Rewrite dynamic fusion
The new version introduces the following major changes:
* Change public interface to simplify and standardize the user experience
- Use the term "Workload" uniformly
- Simplify operator interface to be a set of static methods:
validate_op(), create_op()
* Separate the kernel writing into its own component (template_writer).
This is to allow the co-development of GpuKernelWriter, and to allow
easy replacement once GpuKernelWriter is mature.
* Optimize the core fusion algorithm used by the component graph. The
details can be found in GpuKernelComponentGraph::fuse()
* Use Gpu instead of Cl prefixes for most of the Workload interfaces
(except for runtime and kernel components, which have to be language specific)
This allows the potential extension to other Gpu langauges in the
future.
* Refactor runtime memory interface so that auxiliary tensor handling
is separate from the user tensor passing. This is because the former
is less stable and may require extension in the future.
* Hide source code object from the user as it is not required at the
moment
* Deprecate the old prototype entirely by disabling it in SCons build
Resolves COMPMID-5510, COMPMID-5512, COMPMID-5513
Change-Id: If69d2362856f2de4503546b7b6cf48a525cf3079
Signed-off-by: SiCong Li <sicong.li@arm.com>
Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/8406
Tested-by: Arm Jenkins <bsgcomp@arm.com>
Reviewed-by: Gian Marco Iodice <gianmarco.iodice@arm.com>
Reviewed-by: Jakub Sujak <jakub.sujak@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>
diff --git a/src/dynamic_fusion/sketch/gpu/GpuOperatorGroup.h b/src/dynamic_fusion/sketch/gpu/GpuOperatorGroup.h
new file mode 100644
index 0000000..35abe6c
--- /dev/null
+++ b/src/dynamic_fusion/sketch/gpu/GpuOperatorGroup.h
@@ -0,0 +1,111 @@
+/*
+ * Copyright (c) 2022 Arm Limited.
+ *
+ * SPDX-License-Identifier: MIT
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a copy
+ * of this software and associated documentation files (the "Software"), to
+ * deal in the Software without restriction, including without limitation the
+ * rights to use, copy, modify, merge, publish, distribute, sublicense, and/or
+ * sell copies of the Software, and to permit persons to whom the Software is
+ * furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in all
+ * copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+ * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+ * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ */
+#ifndef SRC_DYNAMIC_FUSION_SKETCH_GPU_GPUOPERATORGROUP
+#define SRC_DYNAMIC_FUSION_SKETCH_GPU_GPUOPERATORGROUP
+
+#include "arm_compute/core/ITensorInfo.h"
+#include "src/dynamic_fusion/sketch/ArgumentPack.h"
+#include "src/dynamic_fusion/sketch/gpu/GpuOperatorProperties.h"
+#include "src/dynamic_fusion/sketch/utils/DependencyGraph.h"
+#include <map>
+
+namespace arm_compute
+{
+namespace experimental
+{
+namespace dynamic_fusion
+{
+using OperatorId = DependencyGraph::OperatorId;
+
+/** An operator for the sole purpose of validating fusion
+ */
+class Operator
+{
+public:
+ /** Default constructor */
+ Operator() = default;
+ /** Get Operator Id */
+ OperatorId id() const;
+ /** Get operator type */
+ GpuOperatorType operator_type() const;
+ /** Get tensor arguments */
+ ArgumentPack<ITensorInfo> tensors() const;
+ friend class GpuOperatorGroup;
+
+private:
+ Operator(OperatorId id, GpuOperatorType operator_type, const ArgumentPack<ITensorInfo> &tensors);
+ OperatorId _id{};
+ GpuOperatorType _operator_type{};
+ ArgumentPack<ITensorInfo> _tensors{};
+};
+
+/** A linear sequence of operators to be fused in a workload
+ * For the time being, this class is only used for validating operator fusion
+ * INVARIANTS:
+ * @note These invariants are exactly the same as operator fusion constraints
+ * 1. Fusion is limited to a linear sequence of operators
+ * 2. Max number of operators that can be fused is @ref GpuOperatorGroup::max_fused_operators
+ * 3. The fusion is subject to the pattern: Complex + Simple * | Simple + Simple * | Un-fusable
+ * 4. All operator but unfusable, have exactly 1 dst tensor
+ * 5. All fused operators share the same dst tensor shape
+ * 6. All fused operators' tensors share the same @ref DataLayout
+ */
+class GpuOperatorGroup
+{
+public:
+ static constexpr size_t max_fused_operators = 32;
+ /** Try adding (without actually adding) an operator to the group
+ *
+ * @param[in] op Operator to be added
+ *
+ * @return true If @p op can be added while maintaining the invariants
+ * @return false Otherwise
+ */
+ bool try_add_operator(const Operator &op) const;
+ /** Add an operator to the group
+ *
+ * @param[in] op Operator to be added
+ */
+ void add_operator(const Operator &op);
+ /** Create a new operator
+ *
+ * @param[in] operator_type @ref GpuOperatorType of the new operator
+ * @param[in] tensors Tensor arguments to the new operator
+ *
+ * @return Operator
+ */
+ Operator new_operator(const GpuOperatorType &operator_type, const ArgumentPack<ITensorInfo> &tensors) const;
+ /** Get the "root operator" of the group, which is the first operator in a linear sequence
+ * @return const Operator* Pointer to the root operator
+ */
+ const Operator *get_root_operator() const;
+
+private:
+ DependencyGraph _graph{};
+ std::map<OperatorId, Operator> _operators{};
+};
+} // namespace dynamic_fusion
+} // namespace experimental
+} // namespace arm_compute
+#endif /* SRC_DYNAMIC_FUSION_SKETCH_GPU_GPUOPERATORGROUP */