sw/kelvin: add max_pooling int16 kernel for Kelvin
Change-Id: I4fb4b0d2fd71f5a12b318b32dfee51bd570acd51
diff --git a/tflm/opt/BUILD b/tflm/opt/BUILD
index cdaf83d..9957e7a 100644
--- a/tflm/opt/BUILD
+++ b/tflm/opt/BUILD
@@ -35,6 +35,7 @@
"leaky_relu_s16.cc",
"leaky_relu_s8.cc",
"logistic_s8.cc",
+ "max_pool_s16.cc",
"max_pool_s8.cc",
"memcpy.cc",
"resize_nearest_neighbor_s8.cc",
diff --git a/tflm/opt/max_pool_s16.cc b/tflm/opt/max_pool_s16.cc
new file mode 100644
index 0000000..4e3aa46
--- /dev/null
+++ b/tflm/opt/max_pool_s16.cc
@@ -0,0 +1,100 @@
+/*
+ * Copyright 2024 Google LLC
+ *
+ * Licensed under the Apache License, Version 2.0 (the "License");
+ * you may not use this file except in compliance with the License.
+ * You may obtain a copy of the License at
+ *
+ * http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+#include "crt/kelvin.h"
+#include "tensorflow/lite/kernels/internal/common.h"
+#include "tensorflow/lite/kernels/internal/runtime_shape.h"
+#include "tensorflow/lite/kernels/internal/types.h"
+
+namespace kelvin::opt {
+void MaxPoolS16(const tflite::PoolParams ¶ms,
+ const tflite::RuntimeShape &input_shape,
+ const int16_t *input_data,
+ const tflite::RuntimeShape &output_shape,
+ int16_t *output_data) {
+ const int batches = MatchingDim(input_shape, 0, output_shape, 0);
+ const int depth = MatchingDim(input_shape, 3, output_shape, 3);
+ const int input_height = input_shape.Dims(1);
+ const int input_width = input_shape.Dims(2);
+ const int output_height = output_shape.Dims(1);
+ const int output_width = output_shape.Dims(2);
+ const int stride_height = params.stride_height;
+ const int stride_width = params.stride_width;
+ for (int batch = 0; batch < batches; ++batch) {
+ for (int out_y = 0; out_y < output_height; ++out_y) {
+ for (int out_x = 0; out_x < output_width; ++out_x) {
+ const int in_x_origin =
+ (out_x * stride_width) - params.padding_values.width;
+ const int in_y_origin =
+ (out_y * stride_height) - params.padding_values.height;
+
+ // Compute the boundaries of the filter region clamped so as to
+ // ensure that the filter window fits in the input array.
+ const int filter_x_start = std::max(0, -in_x_origin);
+ const int filter_x_end =
+ std::min(params.filter_width, input_width - in_x_origin);
+ const int filter_y_start = std::max(0, -in_y_origin);
+ const int filter_y_end =
+ std::min(params.filter_height, input_height - in_y_origin);
+
+ int channel = 0;
+ for (; channel + 16 <= depth; channel += 16) {
+ vdup_h_x(v0, params.quantized_activation_min);
+ for (int filter_y = filter_y_start; filter_y < filter_y_end;
+ ++filter_y) {
+ for (int filter_x = filter_x_start; filter_x < filter_x_end;
+ ++filter_x) {
+ const int in_x = in_x_origin + filter_x;
+ const int in_y = in_y_origin + filter_y;
+ const int16_t *local_input =
+ input_data + Offset(input_shape, batch, in_y, in_x, channel);
+ vld_h_x(v1, local_input);
+ vmax_h_vv(v0, v0, v1);
+ }
+ }
+ vmin_h_vx(v0, v0, params.quantized_activation_max);
+ int16_t *local_output =
+ output_data + Offset(output_shape, batch, out_y, out_x, channel);
+ vst_h_x(v0, local_output);
+ }
+
+ if (channel == depth) {
+ continue;
+ }
+ int remaining_channels = depth - channel;
+ vdup_h_x(v0, params.quantized_activation_min);
+ for (int filter_y = filter_y_start; filter_y < filter_y_end;
+ ++filter_y) {
+ for (int filter_x = filter_x_start; filter_x < filter_x_end;
+ ++filter_x) {
+ const int in_x = in_x_origin + filter_x;
+ const int in_y = in_y_origin + filter_y;
+ const int16_t *local_input =
+ input_data + Offset(input_shape, batch, in_y, in_x, depth - 1);
+ vld_h_l_xx(v1, local_input, remaining_channels);
+ vmax_h_vv(v0, v0, v1);
+ }
+ }
+ vmin_h_vx(v0, v0, params.quantized_activation_max);
+ int16_t *local_output =
+ output_data + Offset(output_shape, batch, out_y, out_x, depth - 1);
+ vst_h_l_xx(v0, local_output, remaining_channels);
+ }
+ }
+ }
+}
+
+} // namespace kelvin::opt
diff --git a/tflm/opt/opt.h b/tflm/opt/opt.h
index 85f5019..053fbd6 100644
--- a/tflm/opt/opt.h
+++ b/tflm/opt/opt.h
@@ -97,6 +97,10 @@
const tflite::RuntimeShape& input_shape,
const int8_t* input_data,
const tflite::RuntimeShape& output_shape, int8_t* output_data);
+void MaxPoolS16(const tflite::PoolParams& params,
+ const tflite::RuntimeShape& input_shape,
+ const int16_t* input_data,
+ const tflite::RuntimeShape& output_shape, int16_t* output_data);
void MulS8(const tflite::ArithmeticParams& params,
const tflite::RuntimeShape& input1_shape, const int8_t* input1_data,
const tflite::RuntimeShape& input2_shape, const int8_t* input2_data,
