Port CLGEMMConvolutionLayer
Details:
port CLWeightsReshapeKernel to ClWeightsReshapeKernel
port CLGEMMConvolutionLayer to ClGemmConvolution
Resolves: COMPMID-4515
Change-Id: I7d5b4ec72db2742f6eb9f3ffc88f717c35b4f2a3
Signed-off-by: Manuel Bottini <manuel.bottini@arm.com>
Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/5983
Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
Reviewed-by: Michele Di Giorgio <michele.digiorgio@arm.com>
Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com>
Tested-by: Arm Jenkins <bsgcomp@arm.com>
diff --git a/src/runtime/gpu/cl/operators/ClGemm.cpp b/src/runtime/gpu/cl/operators/ClGemm.cpp
index cb0eeca..2792dc4 100644
--- a/src/runtime/gpu/cl/operators/ClGemm.cpp
+++ b/src/runtime/gpu/cl/operators/ClGemm.cpp
@@ -208,6 +208,7 @@
_tmp_b(),
_reshape_b_only_on_first_run(false),
_gemm_kernel_type(CLGEMMKernelType::NATIVE_V1),
+ _is_prepared(false),
_aux_mem(AuxTensorIdx::Count)
{
}
@@ -696,6 +697,7 @@
}
ITensorPack gemm_reshaped_pack{ { ACL_SRC_0, lhs_reshaped.get() }, { ACL_SRC_1, rhs_reshaped.get() }, { ACL_SRC_2, src2 }, { ACL_DST, dst } };
+
if(_gemm_kernel_type == CLGEMMKernelType::RESHAPED)
{
CLScheduler::get().enqueue_op(*_mm_reshaped_kernel, gemm_reshaped_pack, true);
@@ -740,19 +742,23 @@
void ClGemm::prepare(ITensorPack &constants)
{
- const ITensor *src1 = constants.get_const_tensor(ACL_SRC_1);
- ICLTensor *rhs_aux = utils::cast::polymorphic_downcast<ICLTensor *>(constants.get_tensor(offset_int_vec(RhsReshape)));
-
- // If memory for RHS is persistent and src1 is provided re-transform else assume that RHS is transformed
- if((_aux_mem[AuxTensorIdx::RhsReshape].lifetime == MemoryLifetime::Persistent) && (src1 != nullptr && rhs_aux != nullptr) && rhs_aux)
+ if(!_is_prepared)
{
- ARM_COMPUTE_LOG_INFO_WITH_FUNCNAME_ACL("Transforming RHS Matrix!");
+ const ITensor *src1 = constants.get_const_tensor(ACL_SRC_1);
+ ICLTensor *rhs_aux = utils::cast::polymorphic_downcast<ICLTensor *>(constants.get_tensor(offset_int_vec(RhsReshape)));
- CLAuxTensorHandler rhs_reshaped(_tmp_b, *rhs_aux);
- ARM_COMPUTE_ERROR_ON(rhs_reshaped.get()->cl_buffer().get() == nullptr);
+ // If memory for RHS is persistent and src1 is provided re-transform else assume that RHS is transformed
+ if((_aux_mem[AuxTensorIdx::RhsReshape].lifetime == MemoryLifetime::Persistent) && (src1 != nullptr && rhs_aux != nullptr) && rhs_aux)
+ {
+ ARM_COMPUTE_LOG_INFO_WITH_FUNCNAME_ACL("Transforming RHS Matrix!");
- ITensorPack reshape_rhs_pack{ { ACL_SRC, src1 }, { ACL_DST, rhs_reshaped.get() } };
- CLScheduler::get().enqueue_op(*_reshape_rhs_kernel, reshape_rhs_pack, true);
+ CLAuxTensorHandler rhs_reshaped(_tmp_b, *rhs_aux);
+ ARM_COMPUTE_ERROR_ON(rhs_reshaped.get()->cl_buffer().get() == nullptr);
+
+ ITensorPack reshape_rhs_pack{ { ACL_SRC, src1 }, { ACL_DST, rhs_reshaped.get() } };
+ CLScheduler::get().enqueue_op(*_reshape_rhs_kernel, reshape_rhs_pack, true);
+ }
+ _is_prepared = true;
}
}
diff --git a/src/runtime/gpu/cl/operators/ClGemm.h b/src/runtime/gpu/cl/operators/ClGemm.h
index aad208b..254344e 100644
--- a/src/runtime/gpu/cl/operators/ClGemm.h
+++ b/src/runtime/gpu/cl/operators/ClGemm.h
@@ -129,8 +129,8 @@
TensorInfo _tmp_b;
bool _reshape_b_only_on_first_run;
CLGEMMKernelType _gemm_kernel_type;
-
- experimental::MemoryRequirements _aux_mem{};
+ bool _is_prepared;
+ experimental::MemoryRequirements _aux_mem{};
};
} // namespace opencl
} // namespace arm_compute
diff --git a/src/runtime/gpu/cl/operators/ClGemmConvolution.cpp b/src/runtime/gpu/cl/operators/ClGemmConvolution.cpp
new file mode 100644
index 0000000..1926cbb
--- /dev/null
+++ b/src/runtime/gpu/cl/operators/ClGemmConvolution.cpp
@@ -0,0 +1,628 @@
+/*
+ * Copyright (c) 2017-2021 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 "src/runtime/gpu/cl/operators/ClGemmConvolution.h"
+
+#include "arm_compute/core/CL/ICLTensor.h"
+#include "arm_compute/core/PixelValue.h"
+#include "arm_compute/core/Size2D.h"
+#include "arm_compute/core/TensorInfo.h"
+#include "arm_compute/core/Utils.h"
+#include "arm_compute/core/Validate.h"
+#include "arm_compute/core/utils/misc/ShapeCalculator.h"
+#include "arm_compute/core/utils/quantization/AsymmHelpers.h"
+#include "arm_compute/runtime/CL/CLScheduler.h"
+#include "src/core/gpu/cl/kernels/ClActivationKernel.h"
+#include "src/core/gpu/cl/kernels/ClCol2ImKernel.h"
+#include "src/core/gpu/cl/kernels/ClIm2ColKernel.h"
+#include "src/core/gpu/cl/kernels/ClWeightsReshapeKernel.h"
+#include "src/core/helpers/AutoConfiguration.h"
