Star677
700
代码
代码
Issues54
Pull Requests297
Actions
项目讨论
Wiki
分析
Star677
700
已合并
HCCL软件目录与软件架构保持一致 #2823
miaoziyi1031创建于 8月29日
HCCL软件目录与软件架构保持一致 #2823
已合并
miaoziyi1031创建于 8月29日
从miaoziyi1031/hcclfeature/m00981902_hccl_dir_refactor合入到cann/hcclmaster
共 762 个文件变更+882-717
M.agents/skills/hccl-review/references/architecture.md+3-3
@@ -18,12 +18,12 @@
18## 新算子落标准结构18## 新算子落标准结构
19 19 
20- 官方新算子落 `src/ops/<op>/`；社区试验性新算子落 `experimental/ops/<op>/`（结构与 src 一致，不保证兼容性、不编入商用版本）20- 官方新算子落 `src/ops/<op>/`；社区试验性新算子落 `experimental/ops/<op>/`（结构与 src 一致，不保证兼容性、不编入商用版本）
21-- 均按 `executor/selector/template` 组织；新算子须提供 selector（算法选择）与 template（引擎模板：aicpu/aiv/ccu）21+- 均按 `selector` + `algorithm/{executor,template}` 组织；新算子须提供 selector（算法选择）与 template（引擎模板：aicpu/aiv/ccu）
22- 禁止散落其他目录22- 禁止散落其他目录
23 23 
24## 目录与结构对齐24## 目录与结构对齐
25 25 
26-- 目录结构对齐 architecture-brief 3.2（`src/ops/` 按算子组织、`src/common/` 通用逻辑、`src/op_common/` 四大通用组件）26+- 目录结构对齐 architecture-brief 3.2（`src/ops/` 按算子组织、`src/common/` 通用逻辑、`src/op_common/` 通用组件）
27- `src/` 重命名/移动时同步检查：`CMakeLists.txt`、测试 include 路径、`#include` 相对路径、`cmake/`、`build.sh`27- `src/` 重命名/移动时同步检查：`CMakeLists.txt`、测试 include 路径、`#include` 相对路径、`cmake/`、`build.sh`
28- 重命名 PR 检查是否混入 brief 未明确要求的无关修改28- 重命名 PR 检查是否混入 brief 未明确要求的无关修改
29 29 
@@ -32,7 +32,7 @@
32| 区域 | 风险 |32| 区域 | 风险 |
33|------|------|33|------|------|
34| 算子主流程（`src/ops/<op>/`） | 算法正确性、selector 覆盖、template 引擎匹配 |34| 算子主流程（`src/ops/<op>/`） | 算法正确性、selector 覆盖、template 引擎匹配 |
35-| 通用组件（`src/ops/op_common/` 的 executor/selector/template/topo） | 算法选择逻辑、执行器引擎分发 |35+| 通用组件（`src/ops/op_common/` 的 algorithm/{executor,template,topo_match} + selector + topo_info） | 算法选择逻辑、执行器引擎分发 |
36| hcomm_dlsym（`src/common/hcomm_dlsym/`） | 符号表与 dlsym 加载、版本兼容 |36| hcomm_dlsym（`src/common/hcomm_dlsym/`） | 符号表与 dlsym 加载、版本兼容 |
37| 环境配置（`src/common/alg_env_config/`） | 环境变量解析与默认值 |37| 环境配置（`src/common/alg_env_config/`） | 环境变量解析与默认值 |
38| 图模式与 MC2（`src/common/op_graph/`、`src/common/hccl_mc2/`） | 图构建正确性、自定义算子框架兼容 |38| 图模式与 MC2（`src/common/op_graph/`、`src/common/hccl_mc2/`） | 图构建正确性、自定义算子框架兼容 |
MAGENTS.md+3-3
@@ -15,9 +15,9 @@ HCCL（Huawei Collective Communication Library）是 CANN 的核心集合通信
15```text15```text
16src/16src/
17├── ops/ # 集合通信算子实现（all_reduce/all_gather/broadcast/reduce_scatter/send/recv/...）17├── ops/ # 集合通信算子实现（all_reduce/all_gather/broadcast/reduce_scatter/send/recv/...）
18-│ └── op_common/ # 四大通用组件：executor/selector/template/topo18+│ └── op_common/ # 公共组件：algorithm/{executor,template,topo_match} + selector + topo_info + inc
19└── common/ # 通用逻辑：adapter_acl/alg_env_config/log/param_check/sal/hcomm_dlsym/op_graph/utils/hccl_mc219└── common/ # 通用逻辑：adapter_acl/alg_env_config/log/param_check/sal/hcomm_dlsym/op_graph/utils/hccl_mc2
20-experimental/ # 社区贡献的试验性代码（结构与 src 一致，含 ops/；不保证兼容性，不编入商用版本）20+experimental/ # 社区贡献的试验性代码（当前尚未完全与 src 对齐，含 ops/；不保证兼容性，不编入商用版本）
21include/ # 对外头文件：hccl.h（算子 API）、hccl_mc2.h（MC2 自定义算子框架）21include/ # 对外头文件：hccl.h（算子 API）、hccl_mc2.h（MC2 自定义算子框架）
22test/ # ut / st22test/ # ut / st
23docs/ # 资料文档23docs/ # 资料文档
@@ -47,7 +47,7 @@ build.sh # 一键编译脚本
47| **分层依赖方向**：上层依赖下层，下层不能反向依赖上层（HCOMM 的 `base_comm` ↛ `coll_communicator_mgr` ↛ `coll_comm_ops`） | HCCL 不得被 HCOMM 反向依赖；HCCL 算子通过 dlsym 调 HCOMM，不得要求 HCOMM 反向 include HCCL 头 |47| **分层依赖方向**：上层依赖下层，下层不能反向依赖上层（HCOMM 的 `base_comm` ↛ `coll_communicator_mgr` ↛ `coll_comm_ops`） | HCCL 不得被 HCOMM 反向依赖；HCCL 算子通过 dlsym 调 HCOMM，不得要求 HCOMM 反向 include HCCL 头 |
48| **控制面/数据面分离**：资源管理、拓扑查询（控制面）与数据搬运/同步（数据面）接口独立演进 | HCCL 算子属数据面消费方；不得在算子层引入对 HCOMM 控制面内部实现的耦合 |48| **控制面/数据面分离**：资源管理、拓扑查询（控制面）与数据搬运/同步（数据面）接口独立演进 | HCCL 算子属数据面消费方；不得在算子层引入对 HCOMM 控制面内部实现的耦合 |
49| **HCCL 与 HCOMM 解耦**：HCCL 算子通过 `dlsym` 动态加载 HCOMM 接口，两仓独立编译、独立版本演进 | HCCL 不得 `#include` HCOMM 私有头；不得引入对 `cann/hcomm` 的编译期硬依赖；跨仓调用走 `src/common/hcomm_dlsym/` 的符号表 + `dlsym` |49| **HCCL 与 HCOMM 解耦**：HCCL 算子通过 `dlsym` 动态加载 HCOMM 接口，两仓独立编译、独立版本演进 | HCCL 不得 `#include` HCOMM 私有头；不得引入对 `cann/hcomm` 的编译期硬依赖；跨仓调用走 `src/common/hcomm_dlsym/` 的符号表 + `dlsym` |
50-| **新算子落标准结构**：官方新算子落 `src/ops/<op>/`；社区贡献的试验性新算子落 `experimental/ops/<op>/`（结构与 `src` 一致，不保证兼容性、不编入商用版本）。均按 `executor/selector/template` 组织 | 新算子须提供 selector（算法选择）与 template（引擎模板：aicpu/aiv/ccu）；官方算子落 `src/ops/`，社区试验算子落 `experimental/ops/`，禁止散落其他目录 |50+| **新算子落标准结构**：官方新算子落 `src/ops/<op>/`；社区贡献的试验性新算子落 `experimental/ops/<op>/`（结构与 `src` 一致，不保证兼容性、不编入商用版本）。均按 `selector` + `algorithm/{executor,template}` 组织 | 新算子须提供 selector（算法选择）与 template（引擎模板：aicpu/aiv/ccu）；官方算子落 `src/ops/`，社区试验算子落 `experimental/ops/`，禁止散落其他目录 |
51 51 
52### 对外 API52### 对外 API
53 53 
MREADME.md+7-4
@@ -44,10 +44,13 @@ HCCL包含HCCL集合通信库与HCOMM（Huawei Communication）通信基础库
44| ├── broadcast # Broadcast算子实现44| ├── broadcast # Broadcast算子实现
45| ├── interface_graph_mode # 图模式接口实现45| ├── interface_graph_mode # 图模式接口实现
46| ├── op_common # 算子通用组件46| ├── op_common # 算子通用组件
47-| │ ├── executor # 执行器47+| │ ├── algorithm # 算法层
48-| │ ├── selector # 算法选择器48+| │ │ ├── executor # 执行器
49-| │ ├── template # 算法模板49+| │ │ ├── template # 算法模板
50-| │ └── topo # 通信域拓扑信息获取和转换50+| │ │ └── topo_match # 算法执行阶段拓扑层级匹配
51+| │ ├── selector # 算法选择、CostModel等选择逻辑
52+| │ ├── topo_info # 公共拓扑信息生成、转换与Rank映射
53+| │ └── inc # 公共头文件
51| ├── recv # Recv算子实现54| ├── recv # Recv算子实现
C
Cclause8月31日

AGENTS.md中也有对目录结构的描述，需要一起刷新

likedislike
miaoziyi1031
miaoziyi1031
9月1日 评论：
52| ├── reduce # Reduce算子实现55| ├── reduce # Reduce算子实现
53| ├── reduce_scatter # ReduceScatter算子实现56| ├── reduce_scatter # ReduceScatter算子实现
MREADME_en.md+7-4
@@ -44,10 +44,13 @@ The key directories of this project are as follows:
44| ├── broadcast # Broadcast operator implementation44| ├── broadcast # Broadcast operator implementation
45| ├── interface_graph_mode # Graph mode interface implementation45| ├── interface_graph_mode # Graph mode interface implementation
46| ├── op_common # Common operator components46| ├── op_common # Common operator components
47-| │ ├── executor # Executor47+| │ ├── algorithm # Algorithm layer
48-| │ ├── selector # Algorithm selector48+| │ │ ├── executor # Executor
49-| │ ├── template # Algorithm template49+| │ │ ├── template # Algorithm template
50-| │ └── topo # Communication domain topology information acquisition and conversion50+| │ │ └── topo_match # Topology level matching during algorithm execution
51+| │ ├── selector # Algorithm selection and cost model
52+| │ ├── topo_info # Public topology info generation, conversion and rank mapping
53+| │ └── inc # Public headers
51| ├── recv # Recv operator implementation54| ├── recv # Recv operator implementation
52| ├── reduce # Reduce operator implementation55| ├── reduce # Reduce operator implementation
53| ├── reduce_scatter # ReduceScatter operator implementation56| ├── reduce_scatter # ReduceScatter operator implementation
Mdocs/en/rfcs/0001-add-batch-invariant-reducescatter.md+1-1
@@ -75,7 +75,7 @@ The BIRS algorithm is integrated into HCCL as an experimental feature, isolated
75```text75```text
76HCCL76HCCL
77├── src/ops/reduce_scatter/ # Existing ReduceScatter implementation77├── src/ops/reduce_scatter/ # Existing ReduceScatter implementation
78-│ └── reduce_scatter_op.cc/.h # Entry function (with BIRS dispatch logic added)78+│ └── reduce_scatter.cc/.h # Entry function (with BIRS dispatch logic added)
79│79│
80├── experimental/ops/ # Experimental features directory (new)80├── experimental/ops/ # Experimental features directory (new)
81│ ├── op_common/ # Common infrastructure81│ ├── op_common/ # Common infrastructure
Mdocs/en/rfcs/0002-HCCL-ALGO-Plugin.md+12-9
@@ -58,16 +58,19 @@ The key directories of the HCCL communication library are shown below:
58| ├── all_gather # AllGather operator implementation58| ├── all_gather # AllGather operator implementation
59| ├── all_reduce # AllReduce operator implementation59| ├── all_reduce # AllReduce operator implementation
60| ├── broadcast # Broadcast operator implementation60| ├── broadcast # Broadcast operator implementation
61-| | ├── executor # Broadcast operator executor61+| | ├── algorithm # Broadcast algorithm layer
62+| | | ├── executor # Broadcast operator executor
63+| | | └── template # Broadcast algorithm template
62| | ├── selector # Broadcast algorithm selector64| | ├── selector # Broadcast algorithm selector
63-| | ├── template # Broadcast algorithm template65+| | └── broadcast.cc # External API implementation of the Broadcast operator
64-| | └── broadcast_op.cc # External API implementation of the Broadcast operator
65| ├── ...... # Other operator implementations66| ├── ...... # Other operator implementations
66| └── op_common # Common operator components67| └── op_common # Common operator components
67-| ├── executor # Executors68+| ├── algorithm # Algorithm layer
69+| | ├── executor # Executors
70+| | ├── template # Algorithm templates
71+| | └── topo_match # Topology hierarchy matching
68| ├── selector # Algorithm selectors72| ├── selector # Algorithm selectors
69-| ├── template # Algorithm templates73+| ├── topo_info # Communication-domain topology acquisition and conversion
70-| ├── topo # Communication-domain topology acquisition and conversion
71| └── op_common.cc # Common operator functions74| └── op_common.cc # Common operator functions
72├── include # HCCL public headers75├── include # HCCL public headers
73├── test # Test code directory76├── test # Test code directory
@@ -76,7 +79,7 @@ The key directories of the HCCL communication library are shown below:
76└── ....... # Other directories79└── ....... # Other directories
77 80 
78```81```
79-The `/ops` directory defines HCCL operator implementations, including common collective communication operators such as `all_gather` and `all_reduce`. Each operator implements its executor (`executor`), algorithm selector (`selector`), algorithm template (`template`), and public API file (`XX_op.cc`).82+The `/ops` directory defines HCCL operator implementations, including common collective communication operators such as `all_gather` and `all_reduce`. Each operator implements its executor (`executor`), algorithm selector (`selector`), algorithm template (`template`), and operator entry implementation file (`XX.cc`).
