Add GpuKernelArgumentBinding for runtime argument setting

* Add flexible runtime argument setting that accept argument bindings
exported from ckw.

* Introduce internal build flag ACL_INTERNAL_TEST_CKW_IN_DF. If set to
true, ckw will be tested in dynamic fusion validation tests. Otherwise
it will not be tested and the dynamic fusion will keep using
ClTemplateWriter instead.

* Fix CKW sampler for elementwise binary to deal with tile sizes > 1
in both dimensions

Resolves: COMPMID-6282
Partially resolves: COMPMID-6260

Signed-off-by: SiCong Li <sicong.li@arm.com>
Change-Id: I0ab225a4484eb2119643d900a4e72806558626ee
Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/9917
Tested-by: Arm Jenkins <bsgcomp@arm.com>
Reviewed-by: Jakub Sujak <jakub.sujak@arm.com>
Reviewed-by: Anitha Raj <Anitha.Raj@arm.com>
Reviewed-by: Viet-Hoa Do <viet-hoa.do@arm.com>
Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
Benchmark: Arm Jenkins <bsgcomp@arm.com>

1 files changed, 3 insertions, 3 deletions

diff --git a/src/dynamic_fusion/sketch/gpu/GpuKernelComponentGraph.cpp b/src/dynamic_fusion/sketch/gpu/GpuKernelComponentGraph.cpp
index b70a192775..5a65ede38b 100644
--- a/src/dynamic_fusion/sketch/gpu/GpuKernelComponentGraph.cpp
+++ b/src/dynamic_fusion/sketch/gpu/GpuKernelComponentGraph.cpp
@@ -44,14 +44,14 @@ std::vector<DependencyGraph::TensorId> GpuKernelComponentGraph::get_tensor_ids(c
     return tensor_ids;
 }
 
-GpuKernelComponentGraph::GpuKernelComponentGraph(GpuComponentServices *services)
-    : _services{ services }, _components{}, _tensors{}, _dependency_graph{}
+GpuKernelComponentGraph::GpuKernelComponentGraph(GpuWorkloadContext *context, GpuComponentServices *services)
+    : _context{ context }, _services{ services }, _components{}, _tensors{}, _dependency_graph{}
 {
 }
 
 GpuKernelComponentStream GpuKernelComponentGraph::fuse(const MemoryDescriptorMap &mem_map) const
 {
-    GpuKernelComponentStream stream{ _services, mem_map };
+    GpuKernelComponentStream stream{ _context, _services, mem_map };
     const auto               op_seq = _dependency_graph.build_operators_sequence();
 
     stream.new_component_group();