+#include "src/core/helpers/MemoryHelpers.h"
+#include "src/runtime/gpu/cl/operators/ClGemm.h"
+#include "src/runtime/gpu/cl/operators/ClGemmLowpMatrixMultiplyCore.h"
+#include "src/runtime/gpu/cl/utils/ClAuxTensorHandler.h"
+#include "support/Cast.h"
+
+namespace arm_compute
+{
+using namespace experimental;
+using namespace misc::shape_calculator;
+using namespace utils::cast;
+namespace opencl
+{
+ClGemmConvolution::ClGemmConvolution()
+ : _weights_reshape_kernel(nullptr), _im2col_kernel(nullptr), _mm_gemm(nullptr), _mm_gemmlowp(nullptr), _col2im_kernel(nullptr), _activation_kernel(nullptr), _im2col_output(), _weights_reshaped(),
+ _gemm_output(), _skip_im2col(false), _skip_col2im(false), _is_quantized(false), _fuse_activation(true), _append_bias(false), _is_prepared(false), _aux_mem(AuxTensorIdx::Count)
+{
+}
+ClGemmConvolution::~ClGemmConvolution() = default;
+
+void ClGemmConvolution::configure_mm(const ClCompileContext &compile_context, const ITensorInfo *src, ITensorInfo *weights, ITensorInfo *biases, ITensorInfo *dst,
+ const GEMMLowpOutputStageInfo &gemmlowp_output_stage,
+ int gemm_3d_depth, const ActivationLayerInfo &act_info)
+{
+ ARM_COMPUTE_ERROR_ON_NULLPTR(src, weights);
+ ARM_COMPUTE_ERROR_THROW_ON(validate_mm(src, weights, biases, dst, gemmlowp_output_stage, gemm_3d_depth, _skip_im2col, act_info));
+
+ const GEMMInfo &gemm_info = GEMMInfo(false, // is_a_reshaped
+ false, // is_b_reshaped
+ true, // reshape_b_only_on_first_run
+ gemm_3d_depth, // depth_output_gemm3d
+ _skip_im2col, // reinterpret_input_as_3d
+ false, // retain_internal_weights
+ gemmlowp_output_stage, // gemmlowp_output_stage
+ false, // fast_math
+ false, // fp_mixed_precision
+ true, // broadcast_bias
+ act_info); // activation_info
+
+ TensorInfo tmp_src{ *src };
+ if(_is_quantized)
+ {
+ // Since we need negative offsets for computing convolution, we need to change QuantizationInfo()
+ // Extract and negate input and weights offset
+ const QuantizationInfo input_quantization_info = src->quantization_info();
+ const QuantizationInfo weights_quantization_info = weights->quantization_info();
+
+ tmp_src.set_quantization_info(QuantizationInfo(input_quantization_info.uniform().scale, -input_quantization_info.uniform().offset));
+ weights->set_quantization_info(QuantizationInfo(weights_quantization_info.uniform().scale, -weights_quantization_info.uniform().offset));
+
+ _mm_gemmlowp = std::make_unique<ClGemmLowpMatrixMultiplyCore>();
+ _mm_gemmlowp->configure(compile_context, &tmp_src, weights, biases, dst, gemm_info);
+
+ // Revert back QuantizatioInfo as weights could be used in other convolution layers
+ weights->set_quantization_info(weights_quantization_info);
+
+ auto mm_mem_req = _mm_gemmlowp->workspace();
+ for(unsigned int cont = 0; cont < mm_mem_req.size(); ++cont)
+ {
+ _aux_mem[cont] = mm_mem_req[cont];
+ }
+ }
+ else
+ {
+ // Configure matrix multiply function
+ _mm_gemm = std::make_unique<ClGemm>();
+ _mm_gemm->configure(compile_context, &tmp_src, weights, biases, dst, 1.0f, 1.0f, gemm_info);
+ auto mm_mem_req = _mm_gemm->workspace();
+ for(unsigned int cont = 0; cont < mm_mem_req.size(); ++cont)
+ {
+ _aux_mem[cont] = mm_mem_req[cont];
+ }
+ }
+}
+
+Status ClGemmConvolution::validate_mm(const ITensorInfo *src, const ITensorInfo *weights, const ITensorInfo *biases, const ITensorInfo *dst,
+ const GEMMLowpOutputStageInfo &gemmlowp_output_stage, int gemm_3d_depth, bool skip_im2col, const ActivationLayerInfo &act_info)
+{
+ const bool is_quantized = is_data_type_quantized_asymmetric(src->data_type());
+
+ const GEMMInfo &gemm_info = GEMMInfo(false, // is_a_reshaped
+ false, // is_b_reshaped
+ true, // reshape_b_only_on_first_run
+ gemm_3d_depth, // depth_output_gemm3d
+ skip_im2col, // reinterpret_input_as_3d
+ false, // retain_internal_weights
+ gemmlowp_output_stage, // gemmlowp_output_stage
+ false, // fast_math
+ false, // fp_mixed_precision
+ true, // broadcast_bias
+ act_info); // activation_info
+
+ if(is_quantized)
+ {
+ // Since we need negative offsets for computing convolution, we need to change QuantizationInfo()
+ // Extract and negate input and weights offset
+ const QuantizationInfo input_quantization_info = src->quantization_info();
+ const QuantizationInfo weights_quantization_info = weights->quantization_info();
+
+ std::unique_ptr<ITensorInfo> src_qa = src->clone();
+ std::unique_ptr<ITensorInfo> weights_qa = weights->clone();
+ src_qa->set_quantization_info(QuantizationInfo(input_quantization_info.uniform().scale, -input_quantization_info.uniform().offset));