80 83 
81The `/op_common` directory under `/ops` defines common operator components, including executor base classes, shared algorithm-selection logic, algorithm-template base classes, communication-domain topology processing, and other infrastructure shared by operators.84The `/op_common` directory under `/ops` defines common operator components, including executor base classes, shared algorithm-selection logic, algorithm-template base classes, communication-domain topology processing, and other infrastructure shared by operators.
82 85 
@@ -109,11 +112,11 @@ Select the algorithm name according to topology information, such as the exact s
109② Create threads and calculate the resources required for communication;112② Create threads and calculate the resources required for communication;
110③ Execute the algorithm by invoking the algorithm orchestration of the `executor`, namely `executor->Orchestrate()`.113③ Execute the algorithm by invoking the algorithm orchestration of the `executor`, namely `executor->Orchestrate()`.
111 114 
112-2) The `Orchestrate()` function of the `executor` is located in `src/ops/broadcast/executor/ins_v2_broadcast_sole_executor.cc`. Its main logic is:115+2) The `Orchestrate()` function of the `executor` is located in `src/ops/broadcast/algorithm/executor/ins_v2_broadcast_sole_executor.cc`. Its main logic is:
113① Further calculate resources, perform data slicing, and complete other preparation steps;116① Further calculate resources, perform data slicing, and complete other preparation steps;
114② Invoke the `KernelRun` function of the `ParallelMesh1DNHR` algorithm template to complete the communication operation, namely `algTemplate->KernelRun()`.117② Invoke the `KernelRun` function of the `ParallelMesh1DNHR` algorithm template to complete the communication operation, namely `algTemplate->KernelRun()`.
115 118 
116-3) The `KernelRun` function of the `ParallelMesh1DNHR` algorithm is located in `src/ops/broadcast/template/aicpu/ins_temp_broadcast_nhr.cc`. Its main logic is:119+3) The `KernelRun` function of the `ParallelMesh1DNHR` algorithm is located in `src/ops/broadcast/algorithm/template/aicpu/ins_temp_broadcast_nhr.cc`. Its main logic is:
117Execute remote reads, remote writes, local thread synchronization, local data copies, and other operations according to the NHR algorithm logic to complete the communication.120Execute remote reads, remote writes, local thread synchronization, local data copies, and other operations according to the NHR algorithm logic to complete the communication.
118 121 
119---122---
Mdocs/zh/architecture/architecture-brief.md+9-6
@@ -213,10 +213,13 @@ hccl
213| ├── batch_send_recv # BatchSendRecv算子实现213| ├── batch_send_recv # BatchSendRecv算子实现
214| ├── broadcast # Broadcast算子实现214| ├── broadcast # Broadcast算子实现
215| ├── op_common # 算子通用组件215| ├── op_common # 算子通用组件
216-| │ ├── executor # 算法执行器216+| │ ├── algorithm # 算法层
严正行
严正行严正行8月30日

[MEDIUM] architecture-brief 断言 experimental 与 src 结构一致，但 PR 未同步对齐 experimental/ops

问题分析：

  • 第 216-222 行：本 PR 更新了 brief 3.2 节 src/ops 的目标目录结构（op_common 下新增 algorithm/executor、algorithm/template、algorithm/topo_match、topo_info、inc）。
  • brief 同节对 experimental 的既有断言（「experimental 内部主要目录结构和src保持一致」）未变，但 experimental/ops/op_common/ 下仍是旧结构 template/、topo/（experimental/ops/CMakeLists.txt 第 22-23、33-34 行仍引用 op_common/template、op_common/topo），与 src 新结构（algorithm/template、topo_info、algorithm/topo_match）不一致。
  • AGENTS.md 目录约束同样要求 experimental 结构与 src 一致。

风险：文档断言与实际目录相悖；experimental 的 alg_template_base_experimental.h 依赖的 alg_template_base.h 等基类头已迁至 algorithm/template，后续维护者按 brief 目录树在 experimental 找对应目录会落空，两套结构长期并存加大认知成本。

建议修复： 在本次结构对齐中同步调整 experimental/ops/op_common/（template → algorithm/template，topo → 拆分对齐 topo_info/topo_match 语义）并更新 experimental/ops/CMakeLists.txt 的路径；若确认 experimental 属于后续单独任务，请在 brief 的 experimental 描述处加注「暂未对齐，后续任务跟踪」并在 Issue#658 中登记，避免文档断言与现状矛盾。

likedislike
miaoziyi1031
miaoziyi1031
9月1日 评论：
217-| │ ├── selector # 算法选择器217+| │ │ ├── executor # 算法执行器
218-| │ ├── template # 算法模板218+| │ │ ├── template # 算法模板
219-| │ └── topo # 通信算子的rankGraph拓扑信息适配219+| │ │ └── topo_match # 算法执行阶段拓扑层级匹配
220+| │ ├── selector # 算法选择、CostModel等选择逻辑
221+| │ ├── topo_info # 公共拓扑信息生成、转换与Rank映射
222+| │ └── inc # 公共头文件
220| ├── recv # Recv算子实现223| ├── recv # Recv算子实现
221| ├── reduce # Reduce算子实现224| ├── reduce # Reduce算子实现
222| ├── reduce_scatter # ReduceScatter算子实现225| ├── reduce_scatter # ReduceScatter算子实现
@@ -224,7 +227,7 @@ hccl
224| ├── scatter # Scatter算子实现227| ├── scatter # Scatter算子实现
225| └── send # Send算子实现228| └── send # Send算子实现
226├── include # HCCL对外头文件229├── include # HCCL对外头文件
227-├── experimental # 社区贡献的试验性代码目录（内部主要目录结构和src保持一致，不保证新接口的兼容性，当前也不会被商用版本采纳）230+├── experimental # 社区贡献的试验性代码目录（内部主要目录结构当前尚未完全与 src 对齐，相关目录调整将在后续任务中跟踪，不保证新接口的兼容性，当前也不会被商用版本采纳）
严正行
严正行严正行9月1日

[MEDIUM] experimental 断言修正不完整：HCOMM 目录树同款失实断言残留

问题分析：

  • 第 230 行：experimental 该处已按上条意见改为「内部主要目录结构当前尚未完全与 src 对齐，相关目录调整将在后续任务中跟踪」，与 experimental/ops/op_common/ 当前实际旧结构（template/、topo/，见 experimental/ops/CMakeLists.txt 第 22-24 行 include 路径）相符，HCCL 目录树这处修正认可。
  • 但同一 3.2 节紧随其后的 HCOMM 仓目录树（当前 head 第 255 行）仍保留旧断言「experimental……内部主要目录结构和src保持一致……」，未同步修改。实测 hcomm 仓 master：experimental/base_comm/ 下为 channel、comm_mem、endpoint、nic_plugin（另有独立 cluster_link_diag/），而 src/base_comm/ 下为 common、config_mgr、dfx、primitives、resources——两者结构并不一致，残留断言与第 230 行新表述在同一节内自相矛盾。
  • 「相关目录调整将在后续任务中跟踪」缺少登记载体：Issue#658（本 PR 对应的 src/ops 目录结构对齐需求）正文与评论均未提及 experimental。

建议修复：

  • 将第 255 行 HCOMM 目录树中 experimental 的描述同步改为与第 230 行一致的「尚未完全与 src 对齐、后续任务跟踪」表述（hcomm 部分请按 hcomm 仓实际目录核实措辞），消除同一文档两处断言一新一旧的矛盾。
  • 在 Issue#658 或新建 Issue 中登记 experimental 目录对齐的后续任务，使「后续任务中跟踪」有实际跟踪载体。
likedislike
228```231```
229 232 
230HCOMM仓的目标目录结构：233HCOMM仓的目标目录结构：
@@ -249,7 +252,7 @@ hcomm
249│ └── ascend950 # A5旧流程兼容代码252│ └── ascend950 # A5旧流程兼容代码
250├── include # 对外头文件253├── include # 对外头文件
251├── pkg_inc # 包间接口头文件254├── pkg_inc # 包间接口头文件
252-├── experimental # 社区贡献的试验性代码目录（内部主要目录结构和src保持一致，不保证新接口的兼容性，当前也不会被商用版本采纳）255+├── experimental # 社区贡献的试验性代码目录（内部主要目录结构当前尚未完全与 src 对齐，相关目录调整将在后续任务中跟踪，不保证新接口的兼容性，当前也不会被商用版本采纳）
253```256```
254 257 
255> **legacy = 历史兼容，不持续演进**258> **legacy = 历史兼容，不持续演进**
Mdocs/zh/rfcs/0001-add-batch-invariant-reducescatter.md+1-1
@@ -75,7 +75,7 @@ BIRS算法作为实验特性集成到HCCL中，通过独立代码路径和编译
75```text75```text
76HCCL76HCCL
77├── src/ops/reduce_scatter/ # 现有 ReduceScatter 实现77├── src/ops/reduce_scatter/ # 现有 ReduceScatter 实现
78-│ └── reduce_scatter_op.cc/.h # 入口函数（新增BIRS分发逻辑）78+│ └── reduce_scatter.cc/.h # 入口函数（新增BIRS分发逻辑）
79│79│
80├── experimental/ops/ # 实验特性目录（新增）80├── experimental/ops/ # 实验特性目录（新增）
81│ ├── op_common/ # 公共基础设施81│ ├── op_common/ # 公共基础设施
Mdocs/zh/rfcs/0002-HCCL-ALGO-Plugin.md+13-10
@@ -58,16 +58,19 @@
58| ├── all_gather # AllGather算子实现58| ├── all_gather # AllGather算子实现
59| ├── all_reduce # AllReduce算子实现59| ├── all_reduce # AllReduce算子实现
60| ├── broadcast # Broadcast算子实现60| ├── broadcast # Broadcast算子实现
61-| | ├── executor # Broadcast算子执行器61+| | ├── algorithm # Broadcast算法层
62+| | | ├── executor # Broadcast算子执行器
63+| | | └── template # Broadcast算法模板
62| | ├── selector # Broadcast算法选择器64| | ├── selector # Broadcast算法选择器
63-| | ├── template # Broadcast算法模板65+| | └── broadcast.cc # Broadcast算子对外提供的API实现
64-| | └── broadcast_op.cc # Broadcast算子对外提供的API实现
65| ├── ...... # 其他算子实现66| ├── ...... # 其他算子实现
66| └── op_common # 算子通用组件67| └── op_common # 算子通用组件
67-|   ├── executor # 执行器68+| ├── algorithm # 算法层
68-|   ├── selector # 算法选择器69+| | ├── executor # 执行器
69-| ├── template # 算法模板70+| | ├── template # 算法模板
70-|   ├── topo # 通信域拓扑信息获取和转换 71+| | └── topo_match # 拓扑层级匹配
72+| ├── selector # 算法选择器
73+| ├── topo_info # 通信域拓扑信息获取和转换
71| └── op_common.cc # 算子通用函数文件74| └── op_common.cc # 算子通用函数文件
