COMPMID-1836: Remove CLGEMMTranspose1xWKernel and replace with CLGEMMReshapeRHSMatrixKernel

Change-Id: Ic5a4f32657a155380684dcd4b44fbb608ef40cb4
Reviewed-on: https://review.mlplatform.org/418
Reviewed-by: Gian Marco Iodice <gianmarco.iodice@arm.com>
Tested-by: Arm Jenkins <bsgcomp@arm.com>

3 files changed, 4 insertions, 8 deletions

diff --git a/arm_compute/runtime/CL/functions/CLGEMM.h b/arm_compute/runtime/CL/functions/CLGEMM.h
index 7d47194e56..c4accde23d 100644
--- a/arm_compute/runtime/CL/functions/CLGEMM.h
+++ b/arm_compute/runtime/CL/functions/CLGEMM.h
@@ -30,7 +30,6 @@
 #include "arm_compute/core/CL/kernels/CLGEMMMatrixMultiplyReshapedKernel.h"
 #include "arm_compute/core/CL/kernels/CLGEMMReshapeLHSMatrixKernel.h"
 #include "arm_compute/core/CL/kernels/CLGEMMReshapeRHSMatrixKernel.h"
-#include "arm_compute/core/CL/kernels/CLGEMMTranspose1xWKernel.h"
 #include "arm_compute/runtime/CL/CLMemoryGroup.h"
 #include "arm_compute/runtime/CL/CLTensor.h"
 #include "arm_compute/runtime/IFunction.h"
@@ -44,8 +43,7 @@ class ICLTensor;
  *
  *  -# @ref CLGEMMInterleave4x4Kernel (only if the reshaped GEMM is selected by the heuristic model and the GPU target is NOT Mali-G76)
  *  -# @ref CLGEMMReshapeLHSMatrixKernel (only if the reshaped GEMM is selected by the heuristic model and the GPU target IS Mali-G76)
- *  -# @ref CLGEMMTranspose1xWKernel (only if the reshaped GEMM is selected by the heuristic model and the GPU target is NOT Mali-G76)
- *  -# @ref CLGEMMReshapeRHSMatrixKernel (only if the reshaped GEMM is selected by the heuristic model and the GPU target IS Mali-G76)
+ *  -# @ref CLGEMMReshapeRHSMatrixKernel (only if the reshaped GEMM is selected by the heuristic model)
  *  -# @ref CLGEMMMatrixMultiplyKernel (if GPU target is NOT G76 or if the reshaped GEMM is NOT selected)
  *  -# @ref CLGEMMMatrixMultiplyReshapedKernel (only if the reshaped GEMM is selected by the heuristic model and the GPU target IS Mali-G76)
  *  -# @ref CLGEMMMatrixAdditionKernel (if c != nullptr and beta != 0.0)
@@ -108,7 +106,6 @@ public:
 private:
     CLMemoryGroup                      _memory_group;
     CLGEMMInterleave4x4Kernel          _interleave_kernel; // TODO - COMPMID-1835: Remove this kernel and use CLGEMMReshapeLHSMatrixKernel
-    CLGEMMTranspose1xWKernel           _transpose_kernel;  // TODO - COMPMID-1836: Remove this kernel and use CLGEMMReshapeRHSMatrixKernel
     CLGEMMMatrixMultiplyKernel         _mm_kernel;
     CLGEMMMatrixAdditionKernel         _ma_kernel;
     CLGEMMReshapeLHSMatrixKernel       _reshape_lhs_kernel;
diff --git a/arm_compute/runtime/CL/functions/CLGEMMConvolutionLayer.h b/arm_compute/runtime/CL/functions/CLGEMMConvolutionLayer.h
index 1468b156eb..d7694a8328 100644
--- a/arm_compute/runtime/CL/functions/CLGEMMConvolutionLayer.h
+++ b/arm_compute/runtime/CL/functions/CLGEMMConvolutionLayer.h
@@ -30,7 +30,6 @@
 #include "arm_compute/core/CL/kernels/CLElementwiseOperationKernel.h"
 #include "arm_compute/core/CL/kernels/CLGEMMInterleave4x4Kernel.h"
 #include "arm_compute/core/CL/kernels/CLGEMMMatrixMultiplyKernel.h"
-#include "arm_compute/core/CL/kernels/CLGEMMTranspose1xWKernel.h"
 #include "arm_compute/core/CL/kernels/CLIm2ColKernel.h"
 #include "arm_compute/core/CL/kernels/CLWeightsReshapeKernel.h"
 #include "arm_compute/core/Types.h"
diff --git a/arm_compute/runtime/CL/functions/CLGEMMLowpMatrixMultiplyCore.h b/arm_compute/runtime/CL/functions/CLGEMMLowpMatrixMultiplyCore.h
index 82f307a773..141354e723 100644
--- a/arm_compute/runtime/CL/functions/CLGEMMLowpMatrixMultiplyCore.h
+++ b/arm_compute/runtime/CL/functions/CLGEMMLowpMatrixMultiplyCore.h
@@ -29,7 +29,7 @@
 #include "arm_compute/core/CL/kernels/CLGEMMLowpOffsetContributionKernel.h"
 #include "arm_compute/core/CL/kernels/CLGEMMLowpOffsetContributionOutputStageKernel.h"
 #include "arm_compute/core/CL/kernels/CLGEMMLowpReductionKernel.h"
-#include "arm_compute/core/CL/kernels/CLGEMMTranspose1xWKernel.h"
+#include "arm_compute/core/CL/kernels/CLGEMMReshapeRHSMatrixKernel.h"
 #include "arm_compute/runtime/CL/CLMemoryGroup.h"
 #include "arm_compute/runtime/CL/CLTensor.h"
 #include "arm_compute/runtime/IFunction.h"
@@ -42,7 +42,7 @@ class ICLTensor;
 /** Basic function to execute GEMMLowpMatrixMultiplyCore on OpenCL. This function calls the following OpenCL kernels:
  *
  *  -# @ref CLGEMMInterleave4x4Kernel  (if the output tensor is a matrix)
- *  -# @ref CLGEMMTranspose1xWKernel  (if the output tensor is a matrix)
+ *  -# @ref CLGEMMReshapeRHSMatrixKernel  (if the output tensor is a matrix)
  *  -# @ref CLGEMMLowpMatrixMultiplyKernel
  *  -# @ref CLGEMMLowpMatrixAReductionKernel (if the offset of matrix B is not 0)
  *  -# @ref CLGEMMLowpMatrixBReductionKernel (if the offset of matrix A is not 0)
@@ -102,7 +102,7 @@ private:
     CLMemoryGroup                                 _memory_group;
     CLGEMMLowpMatrixMultiplyKernel                _mm_kernel;
     CLGEMMInterleave4x4Kernel                     _mtx_a_reshape_kernel;
-    CLGEMMTranspose1xWKernel                      _mtx_b_reshape_kernel;
+    CLGEMMReshapeRHSMatrixKernel                  _mtx_b_reshape_kernel;
     CLGEMMLowpMatrixAReductionKernel              _mtx_a_reduction_kernel;
     CLGEMMLowpMatrixBReductionKernel              _mtx_b_reduction_kernel;
     CLGEMMLowpOffsetContributionKernel            _offset_contribution_kernel;