Decouple NEL2NormalizeLayerKernel
Resolves: COMPMID-4615
Signed-off-by: Yair Schwarzbaum <yair.schwarzbaum@arm.com>
Change-Id: Iadbfb3e45831a5072962b5b9f61e8ae2e674ccc4
Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/7016
Reviewed-by: Giorgio Arena <giorgio.arena@arm.com>
Tested-by: Arm Jenkins <bsgcomp@arm.com>
Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
diff --git a/src/core/NEON/kernels/NEL2NormalizeLayerKernel.cpp b/src/core/NEON/kernels/NEL2NormalizeLayerKernel.cpp
index 9bda82d..8ab0288 100644
--- a/src/core/NEON/kernels/NEL2NormalizeLayerKernel.cpp
+++ b/src/core/NEON/kernels/NEL2NormalizeLayerKernel.cpp
@@ -1,5 +1,5 @@
/*
- * Copyright (c) 2017-2021 Arm Limited.
+ * Copyright (c) 2017-2022 Arm Limited.
*
* SPDX-License-Identifier: MIT
*
@@ -30,11 +30,13 @@
#include "arm_compute/core/Utils.h"
#include "arm_compute/core/Validate.h"
#include "arm_compute/core/Window.h"
+#include "src/common/cpuinfo/CpuIsaInfo.h"
#include "src/core/NEON/NEMath.h"
+#include "src/core/common/Registrars.h"
#include "src/core/helpers/AutoConfiguration.h"
#include "src/core/helpers/WindowHelpers.h"
+#include "src/cpu/kernels/l2normlayer/list.h"
-#include "src/core/NEON/wrapper/wrapper.h"
#include <arm_neon.h>
#include <cmath>
@@ -44,90 +46,64 @@
{
constexpr int max_input_tensor_dim = 3;
-template <typename T, int S>
-void l2_normalize_X(const ITensor *in, const ITensor *sum, ITensor *out, float epsilon, const Window &window)
+struct L2NormalizeLayerSelectorData
{
- using ExactTagType = typename wrapper::traits::neon_vector<T, S>::tag_type;
+ DataType dt;
+ unsigned int actual_axis;
+ cpuinfo::CpuIsaInfo isa;
+};
- const int window_step_x = 16 / data_size_from_type(in->info()->data_type());
- const auto window_start_x = static_cast<int>(window.x().start());
- const auto window_end_x = static_cast<int>(window.x().end());
+using L2NormalizeLayerKernelSelctorPtr = std::add_pointer<bool(const L2NormalizeLayerSelectorData &data)>::type;
- Window win_collapsed = window.collapse_if_possible(window, Window::DimZ);
- win_collapsed.set(Window::DimX, Window::Dimension(0, 1, 1));
+using L2NormalizeLayerPtr = std::add_pointer<void(const ITensor *in, const ITensor *sum, ITensor *out, float epsilon, const Window &window, size_t axis)>::type;
- Iterator input_it(in, win_collapsed);
- Iterator sum_it(sum, win_collapsed);
- Iterator output_it(out, win_collapsed);
-
- execute_window_loop(win_collapsed, [&](const Coordinates &)
- {
- const auto in_ptr = reinterpret_cast<const T *>(input_it.ptr());
- const auto out_ptr = reinterpret_cast<T *>(output_it.ptr());
-
- const T sum_value = *reinterpret_cast<const T *>(sum_it.ptr());
- const T norm_value = static_cast<T>(1.f) / std::sqrt(std::max(sum_value, static_cast<T>(epsilon)));
- const auto vec_norm_value = wrapper::vdup_n(norm_value, ExactTagType{});
-
- // Compute elements over vector steps
- int x = window_start_x;
- for(; x <= (window_end_x - window_step_x); x += window_step_x)
- {
- wrapper::vstore(out_ptr + x, wrapper::vmul(wrapper::vloadq(in_ptr + x), vec_norm_value));
- }
-
- // Compute left-over elements
- for(; x < window_end_x; ++x)
- {
- out_ptr[x] = in_ptr[x] * norm_value;
- }
- },
- input_it, sum_it, output_it);
-}
-
-template <typename T, int S>
-void l2_normalize_YZ(const ITensor *in, const ITensor *sum, ITensor *out, float epsilon, const Window &window, size_t axis)
+struct L2NormalizeLayerKernel
{
- using ExactTagType = typename wrapper::traits::neon_vector<T, S>::tag_type;
+ const char *name;
+ const L2NormalizeLayerKernelSelctorPtr is_selected;
+ L2NormalizeLayerPtr ukernel;
+};
- const int window_step_x = 16 / data_size_from_type(in->info()->data_type());
- const auto window_start_x = static_cast<int>(window.x().start());