+ weights_qa->set_quantization_info(QuantizationInfo(weights_quantization_info.uniform().scale, -weights_quantization_info.uniform().offset));
+
+ // Perform validation step on GEMMLowp
+ return ClGemmLowpMatrixMultiplyCore::validate(src_qa.get(), weights_qa.get(), biases, dst, gemm_info);
+ }
+ else
+ {
+ // Perform validation step on Matrix multiply function
+ return ClGemm::validate(src, weights, biases, dst, 1.0f, 1.0f, gemm_info);
+ }
+}
+
+void ClGemmConvolution::configure(const CLCompileContext &compile_context, ITensorInfo *src, ITensorInfo *weights, ITensorInfo *biases, ITensorInfo *dst,
+ const Conv2dInfo &conv2d_info, const WeightsInfo &weights_info)
+{
+ ARM_COMPUTE_ERROR_ON_NULLPTR(src, weights, dst);
+
+ ARM_COMPUTE_ERROR_THROW_ON(ClGemmConvolution::validate(src, weights, biases, dst,
+ conv2d_info,
+ weights_info));
+
+ const DataType data_type = src->data_type();
+ const DataLayout data_layout = src->data_layout();
+ const int idx_width = get_data_layout_dimension_index(data_layout, DataLayoutDimension::WIDTH);
+ const int idx_height = get_data_layout_dimension_index(data_layout, DataLayoutDimension::HEIGHT);
+ const int idx_kernels = get_data_layout_dimension_index(data_layout, DataLayoutDimension::BATCHES);
+
+ const unsigned int kernel_width = weights->dimension(idx_width);
+ const unsigned int kernel_height = weights->dimension(idx_height);
+ const unsigned int num_kernels = weights->dimension(idx_kernels);
+
+ const UniformQuantizationInfo iq_info = src->quantization_info().uniform();
+ const UniformQuantizationInfo oq_info = dst->quantization_info().uniform();
+
+ _is_prepared = weights_info.retain_internal_weights();
+ _is_quantized = is_data_type_quantized_asymmetric(src->data_type());
+ _skip_im2col = (data_layout == DataLayout::NHWC && kernel_width == 1 && kernel_height == 1 && conv2d_info.conv_info.stride().first == 1 && conv2d_info.conv_info.stride().second == 1);
+ _skip_col2im = data_layout == DataLayout::NHWC;
+
+ // Only for quantize there are few cases where we cannot fuse the activation function in GEMM
+ _fuse_activation = true;
+
+ const ITensorInfo *gemm_input_to_use = src;
+ ITensorInfo *gemm_output_to_use = dst;
+
+ // Get parameters from conv_info
+ unsigned int stride_x = 0;
+ unsigned int stride_y = 0;
+ std::tie(stride_x, stride_y) = conv2d_info.conv_info.stride();
+
+ // Get convolved dimensions
+ unsigned int conv_w = 0;
+ unsigned int conv_h = 0;
+ std::tie(conv_w, conv_h) = scaled_dimensions(src->dimension(idx_width),
+ src->dimension(idx_height),
+ kernel_width,
+ kernel_height,
+ conv2d_info.conv_info,
+ conv2d_info.dilation);
+
+ unsigned int mat_weights_cols = num_kernels / conv2d_info.num_groups;
+
+ ITensorInfo *biases_to_use = biases;
+ _append_bias = false;
+
+ _weights_reshape_kernel = std::make_unique<kernels::ClWeightsReshapeKernel>();
+ if(conv2d_info.num_groups != 1 && biases != nullptr)
+ {
+ // num_groups != 1 can only be for NCHW
+ // Since it is missing an utility function to reshape the biases, we append the biases into the weights tensor
+ biases_to_use = nullptr;
+ _append_bias = true;
+ _weights_reshape_kernel->configure(compile_context, weights, biases, &_weights_reshaped, conv2d_info.num_groups);
+ }
+ else
+ {
+ _weights_reshape_kernel->configure(compile_context, weights, nullptr, &_weights_reshaped, conv2d_info.num_groups);
+ }
+
+ // Create tensor to store im2col reshaped inputs
+ if(!_skip_im2col)
+ {
+ // Configure and tune im2col. im2col output shape is auto-initialized
+ _im2col_kernel = std::make_unique<opencl::kernels::ClIm2ColKernel>();
+
+ // Set the GPU target for im2col
+ _im2col_kernel->set_target(CLScheduler::get().target());
+ _im2col_kernel->configure(compile_context, src, &_im2col_output, Size2D(kernel_width, kernel_height), conv2d_info.conv_info, _append_bias, conv2d_info.dilation, conv2d_info.num_groups);
+
+ // Set quantization info
+ _im2col_output.set_quantization_info(src->quantization_info());
+ CLScheduler::get().tune_kernel_static(*_im2col_kernel);
+
+ // Update GEMM input
+ gemm_input_to_use = &_im2col_output;
+ }
+
+ // Create GEMM output tensor
+ if(!_skip_col2im)
+ {
+ TensorShape shape_gemm;
+
+ // If we cannot skip col2im it means we run im2col as well
+ shape_gemm = _im2col_output.tensor_shape();
+ shape_gemm.set(0, mat_weights_cols);
+ shape_gemm.set(1, conv_w * conv_h);
+
+ _gemm_output = TensorInfo(shape_gemm, 1, data_type);
+ _gemm_output.set_quantization_info(dst->quantization_info()).set_data_layout(src->data_layout());
