Files
2026-07-13 12:40:42 +08:00

60 lines
2.0 KiB
C++

// Copyright (c) 2024 PaddlePaddle Authors. All Rights Reserved.
//
// Licensed under the Apache License, Version 2.0 (the "License");
// you may not use this file except in compliance with the License.
// You may obtain a copy of the License at
//
// http://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing, software
// distributed under the License is distributed on an "AS IS" BASIS,
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
// See the License for the specific language governing permissions and
// limitations under the License.
#pragma once
#include "paddle/cinn/pass/pass.h"
namespace cinn {
namespace optim {
class RemoveScheduleBlockPass : public BlockPass {
public:
RemoveScheduleBlockPass() : BlockPass("remove_schedule_block") {}
LogicalResult Run(ir::stmt::BlockRef block) override;
};
/**
* Removes ScheduleBlock nodes from the IR tree.
*
* This pass is applicable in scenarios where ScheduleBlock nodes are present in
* the IR tree but are no longer needed for further optimization.
*
* When applied, this pass will traverse the IR tree and replace each
* ScheduleBlockRealize node with its body. During this process, it will also
* replace the iter_vars in the body with their corresponding iter_values. This
* effectively removes the ScheduleBlock structure while preserving the
* computational logic within it.
*
* Performance impact: This pass addresses the overhead of maintaining
* ScheduleBlock structures in the IR. By removing these structures, it
* simplifies the IR, which can lead to faster subsequent passes and potentially
* more efficient code generation.
*
* Examples:
* 1. Basic ScheduleBlock removal:
* Input IR:
* ScheduleBlock {
* iter_vars: [i, j]
* iter_values: [0, 1]
* body {
* body: A[i, j] = B[i, j] + C[i, j]
* }
* }
* Output IR:
* A[0, 1] = B[0, 1] + C[0, 1]
*/
std::unique_ptr<BlockPass> CreateRemoveScheduleBlockPass();
} // namespace optim
} // namespace cinn