commit d11de9861e6c32fa389f503e037098f50ffed156
author Gian Marco Iodice <gianmarco.iodice@arm.com> Mon Sep 05 15:35:35 2022 +0100
committer Gian Marco Iodice <gianmarco.iodice@arm.com> Fri Sep 09 08:37:36 2022 +0000
tree edbb84b8a5ecdbc7217c7277321e7c61cd01ea5c
parent e4e3b2ead5b6720af8039f3c9ac15ea6b51b915f

Add a macro guard in all OpenCL kernels in gemmlowp.cl

Resolves COMPMID-5498

Change-Id: I474f3f963257014255d082aab0ccbe3efe5aa067
Signed-off-by: Gian Marco Iodice <gianmarco.iodice@arm.com>
Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/8222
Tested-by: Arm Jenkins <bsgcomp@arm.com>
Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
Comments-Addressed: Ramy Elgammal <ramy.elgammal@arm.com>
Reviewed-by: Ramy Elgammal <ramy.elgammal@arm.com>
Reviewed-by: Gunes Bayir <gunes.bayir@arm.com>
Benchmark: Arm Jenkins <bsgcomp@arm.com>

src/gpu/cl/kernels/ClGemmLowpQuantizeDownInt32ScaleKernel.cpp

diff --git a/src/gpu/cl/kernels/ClGemmLowpQuantizeDownInt32ScaleKernel.cpp b/src/gpu/cl/kernels/ClGemmLowpQuantizeDownInt32ScaleKernel.cpp
index 5469ea9..bad9d96 100644
--- a/src/gpu/cl/kernels/ClGemmLowpQuantizeDownInt32ScaleKernel.cpp
+++ b/src/gpu/cl/kernels/ClGemmLowpQuantizeDownInt32ScaleKernel.cpp
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2020-2021 Arm Limited.
+ * Copyright (c) 2020-2022 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -112,8 +112,13 @@
     build_opts.add_option("-DOUTPUT_DATA_TYPE=" + get_cl_type_from_data_type(dst->data_type()));
     build_opts.add_option_if(bias != nullptr, "-DADD_BIAS");
 
+    const std::string kernel_name = "gemmlowp_output_stage_quantize_down";
+
+    // A macro guard to compile ONLY the kernel of interest
+    build_opts.add_option("-D" + upper_string(kernel_name));
+
     // Create kernel
-    _kernel = create_kernel(compile_context, "gemmlowp_output_stage_quantize_down", build_opts.options());
+    _kernel = create_kernel(compile_context, kernel_name, build_opts.options());
 
     // Configure kernel window
     Window win = calculate_max_window(*src, Steps(num_elems_processed_per_iteration));