72├── include # HCCL对外头文件75├── include # HCCL对外头文件
73├── test # 测试代码目录76├── test # 测试代码目录
@@ -76,7 +79,7 @@
76└── ....... # 其他目录79└── ....... # 其他目录
77 80 
78```81```
79-　　`/ops`目录定义了HCCL算子实现，包含`all_gather`、`all_reduce`等常见集合通信算子，每个算子实现其执行器（`executor`）、算法选择器（`selector`）、算法模板（`template`）和对外提供的API文件（`XX_op.cc`）。 82+　　`/ops`目录定义了HCCL算子实现，包含`all_gather`、`all_reduce`等常见集合通信算子，每个算子实现其执行器（`executor`）、算法选择器（`selector`）、算法模板（`template`）和算子入口实现文件（`XX.cc`）。
80 83 
81　　`/ops`目录下的`/op_common`目录定义了算子通用组件，包括算子执行器基类、算法选择器公共逻辑、算法模板基类、通信域拓扑处理等各算子共用的基础设施。84　　`/ops`目录下的`/op_common`目录定义了算子通用组件，包括算子执行器基类、算法选择器公共逻辑、算法模板基类、通信域拓扑处理等各算子共用的基础设施。
82 85 
@@ -101,10 +104,10 @@
101　　　　　　① 根据操作类型和选定的算法名称(以下都假定选中`ParallelMesh1DNHR`算法)，获取对应的`executor`实例。 104　　　　　　① 根据操作类型和选定的算法名称(以下都假定选中`ParallelMesh1DNHR`算法)，获取对应的`executor`实例。
102　　　　　　　　　　　　② 创建线程、计算通信所需资源等。 105　　　　　　　　　　　　② 创建线程、计算通信所需资源等。
103　　　　　　　　　　　　③ 算法执行：调用`executor`的算法编排，即调用`executor->Orchestrate()`函数。 106　　　　　　　　　　　　③ 算法执行：调用`executor`的算法编排，即调用`executor->Orchestrate()`函数。
104-　　　　2) `executor`的`Orchestrate()`函数位于`src/ops/broadcast/executor/ins_v2_broadcast_sole_executor.cc`中，主要逻辑为： 107+　　　　2) `executor`的`Orchestrate()`函数位于`src/ops/broadcast/algorithm/executor/ins_v2_broadcast_sole_executor.cc`中，主要逻辑为：
105　　　　　　① 进一步计算资源、进行数据分片等步骤。 108　　　　　　① 进一步计算资源、进行数据分片等步骤。
106　　　　　　　　　　　　② 调用`ParallelMesh1DNHR`算法模板的`KernelRun`函数，完成该次通信操作，即`algTemplate->KernelRun()`。 109　　　　　　　　　　　　② 调用`ParallelMesh1DNHR`算法模板的`KernelRun`函数，完成该次通信操作，即`algTemplate->KernelRun()`。
107-　　　　3) `ParallelMesh1DNHR`算法的`KernelRun`函数位于`src/ops/broadcast/template/aicpu/ins_temp_broadcast_nhr.cc`中，其主要逻辑为： 110+　　　　3) `ParallelMesh1DNHR`算法的`KernelRun`函数位于`src/ops/broadcast/algorithm/template/aicpu/ins_temp_broadcast_nhr.cc`中，其主要逻辑为：
108　　　　　　根据NHR算法逻辑执行远端读、远端写、本地线程同步、本地数据拷贝等操作来完成该次通信。111　　　　　　根据NHR算法逻辑执行远端读、远端写、本地线程同步、本地数据拷贝等操作来完成该次通信。
109 112 
110---113---
Mexamples/05_custom_ops_allgather/aiv/op_kernel/CMakeLists.txt+2-2
@@ -69,8 +69,8 @@ set(ASC_INCLUDE_FLAGS
69 69 
70if(DEFINED OPS_BASE_DIR)70if(DEFINED OPS_BASE_DIR)
71 list(APPEND ASC_INCLUDE_FLAGS71 list(APPEND ASC_INCLUDE_FLAGS
72- -I${OPS_BASE_DIR}/src/ops/op_common/template/aiv/72+ -I${OPS_BASE_DIR}/src/ops/op_common/algorithm/template/aiv/
73- -I${OPS_BASE_DIR}/src/ops/op_common/template/aiv/aiv_interface/73+ -I${OPS_BASE_DIR}/src/ops/op_common/algorithm/template/aiv/aiv_interface/
74 )74 )
75endif()75endif()
76 76 
Mexamples/06_custom_ops_reduce_scatter/aiv/op_kernel/CMakeLists.txt+2-2
@@ -69,8 +69,8 @@ set(ASC_INCLUDE_FLAGS
69 69 
70if(DEFINED OPS_BASE_DIR)70if(DEFINED OPS_BASE_DIR)
71 list(APPEND ASC_INCLUDE_FLAGS71 list(APPEND ASC_INCLUDE_FLAGS
72- -I${OPS_BASE_DIR}/src/ops/op_common/template/aiv/72+ -I${OPS_BASE_DIR}/src/ops/op_common/algorithm/template/aiv/
73- -I${OPS_BASE_DIR}/src/ops/op_common/template/aiv/aiv_interface/73+ -I${OPS_BASE_DIR}/src/ops/op_common/algorithm/template/aiv/aiv_interface/
74 )74 )
75endif()75endif()
76 76 
Mexperimental/ops/op_common/op_common_experimental.cc+1-1
@@ -14,7 +14,7 @@
14#include "coll_alg_exec_registry.h"14#include "coll_alg_exec_registry.h"
15#include "config_log.h"15#include "config_log.h"
16#include "op_common_ops.h"16#include "op_common_ops.h"
17-#include "scatter_op.h"17+#include "scatter.h"
18#include "topo.h"18#include "topo.h"
19#include "topo_host.h"19#include "topo_host.h"
20#include "hcomm_host_profiling_dl.h"20#include "hcomm_host_profiling_dl.h"
Mexperimental/ops/op_common/op_common_experimental.h+1-1
@@ -17,7 +17,7 @@
17#include "executor_base.h"17#include "executor_base.h"
18#include "alg_type.h"18#include "alg_type.h"
19#include "op_common.h"19#include "op_common.h"
20-#include "scatter_op.h"20+#include "scatter.h"
21 21 
22#ifdef __cplusplus22#ifdef __cplusplus
23extern "C" {23extern "C" {
Mexperimental/ops/reduce_scatter/reduce_scatter_op_experimental.cc+1-1
@@ -16,7 +16,7 @@
16#include <string>16#include <string>
17#include "op_common_experimental.h"17#include "op_common_experimental.h"
18#include "param_check.h"18#include "param_check.h"
19-#include "reduce_scatter_op.h"19+#include "reduce_scatter.h"
20#include "load_kernel.h"20#include "load_kernel.h"
21#include "hcomm_dlsym.h"21#include "hcomm_dlsym.h"
22#include "hcomm_host_profiling_dl.h"22#include "hcomm_host_profiling_dl.h"
Msrc/CMakeLists.txt+78-79
@@ -88,130 +88,129 @@ set(INCLUDE_LIST
88 ${CMAKE_CURRENT_SOURCE_DIR}/ops/op_common88 ${CMAKE_CURRENT_SOURCE_DIR}/ops/op_common
89 ${CMAKE_CURRENT_SOURCE_DIR}/ops/op_common/inc89 ${CMAKE_CURRENT_SOURCE_DIR}/ops/op_common/inc
90 90 
91- ${CMAKE_CURRENT_SOURCE_DIR}/ops/op_common/executor91+ ${CMAKE_CURRENT_SOURCE_DIR}/ops/op_common/algorithm/executor
92- ${CMAKE_CURRENT_SOURCE_DIR}/ops/op_common/executor/channel92+ ${CMAKE_CURRENT_SOURCE_DIR}/ops/op_common/algorithm/executor/channel
93- ${CMAKE_CURRENT_SOURCE_DIR}/ops/op_common/executor/registry93+ ${CMAKE_CURRENT_SOURCE_DIR}/ops/op_common/algorithm/executor/registry
94 94 
95 ${CMAKE_CURRENT_SOURCE_DIR}/ops/op_common/selector95 ${CMAKE_CURRENT_SOURCE_DIR}/ops/op_common/selector
96 ${CMAKE_CURRENT_SOURCE_DIR}/ops/op_common/selector/inc96 ${CMAKE_CURRENT_SOURCE_DIR}/ops/op_common/selector/inc
97 97 
98- ${CMAKE_CURRENT_SOURCE_DIR}/ops/op_common/template98+ ${CMAKE_CURRENT_SOURCE_DIR}/ops/op_common/algorithm/template
99- ${CMAKE_CURRENT_SOURCE_DIR}/ops/op_common/template/registry99+ ${CMAKE_CURRENT_SOURCE_DIR}/ops/op_common/algorithm/template/registry
100- ${CMAKE_CURRENT_SOURCE_DIR}/ops/op_common/template/aicpu100+ ${CMAKE_CURRENT_SOURCE_DIR}/ops/op_common/algorithm/template/aicpu
101- ${CMAKE_CURRENT_SOURCE_DIR}/ops/op_common/template/aicpu/task_cache101+ ${CMAKE_CURRENT_SOURCE_DIR}/ops/op_common/algorithm/template/aicpu/task_cache
102- ${CMAKE_CURRENT_SOURCE_DIR}/ops/op_common/template/wrapper102+ ${CMAKE_CURRENT_SOURCE_DIR}/ops/op_common/algorithm/template/wrapper
103- ${CMAKE_CURRENT_SOURCE_DIR}/ops/op_common/template/aiv103+ ${CMAKE_CURRENT_SOURCE_DIR}/ops/op_common/algorithm/template/aiv
104- # ${CMAKE_CURRENT_SOURCE_DIR}/ops/op_common/template/hostdpu104+ # ${CMAKE_CURRENT_SOURCE_DIR}/ops/op_common/algorithm/template/hostdpu
105- ${CMAKE_CURRENT_SOURCE_DIR}/ops/op_common/template/dpu105+ ${CMAKE_CURRENT_SOURCE_DIR}/ops/op_common/algorithm/template/dpu
106- ${CMAKE_CURRENT_SOURCE_DIR}/ops/op_common/template/ccu106+ ${CMAKE_CURRENT_SOURCE_DIR}/ops/op_common/algorithm/template/ccu
107- ${CMAKE_CURRENT_SOURCE_DIR}/ops/op_common/template/ccu/kernel
108 107 
109- ${CMAKE_CURRENT_SOURCE_DIR}/ops/op_common/topo108+ ${CMAKE_CURRENT_SOURCE_DIR}/ops/op_common/topo_info
109+ ${CMAKE_CURRENT_SOURCE_DIR}/ops/op_common/algorithm/topo_match
110 ${CMAKE_CURRENT_SOURCE_DIR}/ops/interface_graph_mode110 ${CMAKE_CURRENT_SOURCE_DIR}/ops/interface_graph_mode
111 111 
112 ${CMAKE_CURRENT_SOURCE_DIR}/ops/reduce_scatter112 ${CMAKE_CURRENT_SOURCE_DIR}/ops/reduce_scatter
113- ${CMAKE_CURRENT_SOURCE_DIR}/ops/reduce_scatter/executor113+ ${CMAKE_CURRENT_SOURCE_DIR}/ops/reduce_scatter/algorithm/executor
114 ${CMAKE_CURRENT_SOURCE_DIR}/ops/reduce_scatter/selector114 ${CMAKE_CURRENT_SOURCE_DIR}/ops/reduce_scatter/selector
115- ${CMAKE_CURRENT_SOURCE_DIR}/ops/reduce_scatter/template115+ ${CMAKE_CURRENT_SOURCE_DIR}/ops/reduce_scatter/algorithm/template
116- ${CMAKE_CURRENT_SOURCE_DIR}/ops/reduce_scatter/template/aicpu116+ ${CMAKE_CURRENT_SOURCE_DIR}/ops/reduce_scatter/algorithm/template/aicpu
117- ${CMAKE_CURRENT_SOURCE_DIR}/ops/reduce_scatter/template/aiv117+ ${CMAKE_CURRENT_SOURCE_DIR}/ops/reduce_scatter/algorithm/template/aiv
118- ${CMAKE_CURRENT_SOURCE_DIR}/ops/reduce_scatter/template/ccu118+ ${CMAKE_CURRENT_SOURCE_DIR}/ops/reduce_scatter/algorithm/template/ccu