+
+ // Update GEMM output
+ gemm_output_to_use = &_gemm_output;
+ }
+
+ GEMMLowpOutputStageInfo gemmlowp_output_stage;
+ gemmlowp_output_stage.type = GEMMLowpOutputStageType::QUANTIZE_DOWN_FIXEDPOINT;
+ gemmlowp_output_stage.gemmlowp_offset = 0;
+
+ // Configure output stage for quantized case
+ if(_is_quantized)
+ {
+ const auto output_quant_info = (dst->total_size() == 0) ? iq_info : oq_info;
+ const bool is_quantized_per_channel = is_data_type_quantized_per_channel(weights->data_type());
+ const unsigned int num_filters = (is_quantized_per_channel) ? num_kernels : 1;
+
+ gemmlowp_output_stage.is_quantized_per_channel = is_quantized_per_channel;
+
+ gemmlowp_output_stage.gemmlowp_multipliers.resize(num_filters);
+ gemmlowp_output_stage.gemmlowp_shifts.resize(num_filters);
+ quantization::compute_quantized_multipliers_and_shifts(src, weights, dst,
+ gemmlowp_output_stage.gemmlowp_multipliers.data(),
+ gemmlowp_output_stage.gemmlowp_shifts.data());
+ gemmlowp_output_stage.gemmlowp_multiplier = gemmlowp_output_stage.gemmlowp_multipliers[0];
+ gemmlowp_output_stage.gemmlowp_shift = gemmlowp_output_stage.gemmlowp_shifts[0];
+
+ PixelValue min_val{};
+ PixelValue max_val{};
+ std::tie(min_val, max_val) = get_min_max(dst->data_type());
+
+ auto min_activation = min_val.get<int32_t>();
+ auto max_activation = max_val.get<int32_t>();
+
+ const std::set<ActivationLayerInfo::ActivationFunction> supported_acts = { ActivationLayerInfo::ActivationFunction::RELU,
+ ActivationLayerInfo::ActivationFunction::BOUNDED_RELU,
+ ActivationLayerInfo::ActivationFunction::LU_BOUNDED_RELU
+ };
+
+ if(conv2d_info.act_info.enabled())
+ {
+ if(supported_acts.count(conv2d_info.act_info.activation()) != 0)
+ {
+ std::tie(min_activation, max_activation) = get_quantized_activation_min_max(conv2d_info.act_info, data_type, output_quant_info);
+ }
+ else
+ {
+ _fuse_activation = false;
+ }
+ }
+
+ // Set the GEMMLowp output stage info
+ gemmlowp_output_stage.gemmlowp_offset = output_quant_info.offset;
+ gemmlowp_output_stage.gemmlowp_min_bound = min_activation;
+ gemmlowp_output_stage.gemmlowp_max_bound = max_activation;
+ }
+
+ // Configure and tune GEMM
+ // In case of NHWC, we need to run GEMM3D (gemm_3d_depth != 0) in order to avoid reshaping the output matrix
+ const unsigned int gemm_3d_depth = (data_layout == DataLayout::NHWC) ? conv_h : 0;
+
+ configure_mm(compile_context, gemm_input_to_use, &_weights_reshaped, biases_to_use, gemm_output_to_use, gemmlowp_output_stage, gemm_3d_depth, conv2d_info.act_info);
+
+ if(!_skip_col2im)
+ {
+ // Set the GPU target for col2im
+ _col2im_kernel = std::make_unique<opencl::kernels::ClCol2ImKernel>();
+ _col2im_kernel->set_target(CLScheduler::get().target());
+ // Configure and tune Col2Im
+ _col2im_kernel->configure(compile_context, gemm_output_to_use, dst, Size2D(conv_w, conv_h), conv2d_info.num_groups);
+ CLScheduler::get().tune_kernel_static(*_col2im_kernel.get());
+ }
+
+ ARM_COMPUTE_ERROR_ON_MSG((dst->dimension(idx_width) != conv_w) || (dst->dimension(idx_height) != conv_h),
+ "Output shape does not match the expected one");
+
+ if(!_fuse_activation)
+ {
+ _activation_kernel = std::make_unique<opencl::kernels::ClActivationKernel>();
+ _activation_kernel->configure(compile_context, dst, nullptr, conv2d_info.act_info);
+ }
+
+ _aux_mem[Im2ColOutput] = MemoryInfo(offset_int_vec(Im2ColOutput), MemoryLifetime::Temporary, _im2col_output.total_size());
+ _aux_mem[WeightsReshaped] = MemoryInfo(offset_int_vec(WeightsReshaped), MemoryLifetime::Persistent, _weights_reshaped.total_size());
+ _aux_mem[GemmOutput] = MemoryInfo(offset_int_vec(GemmOutput), MemoryLifetime::Temporary, _gemm_output.total_size());
+}
+
+Status ClGemmConvolution::validate(const ITensorInfo *src, const ITensorInfo *weights, const ITensorInfo *biases, const ITensorInfo *dst, const Conv2dInfo &conv2d_info,
+ const WeightsInfo &weights_info)
+{
+ ARM_COMPUTE_RETURN_ERROR_ON_NULLPTR(src, weights, dst);
+ ARM_COMPUTE_RETURN_ERROR_ON_MSG(weights_info.are_reshaped(), "Weights already reshaped are not supported!");
+ ARM_COMPUTE_RETURN_ERROR_ON_DATA_TYPE_CHANNEL_NOT_IN(src, 1, DataType::QASYMM8, DataType::QASYMM8_SIGNED, DataType::F16, DataType::F32);
+ const bool is_quantized_per_channel = is_data_type_quantized_per_channel(weights->data_type());
+
+ if(!is_quantized_per_channel)
+ {
+ ARM_COMPUTE_RETURN_ERROR_ON_MISMATCHING_DATA_TYPES(src, weights);
+ }
+ ARM_COMPUTE_RETURN_ERROR_ON_MISMATCHING_DATA_LAYOUT(src, weights);