- const auto window_end_x = static_cast<int>(window.x().end());
-
- Window win = window;
- win.set(Window::DimX, Window::Dimension(0, 1, 1));
-
- Window window_sum(win);
- window_sum.set(axis, Window::Dimension(0, 0, 0));
-
- Iterator input_it(in, win);
- Iterator sum_it(sum, window_sum);
- Iterator output_it(out, win);
-
- const auto vec_eps = wrapper::vdup_n(static_cast<T>(epsilon), ExactTagType{});
-
- execute_window_loop(win, [&](const Coordinates &)
+static const L2NormalizeLayerKernel available_kernels[] =
+{
{
- const auto in_ptr = reinterpret_cast<const T *>(input_it.ptr());
- const auto sum_ptr = reinterpret_cast<const T *>(sum_it.ptr());
- const auto out_ptr = reinterpret_cast<T *>(output_it.ptr());
-
- // Compute elements over vector steps
- int x = window_start_x;
- for(; x <= (window_end_x - window_step_x); x += window_step_x)
- {
- const auto vec_norm_value = wrapper::vinvsqrt(wrapper::vmax(wrapper::vloadq(sum_ptr + x), vec_eps));
- wrapper::vstore(out_ptr + x, wrapper::vmul(wrapper::vloadq(in_ptr + x), vec_norm_value));
- }
-
- // Compute left-over elements
- for(; x < window_end_x; ++x)
- {
- const T norm_value = static_cast<T>(1.f) / std::sqrt(std::max(sum_ptr[x], static_cast<T>(epsilon)));
- out_ptr[x] = in_ptr[x] * norm_value;
- }
+ "fp32_neon_l2normalize_x",
+ [](const L2NormalizeLayerSelectorData & data) { return data.dt == DataType::F32 && data.actual_axis == Window::DimX; },
+ REGISTER_FP32_NEON(arm_compute::cpu::neon_fp32_l2_normalize_x)
},
- input_it, sum_it, output_it);
+ {
+ "fp32_neon_l2normalize_yz",
+ [](const L2NormalizeLayerSelectorData & data) { return data.dt == DataType::F32 && data.actual_axis != Window::DimX; },
+ REGISTER_FP32_NEON(arm_compute::cpu::neon_fp32_l2_normalize_yz)
+ },
+ {
+ "fp16_neon_l2normalize_x",
+ [](const L2NormalizeLayerSelectorData & data) { return data.dt == DataType::F16 && data.isa.fp16 && data.actual_axis == Window::DimX; },
+ REGISTER_FP16_NEON(arm_compute::cpu::neon_fp16_l2_normalize_x),
+ },
+ {
+ "fp16_neon_l2normalize_yz",
+ [](const L2NormalizeLayerSelectorData & data) { return data.dt == DataType::F16 && data.isa.fp16 && data.actual_axis != Window::DimX; },
+ REGISTER_FP16_NEON(arm_compute::cpu::neon_fp16_l2_normalize_yz),
+ },
+};
+
+/** Micro-kernel selector
+ *
+ * @param[in] data Selection data passed to help pick the appropriate micro-kernel
+ *
+ * @return A matching micro-kernel else nullptr
+ */
+const L2NormalizeLayerKernel *get_implementation(const L2NormalizeLayerSelectorData &data)
+{
+ for(const auto &uk : available_kernels)
+ {
+ if(uk.is_selected(data))
+ {
+ return &uk;
+ }
+ }
+ return nullptr;
}
Status validate_arguments(const ITensorInfo *input, const ITensorInfo *sum, const ITensorInfo *output, int axis, float epsilon)
@@ -212,18 +188,10 @@
ARM_COMPUTE_ERROR("Unsupported normalization axis");
}
- switch(_input->info()->data_type())
- {
- case DataType::F32:
- (_actual_axis == Window::DimX) ? l2_normalize_X<float, 4>(_input, _sum, _output, _epsilon, window) : l2_normalize_YZ<float, 4>(_input, _sum, _output, _epsilon, window, _actual_axis);
- break;
-#ifdef __ARM_FEATURE_FP16_VECTOR_ARITHMETIC
- case DataType::F16:
- (_actual_axis == Window::DimX) ? l2_normalize_X<float16_t, 8>(_input, _sum, _output, _epsilon, window) : l2_normalize_YZ<float16_t, 8>(_input, _sum, _output, _epsilon, window, _actual_axis);
- break;
-#endif // __ARM_FEATURE_FP16_VECTOR_ARITHMETIC
- default:
- ARM_COMPUTE_ERROR("Not implemented");
- }
+ const auto *uk = get_implementation(L2NormalizeLayerSelectorData{ _output->info()->data_type(), _actual_axis, CPUInfo::get().get_isa() });
+ ARM_COMPUTE_ERROR_ON(uk == nullptr);
+ ARM_COMPUTE_ERROR_ON(uk->ukernel == nullptr);
+
+ uk->ukernel(_input, _sum, _output, _epsilon, window, _actual_axis);
}
} // namespace arm_compute