119- ${CMAKE_CURRENT_SOURCE_DIR}/ops/reduce_scatter/template/ccu/kernel119+ ${CMAKE_CURRENT_SOURCE_DIR}/ops/reduce_scatter/algorithm/template/ccu/kernel
120 120 
121 ${CMAKE_CURRENT_SOURCE_DIR}/ops/broadcast121 ${CMAKE_CURRENT_SOURCE_DIR}/ops/broadcast
122- ${CMAKE_CURRENT_SOURCE_DIR}/ops/broadcast/executor122+ ${CMAKE_CURRENT_SOURCE_DIR}/ops/broadcast/algorithm/executor
123 ${CMAKE_CURRENT_SOURCE_DIR}/ops/broadcast/selector123 ${CMAKE_CURRENT_SOURCE_DIR}/ops/broadcast/selector
124- ${CMAKE_CURRENT_SOURCE_DIR}/ops/broadcast/template124+ ${CMAKE_CURRENT_SOURCE_DIR}/ops/broadcast/algorithm/template
125- ${CMAKE_CURRENT_SOURCE_DIR}/ops/broadcast/template/aicpu125+ ${CMAKE_CURRENT_SOURCE_DIR}/ops/broadcast/algorithm/template/aicpu
126- ${CMAKE_CURRENT_SOURCE_DIR}/ops/broadcast/template/aiv126+ ${CMAKE_CURRENT_SOURCE_DIR}/ops/broadcast/algorithm/template/aiv
127- ${CMAKE_CURRENT_SOURCE_DIR}/ops/broadcast/template/ccu127+ ${CMAKE_CURRENT_SOURCE_DIR}/ops/broadcast/algorithm/template/ccu
128- ${CMAKE_CURRENT_SOURCE_DIR}/ops/broadcast/template/ccu/kernel128+ ${CMAKE_CURRENT_SOURCE_DIR}/ops/broadcast/algorithm/template/ccu/kernel
129 129 
130 ${CMAKE_CURRENT_SOURCE_DIR}/ops/reduce_scatter_v130 ${CMAKE_CURRENT_SOURCE_DIR}/ops/reduce_scatter_v
131- ${CMAKE_CURRENT_SOURCE_DIR}/ops/reduce_scatter_v/executor131+ ${CMAKE_CURRENT_SOURCE_DIR}/ops/reduce_scatter_v/algorithm/executor
132 ${CMAKE_CURRENT_SOURCE_DIR}/ops/reduce_scatter_v/selector132 ${CMAKE_CURRENT_SOURCE_DIR}/ops/reduce_scatter_v/selector
133- ${CMAKE_CURRENT_SOURCE_DIR}/ops/reduce_scatter_v/template133+ ${CMAKE_CURRENT_SOURCE_DIR}/ops/reduce_scatter_v/algorithm/template
134- ${CMAKE_CURRENT_SOURCE_DIR}/ops/reduce_scatter_v/template/aicpu134+ ${CMAKE_CURRENT_SOURCE_DIR}/ops/reduce_scatter_v/algorithm/template/aicpu
135- ${CMAKE_CURRENT_SOURCE_DIR}/ops/reduce_scatter_v/template/ccu135+ ${CMAKE_CURRENT_SOURCE_DIR}/ops/reduce_scatter_v/algorithm/template/ccu
136- ${CMAKE_CURRENT_SOURCE_DIR}/ops/reduce_scatter_v/template/ccu/kernel136+ ${CMAKE_CURRENT_SOURCE_DIR}/ops/reduce_scatter_v/algorithm/template/ccu/kernel
137 137 
138 ${CMAKE_CURRENT_SOURCE_DIR}/ops/scatter138 ${CMAKE_CURRENT_SOURCE_DIR}/ops/scatter
139- ${CMAKE_CURRENT_SOURCE_DIR}/ops/scatter/algo139+ ${CMAKE_CURRENT_SOURCE_DIR}/ops/scatter/algorithm
140- ${CMAKE_CURRENT_SOURCE_DIR}/ops/scatter/algo/template140+ ${CMAKE_CURRENT_SOURCE_DIR}/ops/scatter/algorithm/executor
141- ${CMAKE_CURRENT_SOURCE_DIR}/ops/scatter/executor
142 ${CMAKE_CURRENT_SOURCE_DIR}/ops/scatter/selector141 ${CMAKE_CURRENT_SOURCE_DIR}/ops/scatter/selector
143- ${CMAKE_CURRENT_SOURCE_DIR}/ops/scatter/template142+ ${CMAKE_CURRENT_SOURCE_DIR}/ops/scatter/algorithm/template
144- ${CMAKE_CURRENT_SOURCE_DIR}/ops/scatter/template/aicpu143+ ${CMAKE_CURRENT_SOURCE_DIR}/ops/scatter/algorithm/template/aicpu
145- ${CMAKE_CURRENT_SOURCE_DIR}/ops/scatter/template/aiv144+ ${CMAKE_CURRENT_SOURCE_DIR}/ops/scatter/algorithm/template/aiv
146- ${CMAKE_CURRENT_SOURCE_DIR}/ops/scatter/template/ccu145+ ${CMAKE_CURRENT_SOURCE_DIR}/ops/scatter/algorithm/template/ccu
147- ${CMAKE_CURRENT_SOURCE_DIR}/ops/scatter/template/ccu/kernel146+ ${CMAKE_CURRENT_SOURCE_DIR}/ops/scatter/algorithm/template/ccu/kernel
148 147 
149 ${CMAKE_CURRENT_SOURCE_DIR}/ops/all_gather148 ${CMAKE_CURRENT_SOURCE_DIR}/ops/all_gather
150- ${CMAKE_CURRENT_SOURCE_DIR}/ops/all_gather/executor149+ ${CMAKE_CURRENT_SOURCE_DIR}/ops/all_gather/algorithm/executor
151 ${CMAKE_CURRENT_SOURCE_DIR}/ops/all_gather/selector150 ${CMAKE_CURRENT_SOURCE_DIR}/ops/all_gather/selector
152- ${CMAKE_CURRENT_SOURCE_DIR}/ops/all_gather/template151+ ${CMAKE_CURRENT_SOURCE_DIR}/ops/all_gather/algorithm/template
153- ${CMAKE_CURRENT_SOURCE_DIR}/ops/all_gather/template/aicpu152+ ${CMAKE_CURRENT_SOURCE_DIR}/ops/all_gather/algorithm/template/aicpu
154- ${CMAKE_CURRENT_SOURCE_DIR}/ops/all_gather/template/aiv153+ ${CMAKE_CURRENT_SOURCE_DIR}/ops/all_gather/algorithm/template/aiv
155- ${CMAKE_CURRENT_SOURCE_DIR}/ops/all_gather/template/ccu154+ ${CMAKE_CURRENT_SOURCE_DIR}/ops/all_gather/algorithm/template/ccu
156- ${CMAKE_CURRENT_SOURCE_DIR}/ops/all_gather/template/ccu/kernel155+ ${CMAKE_CURRENT_SOURCE_DIR}/ops/all_gather/algorithm/template/ccu/kernel
157 156 
158 ${CMAKE_CURRENT_SOURCE_DIR}/ops/barrier157 ${CMAKE_CURRENT_SOURCE_DIR}/ops/barrier
159- ${CMAKE_CURRENT_SOURCE_DIR}/ops/barrier/executor158+ ${CMAKE_CURRENT_SOURCE_DIR}/ops/barrier/algorithm/executor
160 ${CMAKE_CURRENT_SOURCE_DIR}/ops/barrier/selector159 ${CMAKE_CURRENT_SOURCE_DIR}/ops/barrier/selector
161- ${CMAKE_CURRENT_SOURCE_DIR}/ops/barrier/template160+ ${CMAKE_CURRENT_SOURCE_DIR}/ops/barrier/algorithm/template
162- ${CMAKE_CURRENT_SOURCE_DIR}/ops/barrier/template/aicpu161+ ${CMAKE_CURRENT_SOURCE_DIR}/ops/barrier/algorithm/template/aicpu
163 162 
164 ${CMAKE_CURRENT_SOURCE_DIR}/ops/all_gather_v163 ${CMAKE_CURRENT_SOURCE_DIR}/ops/all_gather_v
165- ${CMAKE_CURRENT_SOURCE_DIR}/ops/all_gather_v/executor164+ ${CMAKE_CURRENT_SOURCE_DIR}/ops/all_gather_v/algorithm/executor
166 ${CMAKE_CURRENT_SOURCE_DIR}/ops/all_gather_v/selector165 ${CMAKE_CURRENT_SOURCE_DIR}/ops/all_gather_v/selector
167- ${CMAKE_CURRENT_SOURCE_DIR}/ops/all_gather_v/template166+ ${CMAKE_CURRENT_SOURCE_DIR}/ops/all_gather_v/algorithm/template
168- ${CMAKE_CURRENT_SOURCE_DIR}/ops/all_gather_v/template/aicpu167+ ${CMAKE_CURRENT_SOURCE_DIR}/ops/all_gather_v/algorithm/template/aicpu
169- ${CMAKE_CURRENT_SOURCE_DIR}/ops/all_gather_v/template/ccu168+ ${CMAKE_CURRENT_SOURCE_DIR}/ops/all_gather_v/algorithm/template/ccu
170- ${CMAKE_CURRENT_SOURCE_DIR}/ops/all_gather_v/template/ccu/kernel169+ ${CMAKE_CURRENT_SOURCE_DIR}/ops/all_gather_v/algorithm/template/ccu/kernel
171 170 
172 ${CMAKE_CURRENT_SOURCE_DIR}/ops/all_to_all_v171 ${CMAKE_CURRENT_SOURCE_DIR}/ops/all_to_all_v
173- ${CMAKE_CURRENT_SOURCE_DIR}/ops/all_to_all_v/executor172+ ${CMAKE_CURRENT_SOURCE_DIR}/ops/all_to_all_v/algorithm/executor
174 ${CMAKE_CURRENT_SOURCE_DIR}/ops/all_to_all_v/selector173 ${CMAKE_CURRENT_SOURCE_DIR}/ops/all_to_all_v/selector
175- ${CMAKE_CURRENT_SOURCE_DIR}/ops/all_to_all_v/template174+ ${CMAKE_CURRENT_SOURCE_DIR}/ops/all_to_all_v/algorithm/template
176- ${CMAKE_CURRENT_SOURCE_DIR}/ops/all_to_all_v/template/aicpu175+ ${CMAKE_CURRENT_SOURCE_DIR}/ops/all_to_all_v/algorithm/template/aicpu
177- ${CMAKE_CURRENT_SOURCE_DIR}/ops/all_to_all_v/template/aiv176+ ${CMAKE_CURRENT_SOURCE_DIR}/ops/all_to_all_v/algorithm/template/aiv
178- ${CMAKE_CURRENT_SOURCE_DIR}/ops/all_to_all_v/template/ccu177+ ${CMAKE_CURRENT_SOURCE_DIR}/ops/all_to_all_v/algorithm/template/ccu
179- ${CMAKE_CURRENT_SOURCE_DIR}/ops/all_to_all_v/template/ccu/kernel178+ ${CMAKE_CURRENT_SOURCE_DIR}/ops/all_to_all_v/algorithm/template/ccu/kernel
180 179 
181 ${CMAKE_CURRENT_SOURCE_DIR}/ops/all_reduce180 ${CMAKE_CURRENT_SOURCE_DIR}/ops/all_reduce
182- ${CMAKE_CURRENT_SOURCE_DIR}/ops/all_reduce/executor181+ ${CMAKE_CURRENT_SOURCE_DIR}/ops/all_reduce/algorithm/executor
183 ${CMAKE_CURRENT_SOURCE_DIR}/ops/all_reduce/selector182 ${CMAKE_CURRENT_SOURCE_DIR}/ops/all_reduce/selector
184- ${CMAKE_CURRENT_SOURCE_DIR}/ops/all_reduce/template183+ ${CMAKE_CURRENT_SOURCE_DIR}/ops/all_reduce/algorithm/template
185- ${CMAKE_CURRENT_SOURCE_DIR}/ops/all_reduce/template/aicpu184+ ${CMAKE_CURRENT_SOURCE_DIR}/ops/all_reduce/algorithm/template/aicpu
186- ${CMAKE_CURRENT_SOURCE_DIR}/ops/all_reduce/template/aiv185+ ${CMAKE_CURRENT_SOURCE_DIR}/ops/all_reduce/algorithm/template/aiv
187- ${CMAKE_CURRENT_SOURCE_DIR}/ops/all_reduce/template/ccu186+ ${CMAKE_CURRENT_SOURCE_DIR}/ops/all_reduce/algorithm/template/ccu
188- ${CMAKE_CURRENT_SOURCE_DIR}/ops/all_reduce/template/ccu/kernel187+ ${CMAKE_CURRENT_SOURCE_DIR}/ops/all_reduce/algorithm/template/ccu/kernel
189 188 
190 ${CMAKE_CURRENT_SOURCE_DIR}/ops/send189 ${CMAKE_CURRENT_SOURCE_DIR}/ops/send
