From cb0010b02281245c66d5c996fa9ef8b22f036a2d Mon Sep 17 00:00:00 2001
From: Vidhya Sudhan Loganathan <vidhyasudhan.loganathan@arm.com>
Date: Fri, 11 May 2018 16:23:53 +0100
Subject: COMPMID-1102 : Enable the use of 4x4 tile sizes in neon
 implementation of winograd conv.

Change-Id: Ibd2f2c6680b647a066255ea77d4a2a172ef76aa3
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/130418
Reviewed-by: Gian Marco Iodice <gianmarco.iodice@arm.com>
Tested-by: Jenkins <bsgcomp@arm.com>
---
 src/core/NEON/kernels/NEWinogradConvolutionLayerKernel.cpp | 10 ++++++++--
 1 file changed, 8 insertions(+), 2 deletions(-)

(limited to 'src/core/NEON/kernels')

diff --git a/src/core/NEON/kernels/NEWinogradConvolutionLayerKernel.cpp b/src/core/NEON/kernels/NEWinogradConvolutionLayerKernel.cpp
index 7e82dc4ecd..672684d14f 100644
--- a/src/core/NEON/kernels/NEWinogradConvolutionLayerKernel.cpp
+++ b/src/core/NEON/kernels/NEWinogradConvolutionLayerKernel.cpp
@@ -173,8 +173,10 @@ Status validate_arguments_winograd_output_trans(const ITensorInfo *input, const
     const Size2D         kernel_dims = winograd_info.kernel_size;
 
     // Number of tiles along the X and Y direction
-    const unsigned int num_tiles_x = std::ceil((winograd_info.input_dimensions.x() - (kernel_dims.width - 1) + conv_info.pad_left() + conv_info.pad_right()) / 2.f);
-    const unsigned int num_tiles_y = std::ceil((winograd_info.input_dimensions.y() - (kernel_dims.height - 1) + conv_info.pad_top() + conv_info.pad_bottom()) / 2.f);
+    const unsigned int num_tiles_x = std::ceil((winograd_info.input_dimensions.x() - (kernel_dims.width - 1) + conv_info.pad_left() + conv_info.pad_right()) / static_cast<float>
+                                               (winograd_info.output_tile_size.width));
+    const unsigned int num_tiles_y = std::ceil((winograd_info.input_dimensions.y() - (kernel_dims.height - 1) + conv_info.pad_top() + conv_info.pad_bottom()) / static_cast<float>
+                                               (winograd_info.output_tile_size.height));
     const Size2D       num_tiles   = Size2D(num_tiles_x, num_tiles_y);
 
     ARM_COMPUTE_RETURN_ERROR_ON_NULLPTR(input);
@@ -301,6 +303,7 @@ Status NEWinogradLayerBatchedGEMMKernel<TIn, TOut, OutputTileRows, OutputTileCol
 }
 
 template class NEWinogradLayerBatchedGEMMKernel<float, float, 2, 2, 3, 3>;
+template class NEWinogradLayerBatchedGEMMKernel<float, float, 4, 4, 3, 3>;
 template class NEWinogradLayerBatchedGEMMKernel<float, float, 2, 2, 5, 5>;
 
 // Weights transform
@@ -369,6 +372,7 @@ Status NEWinogradLayerTransformWeightsKernel<T, OutputTileRows, OutputTileCols,
 }
 
 template class NEWinogradLayerTransformWeightsKernel<float, 2, 2, 3, 3>;
+template class NEWinogradLayerTransformWeightsKernel<float, 4, 4, 3, 3>;
 template class NEWinogradLayerTransformWeightsKernel<float, 2, 2, 5, 5>;
 
 // Input transform
@@ -442,6 +446,7 @@ Status NEWinogradLayerTransformInputKernel<T, OutputTileRows, OutputTileCols, Ke
 }
 
 template class NEWinogradLayerTransformInputKernel<float, 2, 2, 3, 3>;
+template class NEWinogradLayerTransformInputKernel<float, 4, 4, 3, 3>;
 template class NEWinogradLayerTransformInputKernel<float, 2, 2, 5, 5>;
 
 // Output transform
@@ -544,6 +549,7 @@ Status NEWinogradLayerTransformOutputKernel<T, OutputTileRows, OutputTileCols, K
 }
 
 template class NEWinogradLayerTransformOutputKernel<float, 2, 2, 3, 3>;
+template class NEWinogradLayerTransformOutputKernel<float, 4, 4, 3, 3>;
 template class NEWinogradLayerTransformOutputKernel<float, 2, 2, 5, 5>;
 
 } // namespace arm_compute
-- 
cgit v1.2.1