+ ARM_COMPUTE_RETURN_ERROR_ON_MSG((conv2d_info.num_groups != 1) && (src->data_layout() != DataLayout::NCHW), "Grouping (num_groups != 1) with NHWC data layout is not supported");
+ ARM_COMPUTE_RETURN_ERROR_ON_MSG((conv2d_info.num_groups != 1) && (src->data_type() == DataType::QASYMM8), "Grouping (num_groups != 1) is not supported with QASYMM8");
+ ARM_COMPUTE_RETURN_ERROR_ON(((src->dimension(2) / weights->dimension(2)) != conv2d_info.num_groups) && (src->data_layout() == DataLayout::NCHW));
+
+ const DataLayout data_layout = src->data_layout();
+ const DataType data_type = src->data_type();
+ const int idx_width = get_data_layout_dimension_index(data_layout, DataLayoutDimension::WIDTH);
+ const int idx_height = get_data_layout_dimension_index(data_layout, DataLayoutDimension::HEIGHT);
+ const int idx_channel = get_data_layout_dimension_index(data_layout, DataLayoutDimension::CHANNEL);
+ const int idx_kernels = get_data_layout_dimension_index(data_layout, DataLayoutDimension::BATCHES);
+
+ const unsigned int kernel_width = weights->dimension(idx_width);
+ const unsigned int kernel_height = weights->dimension(idx_height);
+ const unsigned int num_kernels = weights->dimension(idx_kernels);
+
+ TensorInfo im2col_reshaped_info{};
+ TensorInfo info_gemm{};
+ TensorInfo weights_reshaped_info{};
+ const ITensorInfo *gemm_input_to_use = src;
+ const ITensorInfo *gemm_output_to_use = dst;
+ const ITensorInfo *weights_to_use = weights;
+ const bool is_quantized = is_data_type_quantized_asymmetric(data_type);
+ const bool skip_im2col = (data_layout == DataLayout::NHWC && kernel_width == 1 && kernel_height == 1 && conv2d_info.conv_info.stride().first == 1
+ && conv2d_info.conv_info.stride().second == 1);
+ const bool skip_col2im = data_layout == DataLayout::NHWC;
+ bool fuse_activation = true;
+
+ ARM_COMPUTE_RETURN_ERROR_ON((weights->dimension(idx_channel) * conv2d_info.num_groups) != src->dimension(idx_channel));
+ ARM_COMPUTE_RETURN_ERROR_ON(weights->num_dimensions() > 4);
+
+ // Validate biases
+ if(biases != nullptr)
+ {
+ if(is_quantized)
+ {
+ ARM_COMPUTE_RETURN_ERROR_ON_DATA_TYPE_CHANNEL_NOT_IN(biases, 1, DataType::S32);
+ }
+ else
+ {
+ ARM_COMPUTE_RETURN_ERROR_ON_MISMATCHING_DATA_TYPES(src, biases);
+ }
+ ARM_COMPUTE_RETURN_ERROR_ON(biases->dimension(0) != weights->dimension(idx_kernels));
+ ARM_COMPUTE_RETURN_ERROR_ON(biases->num_dimensions() > 1);
+ }
+
+ if(conv2d_info.act_info.enabled())
+ {
+ ARM_COMPUTE_ERROR_ON(conv2d_info.act_info.b() > conv2d_info.act_info.a());
+ }
+
+ // Get convolved dimensions
+ unsigned int conv_w = 0;
+ unsigned int conv_h = 0;
+
+ std::tie(conv_w, conv_h) = scaled_dimensions(src->dimension(idx_width),
+ src->dimension(idx_height),
+ kernel_width,
+ kernel_height,
+ conv2d_info.conv_info,
+ conv2d_info.dilation);
+
+ unsigned int mat_weights_cols = num_kernels / conv2d_info.num_groups;
+
+ const ITensorInfo *biases_to_use = biases;
+ bool append_bias = false;
+
+ if(conv2d_info.num_groups != 1 && biases != nullptr)
+ {
+ // num_groups != 1 can only be for NCHW
+ // Since it is missing an utility function to reshape the biases, we append the biases into the weights tensor
+ biases_to_use = nullptr;
+ append_bias = true;
+ weights_reshaped_info = TensorInfo(compute_weights_reshaped_shape(*weights, true, conv2d_info.num_groups), 1, data_type);
+ }
+ else
+ {
+ weights_reshaped_info = TensorInfo(compute_weights_reshaped_shape(*weights, false, conv2d_info.num_groups), 1, data_type);
+ }
+
+ weights_to_use = &weights_reshaped_info;
+
+ if(!skip_im2col)
+ {
+ const Size2D kernel_dims(kernel_width, kernel_height);
+
+ // Output tensor auto initialization if not yet initialized
+ TensorShape expected_output_shape = compute_im2col_conv_shape(src, kernel_dims, conv2d_info.conv_info, append_bias, conv2d_info.dilation, conv2d_info.num_groups == 1, conv2d_info.num_groups);
+
+ auto_init_if_empty(im2col_reshaped_info, src->clone()->set_tensor_shape(expected_output_shape));
+
+ ARM_COMPUTE_RETURN_ON_ERROR(opencl::kernels::ClIm2ColKernel::validate(src, &im2col_reshaped_info, kernel_dims, conv2d_info.conv_info, append_bias, conv2d_info.dilation, conv2d_info.num_groups));
+ gemm_input_to_use = &im2col_reshaped_info;
+ }
+
+ // Create GEMM output tensor
+ if(!skip_col2im)
+ {
+ TensorShape shape_gemm;
+
+ shape_gemm = gemm_input_to_use->tensor_shape();
+ shape_gemm.set(0, mat_weights_cols);
+ shape_gemm.set(1, conv_w * conv_h);
+
+ info_gemm = TensorInfo(shape_gemm, 1, data_type);