191- ${CMAKE_CURRENT_SOURCE_DIR}/ops/send/executor190+ ${CMAKE_CURRENT_SOURCE_DIR}/ops/send/algorithm/executor
192 ${CMAKE_CURRENT_SOURCE_DIR}/ops/send/selector191 ${CMAKE_CURRENT_SOURCE_DIR}/ops/send/selector
193- ${CMAKE_CURRENT_SOURCE_DIR}/ops/send/template192+ ${CMAKE_CURRENT_SOURCE_DIR}/ops/send/algorithm/template
194- ${CMAKE_CURRENT_SOURCE_DIR}/ops/send/template/host_nic193+ ${CMAKE_CURRENT_SOURCE_DIR}/ops/send/algorithm/template/host_nic
195 194 
196 ${CMAKE_CURRENT_SOURCE_DIR}/ops/recv195 ${CMAKE_CURRENT_SOURCE_DIR}/ops/recv
197- ${CMAKE_CURRENT_SOURCE_DIR}/ops/recv/executor196+ ${CMAKE_CURRENT_SOURCE_DIR}/ops/recv/algorithm/executor
198 ${CMAKE_CURRENT_SOURCE_DIR}/ops/recv/selector197 ${CMAKE_CURRENT_SOURCE_DIR}/ops/recv/selector
199- ${CMAKE_CURRENT_SOURCE_DIR}/ops/recv/template198+ ${CMAKE_CURRENT_SOURCE_DIR}/ops/recv/algorithm/template
200- ${CMAKE_CURRENT_SOURCE_DIR}/ops/recv/template/host_nic199+ ${CMAKE_CURRENT_SOURCE_DIR}/ops/recv/algorithm/template/host_nic
201 200 
202 ${CMAKE_CURRENT_SOURCE_DIR}/ops/reduce201 ${CMAKE_CURRENT_SOURCE_DIR}/ops/reduce
203- ${CMAKE_CURRENT_SOURCE_DIR}/ops/reduce/executor202+ ${CMAKE_CURRENT_SOURCE_DIR}/ops/reduce/algorithm/executor
204 ${CMAKE_CURRENT_SOURCE_DIR}/ops/reduce/selector203 ${CMAKE_CURRENT_SOURCE_DIR}/ops/reduce/selector
205- ${CMAKE_CURRENT_SOURCE_DIR}/ops/reduce/template204+ ${CMAKE_CURRENT_SOURCE_DIR}/ops/reduce/algorithm/template
206- ${CMAKE_CURRENT_SOURCE_DIR}/ops/reduce/template/aicpu205+ ${CMAKE_CURRENT_SOURCE_DIR}/ops/reduce/algorithm/template/aicpu
207- ${CMAKE_CURRENT_SOURCE_DIR}/ops/reduce/template/aiv206+ ${CMAKE_CURRENT_SOURCE_DIR}/ops/reduce/algorithm/template/aiv
208- ${CMAKE_CURRENT_SOURCE_DIR}/ops/reduce/template/ccu207+ ${CMAKE_CURRENT_SOURCE_DIR}/ops/reduce/algorithm/template/ccu
209- ${CMAKE_CURRENT_SOURCE_DIR}/ops/reduce/template/ccu/kernel208+ ${CMAKE_CURRENT_SOURCE_DIR}/ops/reduce/algorithm/template/ccu/kernel
210 209 
211 ${CMAKE_CURRENT_SOURCE_DIR}/ops/batch_send_recv210 ${CMAKE_CURRENT_SOURCE_DIR}/ops/batch_send_recv
212- ${CMAKE_CURRENT_SOURCE_DIR}/ops/batch_send_recv/executor211+ ${CMAKE_CURRENT_SOURCE_DIR}/ops/batch_send_recv/algorithm/executor
213 ${CMAKE_CURRENT_SOURCE_DIR}/ops/batch_send_recv/selector212 ${CMAKE_CURRENT_SOURCE_DIR}/ops/batch_send_recv/selector
214- ${CMAKE_CURRENT_SOURCE_DIR}/ops/batch_send_recv/template213+ ${CMAKE_CURRENT_SOURCE_DIR}/ops/batch_send_recv/algorithm/template
215)214)
216 215 
217# [HCCL-ALGO-Plugin] 仅当显式开启插件开关时，才把 algo_plugin 目录加入包含路径，216# [HCCL-ALGO-Plugin] 仅当显式开启插件开关时，才把 algo_plugin 目录加入包含路径，
Msrc/ops/all_gather/CMakeLists.txt+2-3
@@ -8,13 +8,12 @@
8# See LICENSE in the root of the software repository for the full text of the License.8# See LICENSE in the root of the software repository for the full text of the License.
9# -----------------------------------------------------------------------------------------------------------9# -----------------------------------------------------------------------------------------------------------
10 10 
11-add_subdirectory(executor)11+add_subdirectory(algorithm)
12add_subdirectory(selector)12add_subdirectory(selector)
13-add_subdirectory(template)
14add_subdirectory(op_graph)13add_subdirectory(op_graph)
15 14 
16set(src_list15set(src_list
17- ${CMAKE_CURRENT_SOURCE_DIR}/all_gather_op.cc16+ ${CMAKE_CURRENT_SOURCE_DIR}/all_gather.cc
18)17)
19 18 
20if(TARGET hccl)19if(TARGET hccl)
Rsrc/ops/scatter/template/CMakeLists.txt→src/ops/all_gather/algorithm/CMakeLists.txt+3-6
@@ -1,5 +1,5 @@
1# -----------------------------------------------------------------------------------------------------------1# -----------------------------------------------------------------------------------------------------------
2-# Copyright (c) 2025 Huawei Technologies Co., Ltd.2+# Copyright (c) 2026 Huawei Technologies Co., Ltd.
3# This program is free software, you can redistribute it and/or modify it under the terms and conditions of3# This program is free software, you can redistribute it and/or modify it under the terms and conditions of
4# CANN Open Software License Agreement Version 2.0 (the "License").4# CANN Open Software License Agreement Version 2.0 (the "License").
5# Please refer to the License for details. You may not use this file except in compliance with the License.5# Please refer to the License for details. You may not use this file except in compliance with the License.
@@ -8,8 +8,5 @@
8# See LICENSE in the root of the software repository for the full text of the License.8# See LICENSE in the root of the software repository for the full text of the License.
9# -----------------------------------------------------------------------------------------------------------9# -----------------------------------------------------------------------------------------------------------
10 10 
11-add_subdirectory(aicpu)11+add_subdirectory(executor)
12-add_subdirectory(aiv)12+add_subdirectory(template)
13-if(NOT HCCL_CANN_COMPAT_850)
14- add_subdirectory(ccu)
15-endif()
Rsrc/ops/all_gather/executor/CMakeLists.txt→src/ops/all_gather/algorithm/executor/CMakeLists.txt+0-0
文件重命名但无更改。
Rsrc/ops/all_gather/executor/ins_v2_all_gather_concurrent_executor.cc→src/ops/all_gather/algorithm/executor/ins_v2_all_gather_concurrent_executor.cc+0-0
文件重命名但无更改。
Rsrc/ops/all_gather/executor/ins_v2_all_gather_concurrent_executor.h→src/ops/all_gather/algorithm/executor/ins_v2_all_gather_concurrent_executor.h+0-0
文件重命名但无更改。
Rsrc/ops/all_gather/executor/ins_v2_all_gather_omnipipe_2d_executor.cc→src/ops/all_gather/algorithm/executor/ins_v2_all_gather_omnipipe_2d_executor.cc+0-0
文件重命名但无更改。
Rsrc/ops/all_gather/executor/ins_v2_all_gather_omnipipe_2d_executor.h→src/ops/all_gather/algorithm/executor/ins_v2_all_gather_omnipipe_2d_executor.h+0-0
文件重命名但无更改。
Rsrc/ops/all_gather/executor/ins_v2_all_gather_omnipipe_executor.cc→src/ops/all_gather/algorithm/executor/ins_v2_all_gather_omnipipe_executor.cc+0-0
文件重命名但无更改。
Rsrc/ops/all_gather/executor/ins_v2_all_gather_omnipipe_executor.h→src/ops/all_gather/algorithm/executor/ins_v2_all_gather_omnipipe_executor.h+0-0
文件重命名但无更改。
Rsrc/ops/all_gather/executor/ins_v2_all_gather_parallel_executor.cc→src/ops/all_gather/algorithm/executor/ins_v2_all_gather_parallel_executor.cc+0-0
文件重命名但无更改。
Rsrc/ops/all_gather/executor/ins_v2_all_gather_parallel_executor.h→src/ops/all_gather/algorithm/executor/ins_v2_all_gather_parallel_executor.h+0-0
文件重命名但无更改。
Rsrc/ops/all_gather/executor/ins_v2_all_gather_sequence_executor.cc→src/ops/all_gather/algorithm/executor/ins_v2_all_gather_sequence_executor.cc+0-0
文件重命名但无更改。
Rsrc/ops/all_gather/executor/ins_v2_all_gather_sequence_executor.h→src/ops/all_gather/algorithm/executor/ins_v2_all_gather_sequence_executor.h+0-0
文件重命名但无更改。
Rsrc/ops/all_gather/executor/ins_v2_all_gather_sequence_executor_3level.cc→src/ops/all_gather/algorithm/executor/ins_v2_all_gather_sequence_executor_3level.cc+0-0
文件重命名但无更改。
Rsrc/ops/all_gather/executor/ins_v2_all_gather_sequence_executor_3level.h→src/ops/all_gather/algorithm/executor/ins_v2_all_gather_sequence_executor_3level.h+0-0
文件重命名但无更改。
Rsrc/ops/all_gather/executor/ins_v2_all_gather_sequence_executor_aicpu.cc→src/ops/all_gather/algorithm/executor/ins_v2_all_gather_sequence_executor_aicpu.cc+0-0
文件重命名但无更改。
Rsrc/ops/all_gather/executor/ins_v2_all_gather_sequence_executor_aicpu.h→src/ops/all_gather/algorithm/executor/ins_v2_all_gather_sequence_executor_aicpu.h+0-0
文件重命名但无更改。
Rsrc/ops/all_gather/executor/ins_v2_all_gather_sole_executor.cc→src/ops/all_gather/algorithm/executor/ins_v2_all_gather_sole_executor.cc+0-0
文件重命名但无更改。
Rsrc/ops/all_gather/executor/ins_v2_all_gather_sole_executor.h→src/ops/all_gather/algorithm/executor/ins_v2_all_gather_sole_executor.h+0-0
文件重命名但无更改。
