->cl_buffer()); + const size_t dim1xdim2 = info->tensor_shape()[1] * info->tensor_shape()[2]; + _kernel.setArg<cl_int>(idx++, static_cast<int32_t>(dim1xdim2)); + const size_t stride1 = strides[1]; + _kernel.setArg<cl_int>(idx++, static_cast<int32_t>(stride1)); + break; + } default: { ARM_COMPUTE_ERROR("Unsupported"); diff --git a/src/dynamic_fusion/runtime/gpu/cl/ckw_driver/GpuCkwKernelArgumentsHelpers.cpp b/src/dynamic_fusion/runtime/gpu/cl/ckw_driver/GpuCkwKernelArgumentsHelpers.cpp new file mode 100644 index 0000000000..40934cf6e9 --- /dev/null +++ b/src/dynamic_fusion/runtime/gpu/cl/ckw_driver/GpuCkwKernelArgumentsHelpers.cpp @@ -0,0 +1,96 @@ +/* + * Copyright (c) 2023 Arm Limited. + * + * SPDX-License-Identifier: MIT + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to + * deal in the Software without restriction, including without limitation the + * rights to use, copy, modify, merge, publish, distribute, sublicense, and/or + * sell copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#include "GpuCkwKernelArgumentsHelpers.h" + +namespace arm_compute +{ +void cl_add_tensor_component_argument(cl::Kernel &kernel, unsigned int &idx, ICLTensor *tensor, ckw::TensorComponent component) +{ + ARM_COMPUTE_ERROR_ON(tensor == nullptr); + + const auto *info = tensor->info(); + const auto &strides = info->strides_in_bytes(); + + switch(component) + { + case ckw::TensorComponent::OffsetFirstElement: + kernel.setArg<cl_uint>(idx++, info->offset_first_element_in_bytes()); + break; + case ckw::TensorComponent::Stride0: + kernel.setArg<cl_uint>(idx++, strides[0]); + break; + case ckw::TensorComponent::Stride1: + kernel.setArg<cl_uint>(idx++, strides[1]); + break; + case ckw::TensorComponent::Stride2: + kernel.setArg<cl_uint>(idx++, strides[2]); + break; + case ckw::TensorComponent::Stride3: + kernel.setArg<cl_uint>(idx++, strides[3]); + break; + case ckw::TensorComponent::Stride4: + kernel.setArg<cl_uint>(idx++, strides[4]); + break; + case ckw::TensorComponent::Dim0: + kernel.setArg<cl_uint>(idx++, info->dimension(0)); + break; + case ckw::TensorComponent::Dim1: + kernel.setArg<cl_uint>(idx++, info->dimension(1)); + break; + case ckw::TensorComponent::Dim2: + kernel.setArg<cl_uint>(idx++, info->dimension(2)); + break; + case ckw::TensorComponent::Dim3: + kernel.setArg<cl_uint>(idx++, info->dimension(3)); + break; + case ckw::TensorComponent::Dim4: + kernel.setArg<cl_uint>(idx++, info->dimension(4)); + break; + case ckw::TensorComponent::Dim1xDim2: + kernel.setArg<cl_uint>(idx++, info->dimension(1) * info->dimension(2)); + break; + case ckw::TensorComponent::Dim2xDim3: + kernel.setArg<cl_uint>(idx++, info->dimension(2) * info->dimension(3)); + break; + case ckw::TensorComponent::Dim1xDim2xDim3: + kernel.setArg<cl_uint>(idx++, info->dimension(1) * info->dimension(2) * info->dimension(3)); + break; + case ckw::TensorComponent::Unknown: + default: + ARM_COMPUTE_ERROR("Unknown tensor component"); + } +} + +void cl_add_buffer_argument(cl::Kernel &kernel, unsigned int &idx, const cl::Buffer &buffer) +{ + kernel.setArg(idx++, buffer); +} + +void cl_add_texture_argument(cl::Kernel &kernel, unsigned int &idx, const cl::Image &image) +{ + kernel.setArg(idx++, image); +} + +} // namespace arm_compute diff --git a/src/dynamic_fusion/runtime/gpu/cl/ckw_driver/GpuCkwKernelArgumentsHelpers.h b/src/dynamic_fusion/runtime/gpu/cl/ckw_driver/GpuCkwKernelArgumentsHelpers.h new file mode 100644 index 0000000000..206ad2e401 --- /dev/null +++ b/src/dynamic_fusion/runtime/gpu/cl/ckw_driver/GpuCkwKernelArgumentsHelpers.h @@ -0,0 +1,61 @@ +/* + * Copyright (c) 2023 Arm Limited. + * + * SPDX-License-Identifier: MIT + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to + * deal in the Software without restriction, including without limitation the + * rights to use, copy, modify, merge, publish, distribute, sublicense, and/or + * sell copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#ifndef ACL_SRC_DYNAMIC_FUSION_RUNTIME_GPU_CL_CKW_DRIVER_GPUCKWKERNELARGUMENTSHELPERS +#define ACL_SRC_DYNAMIC_FUSION_RUNTIME_GPU_CL_CKW_DRIVER_GPUCKWKERNELARGUMENTSHELPERS + +#include "arm_compute/core/CL/ICLTensor.h" + +#include "ckw/TensorInfo.h" + +namespace arm_compute +{ +/** Select a Compute Kernel Writer tensor component from a tensor and add to the kernel's arguments at the specified index idx. + * + * @param[in,out] kernel OpenCL kernel to configure with the provided argument. + * @param[in,out] idx Index at which to add the argument. + * @param[in] tensor Tensor from which to access the tensor component. + * @param[in] component Tensor component to select such as tensor dimensions, strides, etc. + */ +void cl_add_tensor_component_argument(cl::Kernel &kernel, unsigned int &idx, ICLTensor *tensor, ckw::TensorComponent component); + +/** Add an OpenCL buffer object to the kernel's arguments at the specified index @p idx. + * + * @param[in,out] kernel OpenCL kernel to configure with the provided argument. + * @param[in,out] idx Index at which to add the argument. + * @param[in] buffer OpenCL buffer containing the tensor's data. + */ +void cl_add_buffer_argument(cl::Kernel &kernel, unsigned int &idx, const cl::Buffer &buffer); + +/** Add an OpenCL image object to the kernel's arguments at the specified index @p idx. + * + * @param[in,out] kernel OpenCL kernel to configure with the provided argument. + * @param[in,out] idx Index at which to add the argument. + * @param[in] image OpenCL image containing the image's data. + */ +void cl_add_texture_argument(cl::Kernel &kernel, unsigned int &idx, const cl::Image &image); + +} // namespace arm_compute + +#endif /* ACL_SRC_DYNAMIC_FUSION_RUNTIME_GPU_CL_CKW_DRIVER_GPUCKWKERNELARGUMENTSHELPERS */ |