+ info_gemm.set_quantization_info(dst->quantization_info()).set_data_layout(src->data_layout());
+ gemm_output_to_use = &info_gemm;
+ }
+
+ GEMMLowpOutputStageInfo gemmlowp_output_stage;
+ gemmlowp_output_stage.type = GEMMLowpOutputStageType::QUANTIZE_DOWN_FIXEDPOINT;
+ gemmlowp_output_stage.gemmlowp_offset = 0;
+ gemmlowp_output_stage.is_quantized_per_channel = is_quantized_per_channel;
+
+ if(is_quantized)
+ {
+ const UniformQuantizationInfo iq_info = src->quantization_info().uniform();
+ const UniformQuantizationInfo oq_info = dst->quantization_info().uniform();
+ const auto output_quant_info = (dst->total_size() == 0) ? iq_info : oq_info;
+ const unsigned int num_filters = (is_quantized_per_channel) ? num_kernels : 1;
+
+ gemmlowp_output_stage.gemmlowp_multipliers.resize(num_filters);
+ gemmlowp_output_stage.gemmlowp_shifts.resize(num_filters);
+ quantization::compute_quantized_multipliers_and_shifts(src, weights, dst,
+ gemmlowp_output_stage.gemmlowp_multipliers.data(),
+ gemmlowp_output_stage.gemmlowp_shifts.data());
+ gemmlowp_output_stage.gemmlowp_multiplier = gemmlowp_output_stage.gemmlowp_multipliers[0];
+ gemmlowp_output_stage.gemmlowp_shift = gemmlowp_output_stage.gemmlowp_shifts[0];
+
+ int min_activation = 0;
+ int max_activation = 0;
+
+ const std::set<ActivationLayerInfo::ActivationFunction> supported_acts = { ActivationLayerInfo::ActivationFunction::RELU,
+ ActivationLayerInfo::ActivationFunction::BOUNDED_RELU,
+ ActivationLayerInfo::ActivationFunction::LU_BOUNDED_RELU
+ };
+
+ if(conv2d_info.act_info.enabled())
+ {
+ if(supported_acts.count(conv2d_info.act_info.activation()) != 0)
+ {
+ std::tie(min_activation, max_activation) = get_quantized_activation_min_max(conv2d_info.act_info, data_type, output_quant_info);
+ }
+ else
+ {
+ fuse_activation = false;
+ }
+ }
+
+ // Set the GEMMLowp output stage info
+ gemmlowp_output_stage.gemmlowp_offset = output_quant_info.offset;
+ gemmlowp_output_stage.gemmlowp_min_bound = min_activation;
+ gemmlowp_output_stage.gemmlowp_max_bound = max_activation;
+ }
+
+ // In case of NHWC, we need to run GEMM3D (gemm_3d_depth != 0) in order to avoid reshaping the output matrix
+ const unsigned int gemm_3d_depth = (data_layout == DataLayout::NHWC) ? conv_h : 0;
+
+ ARM_COMPUTE_RETURN_ON_ERROR(validate_mm(gemm_input_to_use, weights_to_use, biases_to_use, gemm_output_to_use, gemmlowp_output_stage, gemm_3d_depth, skip_im2col, conv2d_info.act_info));
+
+ // Validate Col2Im
+ if(!skip_col2im)
+ {
+ ARM_COMPUTE_RETURN_ON_ERROR(kernels::ClCol2ImKernel::validate(gemm_output_to_use, dst, Size2D(conv_w, conv_h), conv2d_info.num_groups));
+ }
+
+ //Validate Activation Layer
+ if(!fuse_activation)
+ {
+ ARM_COMPUTE_RETURN_ON_ERROR(kernels::ClActivationKernel::validate(dst, nullptr, conv2d_info.act_info));
+ }
+
+ return Status{};
+}
+
+void ClGemmConvolution::run(ITensorPack &tensors)
+{
+ prepare(tensors);
+
+ auto src = tensors.get_const_tensor(ACL_SRC_0);
+ auto biases = tensors.get_const_tensor(ACL_SRC_2);
+ auto dst = tensors.get_tensor(ACL_DST);
+ auto gemm_input_to_use = src;
+ auto gemm_output_to_use = dst;
+
+ CLAuxTensorHandler im2col_output(offset_int_vec(Im2ColOutput), _im2col_output, tensors, false);
+ CLAuxTensorHandler gemm_output(offset_int_vec(GemmOutput), _gemm_output, tensors, false);
+ CLAuxTensorHandler weights_reshaped(offset_int_vec(WeightsReshaped), _weights_reshaped, tensors, false);
+
+ // Run im2col
+ if(!_skip_im2col)
+ {
+ ITensorPack pack =
+ {
+ { TensorType::ACL_SRC, src },
+ { TensorType::ACL_DST, im2col_output.get() }
+ };
+ CLScheduler::get().enqueue_op(*_im2col_kernel, pack, false);
+ gemm_input_to_use = im2col_output.get();
+ }
+ if(!_skip_col2im)
+ {
+ gemm_output_to_use = gemm_output.get();
+ }
+ ITensorPack pack_mm = tensors;
+ pack_mm.add_const_tensor(TensorType::ACL_SRC_0, gemm_input_to_use);
+ pack_mm.add_const_tensor(TensorType::ACL_SRC_1, weights_reshaped.get());
+ if(!_append_bias)
+ {
+ pack_mm.add_const_tensor(TensorType::ACL_SRC_2, biases);
+ }
+ pack_mm.add_tensor(TensorType::ACL_DST, gemm_output_to_use);
+ // Runs ClGemm or ClGemmLowpMatrixMultiplyCore functions
+ if(_is_quantized)
+ {
+ // Run gemmlowp
+ _mm_gemmlowp->run(pack_mm);
+ }
+ else
+ {
+ // Run gemm
+ _mm_gemm->run(pack_mm);
+ }
+
+ // Reshape output matrix
+ if(!_skip_col2im)
+ {
+ ITensorPack pack =
+ {
+ { TensorType::ACL_SRC, gemm_output_to_use },
+ { TensorType::ACL_DST, dst }
+ };
+ CLScheduler::get().enqueue_op(*_col2im_kernel.get(), pack, false);
+ }
+
+ //Run Activation Layer if we cannot fuse in GEMM