Rsrc/ops/all_gather/template/CMakeLists.txt→src/ops/all_gather/algorithm/template/CMakeLists.txt+0-0
文件重命名但无更改。
Rsrc/ops/all_gather/template/aicpu/CMakeLists.txt→src/ops/all_gather/algorithm/template/aicpu/CMakeLists.txt+0-0
文件重命名但无更改。
Rsrc/ops/all_gather/template/aicpu/ins_temp_all_gather_mesh_1D.cc→src/ops/all_gather/algorithm/template/aicpu/ins_temp_all_gather_mesh_1D.cc+0-0
文件重命名但无更改。
Rsrc/ops/all_gather/template/aicpu/ins_temp_all_gather_mesh_1D.h→src/ops/all_gather/algorithm/template/aicpu/ins_temp_all_gather_mesh_1D.h+0-0
文件重命名但无更改。
Rsrc/ops/all_gather/template/aicpu/ins_temp_all_gather_mesh_1D_Z_axis_detour.cc→src/ops/all_gather/algorithm/template/aicpu/ins_temp_all_gather_mesh_1D_Z_axis_detour.cc+0-0
文件重命名但无更改。
Rsrc/ops/all_gather/template/aicpu/ins_temp_all_gather_mesh_1D_Z_axis_detour.h→src/ops/all_gather/algorithm/template/aicpu/ins_temp_all_gather_mesh_1D_Z_axis_detour.h+0-0
文件重命名但无更改。
Rsrc/ops/all_gather/template/aicpu/ins_temp_all_gather_nhr.cc→src/ops/all_gather/algorithm/template/aicpu/ins_temp_all_gather_nhr.cc+0-0
文件重命名但无更改。
Rsrc/ops/all_gather/template/aicpu/ins_temp_all_gather_nhr.h→src/ops/all_gather/algorithm/template/aicpu/ins_temp_all_gather_nhr.h+0-0
文件重命名但无更改。
Rsrc/ops/all_gather/template/aicpu/ins_temp_all_gather_nhr_dpu.cc→src/ops/all_gather/algorithm/template/aicpu/ins_temp_all_gather_nhr_dpu.cc+0-0
文件重命名但无更改。
Rsrc/ops/all_gather/template/aicpu/ins_temp_all_gather_nhr_dpu.h→src/ops/all_gather/algorithm/template/aicpu/ins_temp_all_gather_nhr_dpu.h+0-0
文件重命名但无更改。
Rsrc/ops/all_gather/template/aicpu/ins_temp_all_gather_omnipipe_mesh_1D.cc→src/ops/all_gather/algorithm/template/aicpu/ins_temp_all_gather_omnipipe_mesh_1D.cc+0-0
文件重命名但无更改。
Rsrc/ops/all_gather/template/aicpu/ins_temp_all_gather_omnipipe_mesh_1D.h→src/ops/all_gather/algorithm/template/aicpu/ins_temp_all_gather_omnipipe_mesh_1D.h+0-0
文件重命名但无更改。
Rsrc/ops/all_gather/template/aicpu/ins_temp_all_gather_omnipipe_nhr.cc→src/ops/all_gather/algorithm/template/aicpu/ins_temp_all_gather_omnipipe_nhr.cc+0-0
文件重命名但无更改。
Rsrc/ops/all_gather/template/aicpu/ins_temp_all_gather_omnipipe_nhr.h→src/ops/all_gather/algorithm/template/aicpu/ins_temp_all_gather_omnipipe_nhr.h+0-0
文件重命名但无更改。
Rsrc/ops/all_gather/template/aicpu/ins_temp_all_gather_omnipipe_nhr_dpu.cc→src/ops/all_gather/algorithm/template/aicpu/ins_temp_all_gather_omnipipe_nhr_dpu.cc+0-0
文件重命名但无更改。
Rsrc/ops/all_gather/template/aicpu/ins_temp_all_gather_omnipipe_nhr_dpu.h→src/ops/all_gather/algorithm/template/aicpu/ins_temp_all_gather_omnipipe_nhr_dpu.h+0-0
文件重命名但无更改。
Rsrc/ops/all_gather/template/aiv/CMakeLists.txt→src/ops/all_gather/algorithm/template/aiv/CMakeLists.txt+0-0
文件重命名但无更改。
Rsrc/ops/all_gather/template/aiv/aiv_temp_all_gather_mesh_1D.cc→src/ops/all_gather/algorithm/template/aiv/aiv_temp_all_gather_mesh_1D.cc+0-0
文件重命名但无更改。
Rsrc/ops/all_gather/template/aiv/aiv_temp_all_gather_mesh_1D.h→src/ops/all_gather/algorithm/template/aiv/aiv_temp_all_gather_mesh_1D.h+0-0
文件重命名但无更改。
Rsrc/ops/all_gather/template/aiv/kernel/CMakeLists.txt→src/ops/all_gather/algorithm/template/aiv/kernel/CMakeLists.txt+48-48
@@ -21,8 +21,8 @@ add_custom_command(OUTPUT ${CMAKE_CURRENT_BINARY_DIR}/hccl_aiv_all_gather_op_910
21 --cce-aicore-only -mllvm -cce-aicore-function-stack-size=0x8000 -mllvm -cce-aicore-stack-size=0x800021 --cce-aicore-only -mllvm -cce-aicore-function-stack-size=0x8000 -mllvm -cce-aicore-stack-size=0x8000
22 "-I$<JOIN:$<TARGET_PROPERTY:asc_kernel_headers,INTERFACE_INCLUDE_DIRECTORIES>,;-I>"22 "-I$<JOIN:$<TARGET_PROPERTY:asc_kernel_headers,INTERFACE_INCLUDE_DIRECTORIES>,;-I>"
23 "-I$<JOIN:$<TARGET_PROPERTY:kernel_tiling_headers,INTERFACE_INCLUDE_DIRECTORIES>,;-I>"23 "-I$<JOIN:$<TARGET_PROPERTY:kernel_tiling_headers,INTERFACE_INCLUDE_DIRECTORIES>,;-I>"
24- -I${OPS_BASE_DIR}/src/ops/op_common/template/aiv/24+ -I${OPS_BASE_DIR}/src/ops/op_common/algorithm/template/aiv/
25- -I${OPS_BASE_DIR}/src/ops/op_common/template/aiv/aiv_interface/25+ -I${OPS_BASE_DIR}/src/ops/op_common/algorithm/template/aiv/aiv_interface/
26 -mllvm -cce-aicore-record-overflow=false -mllvm -cce-aicore-addr-transform -mllvm26 -mllvm -cce-aicore-record-overflow=false -mllvm -cce-aicore-addr-transform -mllvm
27 --cce-aicore-jump-expand=true -std=c++17 -fstack-protector-all27 --cce-aicore-jump-expand=true -std=c++17 -fstack-protector-all
28 COMMAND ${CCE_DIR}/ld.lld -m aicorelinux -Ttext=0 ${CMAKE_CURRENT_BINARY_DIR}/hccl_aiv_all_gather_op_910_95_tmp.o -static -o ${CMAKE_CURRENT_BINARY_DIR}/hccl_aiv_all_gather_op_910_95.o28 COMMAND ${CCE_DIR}/ld.lld -m aicorelinux -Ttext=0 ${CMAKE_CURRENT_BINARY_DIR}/hccl_aiv_all_gather_op_910_95_tmp.o -static -o ${CMAKE_CURRENT_BINARY_DIR}/hccl_aiv_all_gather_op_910_95.o
@@ -48,8 +48,8 @@ add_custom_command(OUTPUT ${CMAKE_CURRENT_BINARY_DIR}/hccl_ag_superkernel_mesh_1
48 -DHCCL_DTYPE_INT8 -Dsk_allgather_mesh_1d=sk_allgather_mesh_1d_int848 -DHCCL_DTYPE_INT8 -Dsk_allgather_mesh_1d=sk_allgather_mesh_1d_int8
49 "-I$<JOIN:$<TARGET_PROPERTY:asc_kernel_headers,INTERFACE_INCLUDE_DIRECTORIES>,;-I>"49 "-I$<JOIN:$<TARGET_PROPERTY:asc_kernel_headers,INTERFACE_INCLUDE_DIRECTORIES>,;-I>"
50 "-I$<JOIN:$<TARGET_PROPERTY:kernel_tiling_headers,INTERFACE_INCLUDE_DIRECTORIES>,;-I>"50 "-I$<JOIN:$<TARGET_PROPERTY:kernel_tiling_headers,INTERFACE_INCLUDE_DIRECTORIES>,;-I>"
51- -I${OPS_BASE_DIR}/src/ops/op_common/template/aiv/51+ -I${OPS_BASE_DIR}/src/ops/op_common/algorithm/template/aiv/
52- -I${OPS_BASE_DIR}/src/ops/op_common/template/aiv/aiv_interface/52+ -I${OPS_BASE_DIR}/src/ops/op_common/algorithm/template/aiv/aiv_interface/
53 -I${CMAKE_CURRENT_SOURCE_DIR}/aiv_interface/53 -I${CMAKE_CURRENT_SOURCE_DIR}/aiv_interface/
54 -mllvm -cce-aicore-record-overflow=false -mllvm -cce-aicore-addr-transform -mllvm54 -mllvm -cce-aicore-record-overflow=false -mllvm -cce-aicore-addr-transform -mllvm
55 --cce-aicore-jump-expand=true -std=c++17 -fstack-protector-all55 --cce-aicore-jump-expand=true -std=c++17 -fstack-protector-all
@@ -69,8 +69,8 @@ add_custom_command(OUTPUT ${CMAKE_CURRENT_BINARY_DIR}/hccl_ag_superkernel_mesh_1
69 -DHCCL_DTYPE_INT16 -Dsk_allgather_mesh_1d=sk_allgather_mesh_1d_int1669 -DHCCL_DTYPE_INT16 -Dsk_allgather_mesh_1d=sk_allgather_mesh_1d_int16
70 "-I$<JOIN:$<TARGET_PROPERTY:asc_kernel_headers,INTERFACE_INCLUDE_DIRECTORIES>,;-I>"70 "-I$<JOIN:$<TARGET_PROPERTY:asc_kernel_headers,INTERFACE_INCLUDE_DIRECTORIES>,;-I>"
71 "-I$<JOIN:$<TARGET_PROPERTY:kernel_tiling_headers,INTERFACE_INCLUDE_DIRECTORIES>,;-I>"71 "-I$<JOIN:$<TARGET_PROPERTY:kernel_tiling_headers,INTERFACE_INCLUDE_DIRECTORIES>,;-I>"
72- -I${OPS_BASE_DIR}/src/ops/op_common/template/aiv/72+ -I${OPS_BASE_DIR}/src/ops/op_common/algorithm/template/aiv/
73- -I${OPS_BASE_DIR}/src/ops/op_common/template/aiv/aiv_interface/73+ -I${OPS_BASE_DIR}/src/ops/op_common/algorithm/template/aiv/aiv_interface/
74 -I${CMAKE_CURRENT_SOURCE_DIR}/aiv_interface/74 -I${CMAKE_CURRENT_SOURCE_DIR}/aiv_interface/
75 -mllvm -cce-aicore-record-overflow=false -mllvm -cce-aicore-addr-transform -mllvm75 -mllvm -cce-aicore-record-overflow=false -mllvm -cce-aicore-addr-transform -mllvm
76 --cce-aicore-jump-expand=true -std=c++17 -fstack-protector-all76 --cce-aicore-jump-expand=true -std=c++17 -fstack-protector-all
@@ -90,8 +90,8 @@ add_custom_command(OUTPUT ${CMAKE_CURRENT_BINARY_DIR}/hccl_ag_superkernel_mesh_1
90 -DHCCL_DTYPE_INT32 -Dsk_allgather_mesh_1d=sk_allgather_mesh_1d_int3290 -DHCCL_DTYPE_INT32 -Dsk_allgather_mesh_1d=sk_allgather_mesh_1d_int32
91 "-I$<JOIN:$<TARGET_PROPERTY:asc_kernel_headers,INTERFACE_INCLUDE_DIRECTORIES>,;-I>"91 "-I$<JOIN:$<TARGET_PROPERTY:asc_kernel_headers,INTERFACE_INCLUDE_DIRECTORIES>,;-I>"
92 "-I$<JOIN:$<TARGET_PROPERTY:kernel_tiling_headers,INTERFACE_INCLUDE_DIRECTORIES>,;-I>"92 "-I$<JOIN:$<TARGET_PROPERTY:kernel_tiling_headers,INTERFACE_INCLUDE_DIRECTORIES>,;-I>"
