Add Fused Activation to OpenCL MatMul

- Added fused activation to MatMul function interface
- Added fused activation to CL backend
- Includes tests for supported Activation Functions in MatMul

Resolves: [COMPMID-6192]
Signed-off-by: Mohammed Suhail Munshi <MohammedSuhail.Munshi@arm.com>
Change-Id: Ie103212b600b60699eaf6a6394d609e6e1f5aba6
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/c/VisualCompute/ComputeLibrary/+/522465
Comments-Addressed: bsgcomp <bsgcomp@arm.com>
Reviewed-by: Viet-Hoa Do <viet-hoa.do@arm.com>
Tested-by: bsgcomp <bsgcomp@arm.com>
Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/9714
Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
Reviewed-by: Jakub Sujak <jakub.sujak@arm.com>
Tested-by: Arm Jenkins <bsgcomp@arm.com>
Benchmark: Arm Jenkins <bsgcomp@arm.com>

2 files changed, 12 insertions, 11 deletions

diff --git a/src/runtime/CL/functions/CLMatMul.cpp b/src/runtime/CL/functions/CLMatMul.cpp
index ae5a01f679..bef422fca1 100644
--- a/src/runtime/CL/functions/CLMatMul.cpp
+++ b/src/runtime/CL/functions/CLMatMul.cpp
@@ -42,25 +42,26 @@ CLMatMul::CLMatMul()
 
 CLMatMul::~CLMatMul() = default;
 
-void CLMatMul::configure(ICLTensor *lhs, ICLTensor *rhs, ICLTensor *output, const MatMulInfo &matmul_info, const GpuMatMulSettings &settings)
+void CLMatMul::configure(ICLTensor *lhs, ICLTensor *rhs, ICLTensor *output, const MatMulInfo &matmul_info, const GpuMatMulSettings &settings, const ActivationLayerInfo &act_info)
 {
     ARM_COMPUTE_UNUSED(settings);
-    configure(CLKernelLibrary::get().get_compile_context(), lhs, rhs, output, matmul_info);
+    configure(CLKernelLibrary::get().get_compile_context(), lhs, rhs, output, matmul_info, settings, act_info);
 }
 
-void CLMatMul::configure(const CLCompileContext &compile_context, ICLTensor *lhs, ICLTensor *rhs, ICLTensor *output, const MatMulInfo &matmul_info, const GpuMatMulSettings &settings)
+void CLMatMul::configure(const CLCompileContext &compile_context, ICLTensor *lhs, ICLTensor *rhs, ICLTensor *output, const MatMulInfo &matmul_info, const GpuMatMulSettings &settings,
+                         const ActivationLayerInfo &act_info)
 {
     ARM_COMPUTE_ERROR_ON_NULLPTR(lhs, rhs, output);
     ARM_COMPUTE_UNUSED(settings);
 
     _impl->op = std::make_unique<OperatorType>();
-    _impl->op->configure(compile_context, lhs->info(), rhs->info(), output->info(), matmul_info);
+    _impl->op->configure(compile_context, lhs->info(), rhs->info(), output->info(), matmul_info, act_info);
     _impl->run_pack = { { ACL_SRC_0, lhs }, { ACL_SRC_1, rhs }, { ACL_DST, output } };
 }
 
-Status CLMatMul::validate(const ITensorInfo *lhs, const ITensorInfo *rhs, const ITensorInfo *output, const MatMulInfo &matmul_info)
+Status CLMatMul::validate(const ITensorInfo *lhs, const ITensorInfo *rhs, const ITensorInfo *output, const MatMulInfo &matmul_info, const ActivationLayerInfo &act_info)
 {
-    return OperatorType::validate(lhs, rhs, output, matmul_info);
+    return OperatorType::validate(lhs, rhs, output, matmul_info, act_info);
 }
 
 void CLMatMul::run()
diff --git a/src/runtime/NEON/functions/NEMatMul.cpp b/src/runtime/NEON/functions/NEMatMul.cpp
index 0c46516f1e..58640f40ea 100644
--- a/src/runtime/NEON/functions/NEMatMul.cpp
+++ b/src/runtime/NEON/functions/NEMatMul.cpp
@@ -25,9 +25,9 @@
 
 #include "arm_compute/core/Validate.h"
 #include "arm_compute/runtime/MemoryGroup.h"
+#include "arm_compute/runtime/Tensor.h"
 #include "src/core/helpers/MemoryHelpers.h"
 #include "src/cpu/operators/CpuMatMul.h"
-#include "arm_compute/runtime/Tensor.h"
 
 namespace arm_compute
 {
@@ -49,7 +49,7 @@ NEMatMul::NEMatMul()
 
 NEMatMul::~NEMatMul() = default;
 
-void NEMatMul::configure(ITensor *lhs, ITensor *rhs, ITensor *output, const MatMulInfo &info, const CpuMatMulSettings &settings)
+void NEMatMul::configure(ITensor *lhs, ITensor *rhs, ITensor *output, const MatMulInfo &info, const CpuMatMulSettings &settings, const ActivationLayerInfo &act_info)
 {
     _impl->lhs    = lhs;
     _impl->rhs    = rhs;
@@ -57,14 +57,14 @@ void NEMatMul::configure(ITensor *lhs, ITensor *rhs, ITensor *output, const MatM
 
     ARM_COMPUTE_ERROR_ON_NULLPTR(_impl->lhs, _impl->rhs, _impl->output);
     _impl->op = std::make_unique<cpu::CpuMatMul>();
-    _impl->op->configure(lhs->info(), rhs->info(), output->info(), info, settings);
+    _impl->op->configure(lhs->info(), rhs->info(), output->info(), info, settings, act_info);
     _impl->run_pack          = { { ACL_SRC_0, lhs }, { ACL_SRC_1, rhs }, { ACL_DST, output } };
     _impl->workspace_tensors = manage_workspace<Tensor>(_impl->op->workspace(), _impl->memory_group, _impl->run_pack);
 }
 
-Status NEMatMul::validate(const ITensorInfo *lhs, const ITensorInfo *rhs, const ITensorInfo *output, const MatMulInfo &info, const CpuMatMulSettings &settings)
+Status NEMatMul::validate(const ITensorInfo *lhs, const ITensorInfo *rhs, const ITensorInfo *output, const MatMulInfo &info, const CpuMatMulSettings &settings, const ActivationLayerInfo &act_info)
 {
-    return cpu::CpuMatMul::validate(lhs, rhs, output, info, settings);
+    return cpu::CpuMatMul::validate(lhs, rhs, output, info, settings, act_info);
 }
 
 void NEMatMul::run()