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, 15 insertions, 6 deletions

diff --git a/src/dynamic_fusion/sketch/gpu/components/cl/ClComponentElementwiseBinary.h b/src/dynamic_fusion/sketch/gpu/components/cl/ClComponentElementwiseBinary.h
index a56dd8b37d..f7175903d0 100644
--- a/src/dynamic_fusion/sketch/gpu/components/cl/ClComponentElementwiseBinary.h
+++ b/src/dynamic_fusion/sketch/gpu/components/cl/ClComponentElementwiseBinary.h
@@ -21,8 +21,8 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#ifndef SRC_DYNAMIC_FUSION_SKETCH_GPU_COMPONENTS_CL_CLCOMPONENTELEMENTWISEBINARY
-#define SRC_DYNAMIC_FUSION_SKETCH_GPU_COMPONENTS_CL_CLCOMPONENTELEMENTWISEBINARY
+#ifndef ACL_SRC_DYNAMIC_FUSION_SKETCH_GPU_COMPONENTS_CL_CLCOMPONENTELEMENTWISEBINARY
+#define ACL_SRC_DYNAMIC_FUSION_SKETCH_GPU_COMPONENTS_CL_CLCOMPONENTELEMENTWISEBINARY
 
 #include "src/dynamic_fusion/sketch/gpu/components/IGpuKernelComponent.h"
 #include "src/dynamic_fusion/sketch/gpu/operators/internal/GpuElementwiseBinaryCommon.h"
@@ -40,8 +40,11 @@ template <typename T>
 class ArgumentPack;
 
 /** Forward declaration */
+#ifndef ACL_INTERNAL_TEST_CKW_IN_DF
 class ClTemplateElementwiseBinary;
+#else  //ACL_INTERNAL_TEST_CKW_IN_DF
 class GpuCkwElementwiseBinary;
+#endif //ACL_INTERNAL_TEST_CKW_IN_DF
 
 class ClComponentElementwiseBinary final : public IGpuKernelComponent
 {
@@ -101,10 +104,13 @@ public:
     ClComponentElementwiseBinary(ClComponentElementwiseBinary &&component) = default;
     /** Allow instances of this class to be moved */
     ClComponentElementwiseBinary &operator=(ClComponentElementwiseBinary &&component) = default;
-    /** Get template writer for the component */
+    /** Get writer for the component */
+#ifndef ACL_INTERNAL_TEST_CKW_IN_DF
     const IGpuTemplateComponentWriter *template_writer() const override;
+#else  //ACL_INTERNAL_TEST_CKW_IN_DF
+    const IGpuCkwComponentDriver            *ckw_component_driver() const override;
+#endif //ACL_INTERNAL_TEST_CKW_IN_DF
 
-    const IGpuCkwComponentDriver *ckw_component_driver() const override;
     /** Get component type */
     GpuComponentType type() const override
     {
@@ -112,10 +118,13 @@ public:
     }
 
 private:
+#ifndef ACL_INTERNAL_TEST_CKW_IN_DF
     std::unique_ptr<ClTemplateElementwiseBinary> _component_writer;
-    std::unique_ptr<GpuCkwElementwiseBinary>     _ckw_driver;
+#else  //ACL_INTERNAL_TEST_CKW_IN_DF
+    std::unique_ptr<GpuCkwElementwiseBinary> _component_writer;
+#endif //ACL_INTERNAL_TEST_CKW_IN_DF
 };
 } // namespace dynamic_fusion
 } // namespace experimental
 } // namespace arm_compute
-#endif /* SRC_DYNAMIC_FUSION_SKETCH_GPU_COMPONENTS_CL_CLCOMPONENTELEMENTWISEBINARY */
+#endif /* ACL_SRC_DYNAMIC_FUSION_SKETCH_GPU_COMPONENTS_CL_CLCOMPONENTELEMENTWISEBINARY */