93- -I${OPS_BASE_DIR}/src/ops/op_common/template/aiv/93+ -I${OPS_BASE_DIR}/src/ops/op_common/algorithm/template/aiv/
94- -I${OPS_BASE_DIR}/src/ops/op_common/template/aiv/aiv_interface/94+ -I${OPS_BASE_DIR}/src/ops/op_common/algorithm/template/aiv/aiv_interface/
95 -I${CMAKE_CURRENT_SOURCE_DIR}/aiv_interface/95 -I${CMAKE_CURRENT_SOURCE_DIR}/aiv_interface/
96 -mllvm -cce-aicore-record-overflow=false -mllvm -cce-aicore-addr-transform -mllvm96 -mllvm -cce-aicore-record-overflow=false -mllvm -cce-aicore-addr-transform -mllvm
97 --cce-aicore-jump-expand=true -std=c++17 -fstack-protector-all97 --cce-aicore-jump-expand=true -std=c++17 -fstack-protector-all
@@ -111,8 +111,8 @@ add_custom_command(OUTPUT ${CMAKE_CURRENT_BINARY_DIR}/hccl_ag_superkernel_mesh_1
111 -DHCCL_DTYPE_FP16 -Dsk_allgather_mesh_1d=sk_allgather_mesh_1d_float16111 -DHCCL_DTYPE_FP16 -Dsk_allgather_mesh_1d=sk_allgather_mesh_1d_float16
112 "-I$<JOIN:$<TARGET_PROPERTY:asc_kernel_headers,INTERFACE_INCLUDE_DIRECTORIES>,;-I>"112 "-I$<JOIN:$<TARGET_PROPERTY:asc_kernel_headers,INTERFACE_INCLUDE_DIRECTORIES>,;-I>"
113 "-I$<JOIN:$<TARGET_PROPERTY:kernel_tiling_headers,INTERFACE_INCLUDE_DIRECTORIES>,;-I>"113 "-I$<JOIN:$<TARGET_PROPERTY:kernel_tiling_headers,INTERFACE_INCLUDE_DIRECTORIES>,;-I>"
114- -I${OPS_BASE_DIR}/src/ops/op_common/template/aiv/114+ -I${OPS_BASE_DIR}/src/ops/op_common/algorithm/template/aiv/
115- -I${OPS_BASE_DIR}/src/ops/op_common/template/aiv/aiv_interface/115+ -I${OPS_BASE_DIR}/src/ops/op_common/algorithm/template/aiv/aiv_interface/
116 -I${CMAKE_CURRENT_SOURCE_DIR}/aiv_interface/116 -I${CMAKE_CURRENT_SOURCE_DIR}/aiv_interface/
117 -mllvm -cce-aicore-record-overflow=false -mllvm -cce-aicore-addr-transform -mllvm117 -mllvm -cce-aicore-record-overflow=false -mllvm -cce-aicore-addr-transform -mllvm
118 --cce-aicore-jump-expand=true -std=c++17 -fstack-protector-all118 --cce-aicore-jump-expand=true -std=c++17 -fstack-protector-all
@@ -132,8 +132,8 @@ add_custom_command(OUTPUT ${CMAKE_CURRENT_BINARY_DIR}/hccl_ag_superkernel_mesh_1
132 -DHCCL_DTYPE_FP32 -Dsk_allgather_mesh_1d=sk_allgather_mesh_1d_float32132 -DHCCL_DTYPE_FP32 -Dsk_allgather_mesh_1d=sk_allgather_mesh_1d_float32
133 "-I$<JOIN:$<TARGET_PROPERTY:asc_kernel_headers,INTERFACE_INCLUDE_DIRECTORIES>,;-I>"133 "-I$<JOIN:$<TARGET_PROPERTY:asc_kernel_headers,INTERFACE_INCLUDE_DIRECTORIES>,;-I>"
134 "-I$<JOIN:$<TARGET_PROPERTY:kernel_tiling_headers,INTERFACE_INCLUDE_DIRECTORIES>,;-I>"134 "-I$<JOIN:$<TARGET_PROPERTY:kernel_tiling_headers,INTERFACE_INCLUDE_DIRECTORIES>,;-I>"
135- -I${OPS_BASE_DIR}/src/ops/op_common/template/aiv/135+ -I${OPS_BASE_DIR}/src/ops/op_common/algorithm/template/aiv/
136- -I${OPS_BASE_DIR}/src/ops/op_common/template/aiv/aiv_interface/136+ -I${OPS_BASE_DIR}/src/ops/op_common/algorithm/template/aiv/aiv_interface/
137 -I${CMAKE_CURRENT_SOURCE_DIR}/aiv_interface/137 -I${CMAKE_CURRENT_SOURCE_DIR}/aiv_interface/
138 -mllvm -cce-aicore-record-overflow=false -mllvm -cce-aicore-addr-transform -mllvm138 -mllvm -cce-aicore-record-overflow=false -mllvm -cce-aicore-addr-transform -mllvm
139 --cce-aicore-jump-expand=true -std=c++17 -fstack-protector-all139 --cce-aicore-jump-expand=true -std=c++17 -fstack-protector-all
@@ -153,8 +153,8 @@ add_custom_command(OUTPUT ${CMAKE_CURRENT_BINARY_DIR}/hccl_ag_superkernel_mesh_1
153 -DHCCL_DTYPE_BFP16 -Dsk_allgather_mesh_1d=sk_allgather_mesh_1d_bfloat16153 -DHCCL_DTYPE_BFP16 -Dsk_allgather_mesh_1d=sk_allgather_mesh_1d_bfloat16
154 "-I$<JOIN:$<TARGET_PROPERTY:asc_kernel_headers,INTERFACE_INCLUDE_DIRECTORIES>,;-I>"154 "-I$<JOIN:$<TARGET_PROPERTY:asc_kernel_headers,INTERFACE_INCLUDE_DIRECTORIES>,;-I>"
155 "-I$<JOIN:$<TARGET_PROPERTY:kernel_tiling_headers,INTERFACE_INCLUDE_DIRECTORIES>,;-I>"155 "-I$<JOIN:$<TARGET_PROPERTY:kernel_tiling_headers,INTERFACE_INCLUDE_DIRECTORIES>,;-I>"
156- -I${OPS_BASE_DIR}/src/ops/op_common/template/aiv/156+ -I${OPS_BASE_DIR}/src/ops/op_common/algorithm/template/aiv/
157- -I${OPS_BASE_DIR}/src/ops/op_common/template/aiv/aiv_interface/157+ -I${OPS_BASE_DIR}/src/ops/op_common/algorithm/template/aiv/aiv_interface/
158 -I${CMAKE_CURRENT_SOURCE_DIR}/aiv_interface/158 -I${CMAKE_CURRENT_SOURCE_DIR}/aiv_interface/
159 -mllvm -cce-aicore-record-overflow=false -mllvm -cce-aicore-addr-transform -mllvm159 -mllvm -cce-aicore-record-overflow=false -mllvm -cce-aicore-addr-transform -mllvm
160 --cce-aicore-jump-expand=true -std=c++17 -fstack-protector-all160 --cce-aicore-jump-expand=true -std=c++17 -fstack-protector-all
@@ -174,8 +174,8 @@ add_custom_command(OUTPUT ${CMAKE_CURRENT_BINARY_DIR}/hccl_ag_superkernel_mesh_1
174 -DHCCL_DTYPE_UINT8 -Dsk_allgather_mesh_1d=sk_allgather_mesh_1d_uint8174 -DHCCL_DTYPE_UINT8 -Dsk_allgather_mesh_1d=sk_allgather_mesh_1d_uint8
175 "-I$<JOIN:$<TARGET_PROPERTY:asc_kernel_headers,INTERFACE_INCLUDE_DIRECTORIES>,;-I>"175 "-I$<JOIN:$<TARGET_PROPERTY:asc_kernel_headers,INTERFACE_INCLUDE_DIRECTORIES>,;-I>"
176 "-I$<JOIN:$<TARGET_PROPERTY:kernel_tiling_headers,INTERFACE_INCLUDE_DIRECTORIES>,;-I>"176 "-I$<JOIN:$<TARGET_PROPERTY:kernel_tiling_headers,INTERFACE_INCLUDE_DIRECTORIES>,;-I>"
177- -I${OPS_BASE_DIR}/src/ops/op_common/template/aiv/177+ -I${OPS_BASE_DIR}/src/ops/op_common/algorithm/template/aiv/
178- -I${OPS_BASE_DIR}/src/ops/op_common/template/aiv/aiv_interface/178+ -I${OPS_BASE_DIR}/src/ops/op_common/algorithm/template/aiv/aiv_interface/
179 -I${CMAKE_CURRENT_SOURCE_DIR}/aiv_interface/179 -I${CMAKE_CURRENT_SOURCE_DIR}/aiv_interface/
180 -mllvm -cce-aicore-record-overflow=false -mllvm -cce-aicore-addr-transform -mllvm180 -mllvm -cce-aicore-record-overflow=false -mllvm -cce-aicore-addr-transform -mllvm
181 --cce-aicore-jump-expand=true -std=c++17 -fstack-protector-all181 --cce-aicore-jump-expand=true -std=c++17 -fstack-protector-all
@@ -195,8 +195,8 @@ add_custom_command(OUTPUT ${CMAKE_CURRENT_BINARY_DIR}/hccl_ag_superkernel_mesh_1
195 -DHCCL_DTYPE_UINT16 -Dsk_allgather_mesh_1d=sk_allgather_mesh_1d_uint16195 -DHCCL_DTYPE_UINT16 -Dsk_allgather_mesh_1d=sk_allgather_mesh_1d_uint16
196 "-I$<JOIN:$<TARGET_PROPERTY:asc_kernel_headers,INTERFACE_INCLUDE_DIRECTORIES>,;-I>"196 "-I$<JOIN:$<TARGET_PROPERTY:asc_kernel_headers,INTERFACE_INCLUDE_DIRECTORIES>,;-I>"
197 "-I$<JOIN:$<TARGET_PROPERTY:kernel_tiling_headers,INTERFACE_INCLUDE_DIRECTORIES>,;-I>"197 "-I$<JOIN:$<TARGET_PROPERTY:kernel_tiling_headers,INTERFACE_INCLUDE_DIRECTORIES>,;-I>"
198- -I${OPS_BASE_DIR}/src/ops/op_common/template/aiv/198+ -I${OPS_BASE_DIR}/src/ops/op_common/algorithm/template/aiv/
199- -I${OPS_BASE_DIR}/src/ops/op_common/template/aiv/aiv_interface/199+ -I${OPS_BASE_DIR}/src/ops/op_common/algorithm/template/aiv/aiv_interface/
200 -I${CMAKE_CURRENT_SOURCE_DIR}/aiv_interface/200 -I${CMAKE_CURRENT_SOURCE_DIR}/aiv_interface/
201 -mllvm -cce-aicore-record-overflow=false -mllvm -cce-aicore-addr-transform -mllvm201 -mllvm -cce-aicore-record-overflow=false -mllvm -cce-aicore-addr-transform -mllvm
202 --cce-aicore-jump-expand=true -std=c++17 -fstack-protector-all202 --cce-aicore-jump-expand=true -std=c++17 -fstack-protector-all
@@ -216,8 +216,8 @@ add_custom_command(OUTPUT ${CMAKE_CURRENT_BINARY_DIR}/hccl_ag_superkernel_mesh_1
216 -DHCCL_DTYPE_UINT32 -Dsk_allgather_mesh_1d=sk_allgather_mesh_1d_uint32216 -DHCCL_DTYPE_UINT32 -Dsk_allgather_mesh_1d=sk_allgather_mesh_1d_uint32
217 "-I$<JOIN:$<TARGET_PROPERTY:asc_kernel_headers,INTERFACE_INCLUDE_DIRECTORIES>,;-I>"217 "-I$<JOIN:$<TARGET_PROPERTY:asc_kernel_headers,INTERFACE_INCLUDE_DIRECTORIES>,;-I>"
218 "-I$<JOIN:$<TARGET_PROPERTY:kernel_tiling_headers,INTERFACE_INCLUDE_DIRECTORIES>,;-I>"218 "-I$<JOIN:$<TARGET_PROPERTY:kernel_tiling_headers,INTERFACE_INCLUDE_DIRECTORIES>,;-I>"
219- -I${OPS_BASE_DIR}/src/ops/op_common/template/aiv/219+ -I${OPS_BASE_DIR}/src/ops/op_common/algorithm/template/aiv/