+ if(!_fuse_activation)
+ {
+ ITensorPack pack =
+ {
+ { TensorType::ACL_SRC, dst },
+ { TensorType::ACL_DST, dst }
+ };
+ CLScheduler::get().enqueue_op(*_activation_kernel.get(), pack, false);
+ }
+}
+
+void ClGemmConvolution::prepare(ITensorPack &tensors)
+{
+ if(!_is_prepared)
+ {
+ // Run weights reshaping and mark original weights tensor as unused
+ ICLTensor *weights_reshaped_p = utils::cast::polymorphic_downcast<ICLTensor *>(tensors.get_tensor(offset_int_vec(WeightsReshaped)));
+ CLAuxTensorHandler weights_reshaped(_weights_reshaped, *weights_reshaped_p);
+ auto weights = tensors.get_const_tensor(TensorType::ACL_SRC_1);
+ ITensorPack pack =
+ {
+ { TensorType::ACL_SRC, weights },
+ { TensorType::ACL_DST, weights_reshaped.get() }
+ };
+
+ if(_append_bias)
+ {
+ const auto biases = tensors.get_const_tensor(TensorType::ACL_SRC_2);
+ pack.add_const_tensor(TensorType::ACL_BIAS, biases);
+ }
+ CLScheduler::get().enqueue_op(*_weights_reshape_kernel.get(), pack, true);
+ tensors.add_const_tensor(TensorType::ACL_SRC_1, weights_reshaped.get());
+
+ // Prepare GEMM
+ _is_quantized ? _mm_gemmlowp->prepare(tensors) : _mm_gemm->prepare(tensors);
+ _is_prepared = true;
+ }
+}
+experimental::MemoryRequirements ClGemmConvolution::workspace() const
+{
+ return _aux_mem;
+}
+} // namespace opencl
+} // namespace arm_compute
diff --git a/src/runtime/gpu/cl/operators/ClGemmConvolution.h b/src/runtime/gpu/cl/operators/ClGemmConvolution.h
new file mode 100644
index 0000000..444516e
--- /dev/null
+++ b/src/runtime/gpu/cl/operators/ClGemmConvolution.h
@@ -0,0 +1,185 @@
+/*
+ * Copyright (c) 2021 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 ARM_COMPUTE_CL_GEMMCONVOLUTION_H
+#define ARM_COMPUTE_CL_GEMMCONVOLUTION_H
+
+#include "arm_compute/core/TensorInfo.h"
+#include "arm_compute/core/Types.h"
+#include "arm_compute/runtime/FunctionDescriptors.h"
+#include "src/core/gpu/cl/ClCompileContext.h"
+#include "src/runtime/gpu/cl/IClOperator.h"
+
+#include <memory>
+
+namespace arm_compute
+{
+namespace opencl
+{
+class ClGemm;
+class ClGemmLowpMatrixMultiplyCore;
+namespace kernels
+{
+class ClIm2ColKernel;
+class ClCol2ImKernel;
+class ClWeightsReshapeKernel;
+class ClActivationKernel;
+} // namespace kernels
+
+/** Basic function to compute the convolution layer. This function calls the following OpenCL kernels/functions:
+ *
+ * -# @ref opencl::kernels::ClIm2ColKernel
+ * -# @ref ClGemm (if the data type is FP32 or FP16)
+ * -# @ref CLGEMMLowpMatrixMultiplyCore (if the data type is QASYMM8/QASYMM8_SIGNED)
+ * -# @ref ClGemmLowpOutputStage with QUANTIZE_DOWN_FIXEDPOINT type of quantization (if the data type is QASYMM8/QASYMM8_SIGNED)
+ * -# @ref opencl::kernels::ClCol2ImKernel (if NCHW data layout)
+ * -# @ref opencl::kernels::ClActivationKernel
+ */
+class ClGemmConvolution : public IClOperator
+{
+public:
+ /** Constructor */
+ ClGemmConvolution();
+ /** Prevent instances of this class from being copied (As this class contains pointers) */
+ ClGemmConvolution(const ClGemmConvolution &) = delete;
+ /** Default move constructor */
+ ClGemmConvolution(ClGemmConvolution &&) = default;
+ /** Prevent instances of this class from being copied (As this class contains pointers) */
+ ClGemmConvolution &operator=(const ClGemmConvolution &) = delete;
+ /** Default move assignment operator */
+ ClGemmConvolution &operator=(ClGemmConvolution &&) = default;
+ /**Default destructor */
+ ~ClGemmConvolution();
+ /** Set the input and output tensors.
+ *
+ * Valid data layouts:
+ * - NHWC
+ * - NCHW
+ *
+ * Valid data type configurations:
+ * |src0 |src1 |src2 |dst |
+ * |:--------------|:------------------|:--------|:--------------|
+ * |F16 |F16 |F16 |F16 |
+ * |F32 |F32 |F32 |F32 |
+ * |QASYMM8 |QASYMM8 |S32 |QASYMM8 |
+ * |QASYMM8 |QSYMM8_PER_CHANNEL |S32 |QASYMM8 |
+ * |QASYMM8_SIGNED |QASYMM8_SIGNED |S32 |QASYMM8_SIGNED |
+ * |QASYMM8_SIGNED |QSYMM8_PER_CHANNEL |S32 |QASYMM8_SIGNED |
+ *
+ * @param[in] compile_context The compile context to be used.
+ * @param[in] src Source tensor info. 3 lower dimensions represent a single input [width, height, IFM],
+ * while every optional dimension from 4 and above represent a batch of inputs.
+ * Data types supported: QASYMM8/QASYMM8_SIGNED/F16/F32.
+ * @param[in] weights Weights tensor info. Weights are 4D tensor with dimensions [kernel_x, kernel_y, IFM, OFM].
+ * Data type supported: Same as @p input or QASYMM8/QSYMM8_PER_CHANNEL when @p input is QASYMM8 or QASYMM8_SIGNED/QSYMM8_PER_CHANNEL when @p input is QASYMM8_SIGNED.
