From edfa9f463bed084f8b0953557202b2a1e56da817 Mon Sep 17 00:00:00 2001
From: Gian Marco Iodice <gianmarco.iodice@arm.com>
Date: Tue, 15 Aug 2017 11:45:22 +0100
Subject: COMPMID-477 - Optimized batched case in CLConvolutionLayer

Change-Id: I4ef18f49f1da0cb816aaa0762466b940792c15ed
Reviewed-on: http://mpd-gerrit.cambridge.arm.com/84162
Tested-by: Kaizen <jeremy.johnson+kaizengerrit@arm.com>
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
---
 src/core/CL/kernels/CLGEMMLowpMatrixMultiplyKernel.cpp | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

(limited to 'src/core/CL/kernels/CLGEMMLowpMatrixMultiplyKernel.cpp')

diff --git a/src/core/CL/kernels/CLGEMMLowpMatrixMultiplyKernel.cpp b/src/core/CL/kernels/CLGEMMLowpMatrixMultiplyKernel.cpp
index ce68c1f9cd..ef572cfc7e 100644
--- a/src/core/CL/kernels/CLGEMMLowpMatrixMultiplyKernel.cpp
+++ b/src/core/CL/kernels/CLGEMMLowpMatrixMultiplyKernel.cpp
@@ -64,8 +64,8 @@ void CLGEMMLowpMatrixMultiplyKernel::configure(const ICLTensor *input0, const IC
     _output = output;
 
     // Create kernel and set static arguments
-    std::set<std::string> build_opts = { ("-DWIDTH_MATRIX_B=" + support::cpp11::to_string(input1->info()->dimension(0))) };
-    _kernel                          = static_cast<cl::Kernel>(CLKernelLibrary::get().create_kernel("gemm_mm_u8", build_opts));
+    std::set<std::string> build_opts = { ("-DCOLS_B=" + support::cpp11::to_string(input1->info()->dimension(0))) };
+    _kernel                          = static_cast<cl::Kernel>(CLKernelLibrary::get().create_kernel("gemm_mm_interleaved_transposed_u8", build_opts));
     unsigned int idx                 = 3 * num_arguments_per_2D_tensor(); //Skip the input and output parameters
     _kernel.setArg<int32_t>(idx++, a_offset);
     _kernel.setArg<int32_t>(idx++, b_offset);
-- 
cgit v1.2.1