COMPMID-2336 Fix valgrind error for BatchNormalizationLayer on NEON with NHWC layout
Change-Id: I9ed2d0647ae3c33bce6290acfdac356ffffcb709
Signed-off-by: Giorgio Arena <giorgio.arena@arm.com>
Reviewed-on: https://review.mlplatform.org/c/1697
Reviewed-by: Gian Marco Iodice <gianmarco.iodice@arm.com>
Tested-by: Arm Jenkins <bsgcomp@arm.com>
diff --git a/src/core/CL/kernels/CLBatchNormalizationLayerKernel.cpp b/src/core/CL/kernels/CLBatchNormalizationLayerKernel.cpp
index f9b9753..ae646a7 100644
--- a/src/core/CL/kernels/CLBatchNormalizationLayerKernel.cpp
+++ b/src/core/CL/kernels/CLBatchNormalizationLayerKernel.cpp
@@ -107,6 +107,7 @@
window_changed = update_window_and_padding(win, input_access);
}
+ // Mean, var, gamma and beta get parallelized for the NHWC case as they follow the channel dimension, which is along the first axis
if(input->data_layout() == DataLayout::NHWC)
{
AccessWindowHorizontal mean_access(mean, 0, num_elems_processed_per_iteration);