+ * @param[in] biases Biases tensor info. Shared biases supported. Biases are 1D tensor with dimensions [OFM].
+ * Data type supported: Should match @p input data type, except for input of quantized type where biases should be of S32 type.
+ * @param[out] dst Destination tensor info. 3 lower dimensions represent a single output [width, height, OFM], while the rest represent batch of outputs.
+ * Data types supported: Same as @p input.
+ * @param[in] conv2d_info Contains convolution 2d info described in @ref Conv2dInfo.
+ * @param[in] weights_info Specifies if the weights tensor has been reshaped with CLWeightsReshapeKernel. If this is not part of the fully connected layer the weights
+ * tensor has also been transposed with CLGEMMReshapeRHSMatrixKernel. Data type supported: Same as @p input.
+ */
+ void configure(const ClCompileContext &compile_context, ITensorInfo *src, ITensorInfo *weights, ITensorInfo *biases, ITensorInfo *dst, const Conv2dInfo &conv2d_info,
+ const WeightsInfo &weights_info = WeightsInfo());
+ /** Static function to check if given info will lead to a valid configuration
+ *
+ * Similar to ClGemmConvolution::configure()
+ *
+ * @return a status
+ */
+ static Status validate(const ITensorInfo *input, const ITensorInfo *weights, const ITensorInfo *biases, const ITensorInfo *output, const Conv2dInfo &conv2d_info,
+ const WeightsInfo &weights_info = WeightsInfo());
+
+ // Inherited methods overridden:
+ void run(ITensorPack &tensors) override;
+ void prepare(ITensorPack &constants) override;
+ experimental::MemoryRequirements workspace() const override;
+
+private:
+ /** Configures the appropriate matrix multiply routine
+ *
+ * @param[in] compile_context The compile context to be used.
+ * @param[in] src Input tensor info. Data types supported: QASYMM8/QASYMM8_SIGNED/F16/F32.
+ * @param[in] weights Weights tensor info. Data type supported: Same as @p input or QASYMM8/QSYMM8_PER_CHANNEL when @p input is QASYMM8 or
+ * QASYMM8_SIGNED/QSYMM8_PER_CHANNEL when @p input is QASYMM8_SIGNED.
+ * @param[in] biases Biases tensor info. Shared biases supported. Biases are 1D tensor with dimensions [OFM].
+ * Data type supported: Should match @p input data type, except for input of quantized type where biases should be of S32 type.
+ * @param[in, out] dst Output tensor info. Data types supported: same as @p input.
+ * @param[in] gemmlowp_output_stage GEMMLowp output stage info
+ * @param[in] gemm_3d_depth Depth of GEMM 3D
+ * @param[in] act_info Activation to apply after the matrix multiplication
+ */
+ void configure_mm(const CLCompileContext &compile_context, const ITensorInfo *src, ITensorInfo *weights, ITensorInfo *biases, ITensorInfo *dst,
+ const GEMMLowpOutputStageInfo &gemmlowp_output_stage,
+ int gemm_3d_depth, const ActivationLayerInfo &act_info);
+ /** Static function to check if given info will lead to a valid configuration of @ref CLGEMMConvolutionLayer matrix multiply routines
+ *
+ * @param[in] src Input tensor info. Data types supported: QASYMM8/QASYMM8_SIGNED/F16/F32.
+ * @param[in] weights Weights tensor info. Data type supported: Same as @p input or QASYMM8/QSYMM8_PER_CHANNEL when @p input is QASYMM8 or
+ * QASYMM8_SIGNED/QSYMM8_PER_CHANNEL when @p input is QASYMM8_SIGNED.
+ * @param[in] biases Biases tensor info. Shared biases supported. Biases are 1D tensor with dimensions [OFM].
+ * Data type supported: Should match @p input data type, except for input of quantized type where biases should be of S32 type.
+ * @param[in] dst Output tensor info. Data types supported: same as @p input.
+ * @param[in] gemmlowp_output_stage GEMMLowp output stage info
+ * @param[in] gemm_3d_depth Depth of GEMM 3D
+ * @param[in] skip_im2col Flag which specifies if im2col has to be skipped. i.e. 1x1 convolution with NHWC data layout.
+ * @param[in] act_info Activation to apply after the matrix multiplication
+ *
+ * @return a status
+ */
+ static Status validate_mm(const ITensorInfo *src, const ITensorInfo *weights, const ITensorInfo *biases, const ITensorInfo *dst, const GEMMLowpOutputStageInfo &gemmlowp_output_stage,
+ int gemm_3d_depth, bool skip_im2col, const ActivationLayerInfo &act_info);
+
+ enum AuxTensorIdx
+ {
+ // ClGemmLowpMatrixMultiplyCore has up to 7 internal tensors
+ Im2ColOutput = 8,
+ WeightsReshaped,
+ GemmOutput,
+ Count
+ };
+
+ std::unique_ptr<kernels::ClWeightsReshapeKernel> _weights_reshape_kernel;
+ std::unique_ptr<kernels::ClIm2ColKernel> _im2col_kernel;
+ std::unique_ptr<ClGemm> _mm_gemm;
+ std::unique_ptr<ClGemmLowpMatrixMultiplyCore> _mm_gemmlowp;
+ std::unique_ptr<opencl::kernels::ClCol2ImKernel> _col2im_kernel;
+ std::unique_ptr<kernels::ClActivationKernel> _activation_kernel;
+
+ TensorInfo _im2col_output;
+ TensorInfo _weights_reshaped;
+ TensorInfo _gemm_output;
+
+ bool _skip_im2col;
+ bool _skip_col2im;
+ bool _is_quantized;
+ bool _fuse_activation;
+ bool _append_bias;
+ bool _is_prepared;
+
+ experimental::MemoryRequirements _aux_mem;
+};
+} // namespace opencl
+} // namespace arm_compute
+#endif /* ARM_COMPUTE_CL_GEMMCONVOLUTION_H */