From b2d65b447f764928499f3c86999fff23d6e68ba2 Mon Sep 17 00:00:00 2001
From: Hariharan Seshadri <shariharan91@gmail.com>
Date: Wed, 11 Dec 2019 11:15:12 -0800
Subject: [PATCH] Remove unnecessary parameter in some places in GatherElements
 implementation (#2612)

* Remove unnecessary parameter in some places

* Update

* Update
---
 onnxruntime/core/providers/cuda/tensor/gather_elements.cc      | 3 ---
 onnxruntime/core/providers/cuda/tensor/gather_elements_impl.cu | 3 ---
 onnxruntime/core/providers/cuda/tensor/gather_elements_impl.h  | 1 -
 3 files changed, 7 deletions(-)

diff --git a/onnxruntime/core/providers/cuda/tensor/gather_elements.cc b/onnxruntime/core/providers/cuda/tensor/gather_elements.cc
index 4fa8185f2e..1a2edb43b9 100644
--- a/onnxruntime/core/providers/cuda/tensor/gather_elements.cc
+++ b/onnxruntime/core/providers/cuda/tensor/gather_elements.cc
@@ -27,7 +27,6 @@ Status GatherElements::ComputeInternal(OpKernelContext* context) const {
   const auto& input_shape = input_tensor->Shape();
   const auto& input_dims = input_shape.GetDims();
   const int64_t input_rank = static_cast<int64_t>(input_dims.size());
-  const int64_t input_size = input_shape.Size();
 
   // Process indices tensor
   const auto* indices_tensor = context->Input<Tensor>(1);
@@ -67,7 +66,6 @@ Status GatherElements::ComputeInternal(OpKernelContext* context) const {
     GatherElementsImpl<int32_t>(
         input_rank,
         input_tensor->DataRaw(),
-        input_size,
         input_dims[axis],
         gpu_input_strides.GpuPtr(),
         indices_data,
@@ -82,7 +80,6 @@ Status GatherElements::ComputeInternal(OpKernelContext* context) const {
     GatherElementsImpl<int64_t>(
         input_rank,
         input_tensor->DataRaw(),
-        input_size,
         input_dims[axis],
         gpu_input_strides.GpuPtr(),
         indices_data,
diff --git a/onnxruntime/core/providers/cuda/tensor/gather_elements_impl.cu b/onnxruntime/core/providers/cuda/tensor/gather_elements_impl.cu
index a0ec64735e..b9d989c413 100644
--- a/onnxruntime/core/providers/cuda/tensor/gather_elements_impl.cu
+++ b/onnxruntime/core/providers/cuda/tensor/gather_elements_impl.cu
@@ -42,7 +42,6 @@ template <typename Tin>
 void GatherElementsImpl(
     const int64_t rank,
     const void* input_data,
-    const int64_t input_size,
     const int64_t input_dim_along_axis,
     const int64_t* input_strides,
     const Tin* indices_data,
@@ -95,7 +94,6 @@ void GatherElementsImpl(
 template void GatherElementsImpl<int32_t>(
     const int64_t rank,
     const void* input_data,
-    const int64_t input_size,
     const int64_t input_dim_along_axis,
     const int64_t* input_strides,
     const int32_t* indices_data,
@@ -108,7 +106,6 @@ template void GatherElementsImpl<int32_t>(
 template void GatherElementsImpl<int64_t>(
     const int64_t rank,
     const void* input_data,
-    const int64_t input_size,
     const int64_t input_dim_along_axis,
     const int64_t* input_strides,
     const int64_t* indices_data,
diff --git a/onnxruntime/core/providers/cuda/tensor/gather_elements_impl.h b/onnxruntime/core/providers/cuda/tensor/gather_elements_impl.h
index aba9ef044e..e856f0f259 100644
--- a/onnxruntime/core/providers/cuda/tensor/gather_elements_impl.h
+++ b/onnxruntime/core/providers/cuda/tensor/gather_elements_impl.h
@@ -13,7 +13,6 @@ template <typename Tin>
 void GatherElementsImpl(
     const int64_t rank,  // both inputs have same rank and this is validated in the main Compute
     const void* input_data,
-    const int64_t input_size,
     const int64_t input_dim_along_axis,
     const int64_t* input_strides,
     const Tin* indices_data,