From 64f1a908841913049ccc0eb941b5b213290d7bf7 Mon Sep 17 00:00:00 2001
From: Georgios Pinitas <georgios.pinitas@arm.com>
Date: Tue, 18 Sep 2018 13:42:51 +0100
Subject: COMPMID-1591: Fix NEPoolingLayer for NHWC

Restore window step across width to 4 for FP32 instead of the whole row
as the kernel code was inconsistent with this decision.

Change-Id: I7c4dcdf960b8cbc970a36fa1df39df2c6f000c86
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/148908
Reviewed-by: Michalis Spyrou <michalis.spyrou@arm.com>
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
Tested-by: bsgcomp <bsgcomp@arm.com>
---
 src/core/NEON/kernels/NEPoolingLayerKernel.cpp | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'src/core/NEON/kernels/NEPoolingLayerKernel.cpp')

diff --git a/src/core/NEON/kernels/NEPoolingLayerKernel.cpp b/src/core/NEON/kernels/NEPoolingLayerKernel.cpp
index a8e3be28fe..1fa8f47332 100644
--- a/src/core/NEON/kernels/NEPoolingLayerKernel.cpp
+++ b/src/core/NEON/kernels/NEPoolingLayerKernel.cpp
@@ -242,7 +242,7 @@ std::pair<Status, Window> validate_and_configure_window(ITensorInfo *input, ITen
             case DataType::F32:
                 if(is_nhwc)
                 {
-                    num_elems_processed_per_iteration = std::max(4, ceil_to_multiple<int>(input->dimension(0), 2));
+                    num_elems_processed_per_iteration = 4;
                     break;
                 }
                 switch(pool_size_x)
-- 
cgit v1.2.1