Implement dynamic fusion softmax operator

 - Return aux tensorInfo by get_aux_tensors() at runtime to init the aux
   tensor with the right size.
 - Keep softmax unfusable for this commit
 - Hence, added Tensor3D to template writer arguments declaration, for sake of
   keeping dynamic fusion softmax componenets' kernels matching their cl
   counterparts.

Resolves: COMPMID-5523
Change-Id: I667f39545db925f667036ef448302c79a0330373
Signed-off-by: Ramy Elgammal <ramy.elgammal@arm.com>
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/c/VisualCompute/ComputeLibrary/+/483924
Tested-by: bsgcomp <bsgcomp@arm.com>
Reviewed-by: Gunes Bayir <gunes.bayir@arm.com>
Comments-Addressed: bsgcomp <bsgcomp@arm.com>
Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/8986
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>

1 files changed, 7 insertions, 2 deletions

diff --git a/src/dynamic_fusion/sketch/gpu/GpuKernelComponentGraph.cpp b/src/dynamic_fusion/sketch/gpu/GpuKernelComponentGraph.cpp
index 4cf7a7fece..b70a192775 100644
--- a/src/dynamic_fusion/sketch/gpu/GpuKernelComponentGraph.cpp
+++ b/src/dynamic_fusion/sketch/gpu/GpuKernelComponentGraph.cpp
@@ -59,8 +59,13 @@ GpuKernelComponentStream GpuKernelComponentGraph::fuse(const MemoryDescriptorMap
     {
         const auto component = _components.at(op.op).get();
         const auto success   = stream.add_component(component);
-        ARM_COMPUTE_ERROR_ON(!success);
-        ARM_COMPUTE_UNUSED(success);
+        if(!success) // Assume first failure was because the root component is unfusable
+        {
+            stream.new_component_group();
+            const auto success = stream.add_component(component);
+            ARM_COMPUTE_ERROR_ON(!success);
+            ARM_COMPUTE_UNUSED(success);
+        }
     }
 
     return stream;