220- -I${OPS_BASE_DIR}/src/ops/op_common/template/aiv/aiv_interface/220+ -I${OPS_BASE_DIR}/src/ops/op_common/algorithm/template/aiv/aiv_interface/
221 -I${CMAKE_CURRENT_SOURCE_DIR}/aiv_interface/221 -I${CMAKE_CURRENT_SOURCE_DIR}/aiv_interface/
222 -mllvm -cce-aicore-record-overflow=false -mllvm -cce-aicore-addr-transform -mllvm222 -mllvm -cce-aicore-record-overflow=false -mllvm -cce-aicore-addr-transform -mllvm
223 --cce-aicore-jump-expand=true -std=c++17 -fstack-protector-all223 --cce-aicore-jump-expand=true -std=c++17 -fstack-protector-all
@@ -237,8 +237,8 @@ add_custom_command(OUTPUT ${CMAKE_CURRENT_BINARY_DIR}/hccl_ag_superkernel_mesh_1
237 -DHCCL_DTYPE_INT64 -Dsk_allgather_mesh_1d=sk_allgather_mesh_1d_int64237 -DHCCL_DTYPE_INT64 -Dsk_allgather_mesh_1d=sk_allgather_mesh_1d_int64
238 "-I$<JOIN:$<TARGET_PROPERTY:asc_kernel_headers,INTERFACE_INCLUDE_DIRECTORIES>,;-I>"238 "-I$<JOIN:$<TARGET_PROPERTY:asc_kernel_headers,INTERFACE_INCLUDE_DIRECTORIES>,;-I>"
239 "-I$<JOIN:$<TARGET_PROPERTY:kernel_tiling_headers,INTERFACE_INCLUDE_DIRECTORIES>,;-I>"239 "-I$<JOIN:$<TARGET_PROPERTY:kernel_tiling_headers,INTERFACE_INCLUDE_DIRECTORIES>,;-I>"
240- -I${OPS_BASE_DIR}/src/ops/op_common/template/aiv/240+ -I${OPS_BASE_DIR}/src/ops/op_common/algorithm/template/aiv/
241- -I${OPS_BASE_DIR}/src/ops/op_common/template/aiv/aiv_interface/241+ -I${OPS_BASE_DIR}/src/ops/op_common/algorithm/template/aiv/aiv_interface/
242 -I${CMAKE_CURRENT_SOURCE_DIR}/aiv_interface/242 -I${CMAKE_CURRENT_SOURCE_DIR}/aiv_interface/
243 -mllvm -cce-aicore-record-overflow=false -mllvm -cce-aicore-addr-transform -mllvm243 -mllvm -cce-aicore-record-overflow=false -mllvm -cce-aicore-addr-transform -mllvm
244 --cce-aicore-jump-expand=true -std=c++17 -fstack-protector-all244 --cce-aicore-jump-expand=true -std=c++17 -fstack-protector-all
@@ -258,8 +258,8 @@ add_custom_command(OUTPUT ${CMAKE_CURRENT_BINARY_DIR}/hccl_ag_superkernel_mesh_1
258 -DHCCL_DTYPE_UINT64 -Dsk_allgather_mesh_1d=sk_allgather_mesh_1d_uint64258 -DHCCL_DTYPE_UINT64 -Dsk_allgather_mesh_1d=sk_allgather_mesh_1d_uint64
259 "-I$<JOIN:$<TARGET_PROPERTY:asc_kernel_headers,INTERFACE_INCLUDE_DIRECTORIES>,;-I>"259 "-I$<JOIN:$<TARGET_PROPERTY:asc_kernel_headers,INTERFACE_INCLUDE_DIRECTORIES>,;-I>"
260 "-I$<JOIN:$<TARGET_PROPERTY:kernel_tiling_headers,INTERFACE_INCLUDE_DIRECTORIES>,;-I>"260 "-I$<JOIN:$<TARGET_PROPERTY:kernel_tiling_headers,INTERFACE_INCLUDE_DIRECTORIES>,;-I>"
261- -I${OPS_BASE_DIR}/src/ops/op_common/template/aiv/261+ -I${OPS_BASE_DIR}/src/ops/op_common/algorithm/template/aiv/
262- -I${OPS_BASE_DIR}/src/ops/op_common/template/aiv/aiv_interface/262+ -I${OPS_BASE_DIR}/src/ops/op_common/algorithm/template/aiv/aiv_interface/
263 -I${CMAKE_CURRENT_SOURCE_DIR}/aiv_interface/263 -I${CMAKE_CURRENT_SOURCE_DIR}/aiv_interface/
264 -mllvm -cce-aicore-record-overflow=false -mllvm -cce-aicore-addr-transform -mllvm264 -mllvm -cce-aicore-record-overflow=false -mllvm -cce-aicore-addr-transform -mllvm
265 --cce-aicore-jump-expand=true -std=c++17 -fstack-protector-all265 --cce-aicore-jump-expand=true -std=c++17 -fstack-protector-all
@@ -346,8 +346,8 @@ add_custom_command(OUTPUT ${CMAKE_CURRENT_BINARY_DIR}/hccl_aiv_all_gather_op_910
346 -I${ASCEND_CANN_PACKAGE_PATH}/tools/tikcpp/tikcfw/interface/346 -I${ASCEND_CANN_PACKAGE_PATH}/tools/tikcpp/tikcfw/interface/
347 -I${ASCEND_CANN_PACKAGE_PATH}/tools/tikcpp/tikcfw/impl/347 -I${ASCEND_CANN_PACKAGE_PATH}/tools/tikcpp/tikcfw/impl/
348 -I${ASCEND_CANN_PACKAGE_PATH}/include/hccl/348 -I${ASCEND_CANN_PACKAGE_PATH}/include/hccl/
349- -I${OPS_BASE_DIR}/src/ops/op_common/template/aiv/349+ -I${OPS_BASE_DIR}/src/ops/op_common/algorithm/template/aiv/
350- -I${OPS_BASE_DIR}/src/ops/op_common/template/aiv/aiv_interface/350+ -I${OPS_BASE_DIR}/src/ops/op_common/algorithm/template/aiv/aiv_interface/
351 -mllvm -cce-aicore-record-overflow=false -mllvm -cce-aicore-addr-transform -mllvm351 -mllvm -cce-aicore-record-overflow=false -mllvm -cce-aicore-addr-transform -mllvm
352 --cce-aicore-jump-expand=true -std=c++17 -fstack-protector-all352 --cce-aicore-jump-expand=true -std=c++17 -fstack-protector-all
353 COMMAND ${CCE_DIR}/ld.lld -m aicorelinux -Ttext=0 ${CMAKE_CURRENT_BINARY_DIR}/hccl_aiv_all_gather_op_910_96_tmp.o -static -o ${CMAKE_CURRENT_BINARY_DIR}/hccl_aiv_all_gather_op_910_96.o353 COMMAND ${CCE_DIR}/ld.lld -m aicorelinux -Ttext=0 ${CMAKE_CURRENT_BINARY_DIR}/hccl_aiv_all_gather_op_910_96_tmp.o -static -o ${CMAKE_CURRENT_BINARY_DIR}/hccl_aiv_all_gather_op_910_96.o
@@ -374,8 +374,8 @@ add_custom_command(OUTPUT ${CMAKE_CURRENT_BINARY_DIR}/hccl_ag_superkernel_mesh_1
374 -I${ASCEND_CANN_PACKAGE_PATH}/tools/tikcpp/tikcfw/interface/374 -I${ASCEND_CANN_PACKAGE_PATH}/tools/tikcpp/tikcfw/interface/
375 -I${ASCEND_CANN_PACKAGE_PATH}/tools/tikcpp/tikcfw/impl/375 -I${ASCEND_CANN_PACKAGE_PATH}/tools/tikcpp/tikcfw/impl/
376 -I${ASCEND_CANN_PACKAGE_PATH}/include/hccl/376 -I${ASCEND_CANN_PACKAGE_PATH}/include/hccl/
377- -I${OPS_BASE_DIR}/src/ops/op_common/template/aiv/377+ -I${OPS_BASE_DIR}/src/ops/op_common/algorithm/template/aiv/
378- -I${OPS_BASE_DIR}/src/ops/op_common/template/aiv/aiv_interface/378+ -I${OPS_BASE_DIR}/src/ops/op_common/algorithm/template/aiv/aiv_interface/
379 -I${CMAKE_CURRENT_SOURCE_DIR}/aiv_interface/379 -I${CMAKE_CURRENT_SOURCE_DIR}/aiv_interface/
380 -mllvm -cce-aicore-record-overflow=false -mllvm -cce-aicore-addr-transform -mllvm380 -mllvm -cce-aicore-record-overflow=false -mllvm -cce-aicore-addr-transform -mllvm
381 --cce-aicore-jump-expand=true -std=c++17 -fstack-protector-all381 --cce-aicore-jump-expand=true -std=c++17 -fstack-protector-all
@@ -402,8 +402,8 @@ add_custom_command(OUTPUT ${CMAKE_CURRENT_BINARY_DIR}/hccl_ag_superkernel_mesh_1
402 -I${ASCEND_CANN_PACKAGE_PATH}/tools/tikcpp/tikcfw/interface/402 -I${ASCEND_CANN_PACKAGE_PATH}/tools/tikcpp/tikcfw/interface/
403 -I${ASCEND_CANN_PACKAGE_PATH}/tools/tikcpp/tikcfw/impl/403 -I${ASCEND_CANN_PACKAGE_PATH}/tools/tikcpp/tikcfw/impl/
404 -I${ASCEND_CANN_PACKAGE_PATH}/include/hccl/404 -I${ASCEND_CANN_PACKAGE_PATH}/include/hccl/
405- -I${OPS_BASE_DIR}/src/ops/op_common/template/aiv/405+ -I${OPS_BASE_DIR}/src/ops/op_common/algorithm/template/aiv/
406- -I${OPS_BASE_DIR}/src/ops/op_common/template/aiv/aiv_interface/406+ -I${OPS_BASE_DIR}/src/ops/op_common/algorithm/template/aiv/aiv_interface/
407 -I${CMAKE_CURRENT_SOURCE_DIR}/aiv_interface/407 -I${CMAKE_CURRENT_SOURCE_DIR}/aiv_interface/
408 -mllvm -cce-aicore-record-overflow=false -mllvm -cce-aicore-addr-transform -mllvm408 -mllvm -cce-aicore-record-overflow=false -mllvm -cce-aicore-addr-transform -mllvm
409 --cce-aicore-jump-expand=true -std=c++17 -fstack-protector-all409 --cce-aicore-jump-expand=true -std=c++17 -fstack-protector-all
@@ -430,8 +430,8 @@ add_custom_command(OUTPUT ${CMAKE_CURRENT_BINARY_DIR}/hccl_ag_superkernel_mesh_1
430 -I${ASCEND_CANN_PACKAGE_PATH}/tools/tikcpp/tikcfw/interface/430 -I${ASCEND_CANN_PACKAGE_PATH}/tools/tikcpp/tikcfw/interface/
431 -I${ASCEND_CANN_PACKAGE_PATH}/tools/tikcpp/tikcfw/impl/431 -I${ASCEND_CANN_PACKAGE_PATH}/tools/tikcpp/tikcfw/impl/
432 -I${ASCEND_CANN_PACKAGE_PATH}/include/hccl/432 -I${ASCEND_CANN_PACKAGE_PATH}/include/hccl/
433- -I${OPS_BASE_DIR}/src/ops/op_common/template/aiv/433+ -I${OPS_BASE_DIR}/src/ops/op_common/algorithm/template/aiv/
434- -I${OPS_BASE_DIR}/src/ops/op_common/template/aiv/aiv_interface/434+ -I${OPS_BASE_DIR}/src/ops/op_common/algorithm/template/aiv/aiv_interface/
435 -I${CMAKE_CURRENT_SOURCE_DIR}/aiv_interface/435 -I${CMAKE_CURRENT_SOURCE_DIR}/aiv_interface/
436 -mllvm -cce-aicore-record-overflow=false -mllvm -cce-aicore-addr-transform -mllvm436 -mllvm -cce-aicore-record-overflow=false -mllvm -cce-aicore-addr-transform -mllvm
437 --cce-aicore-jump-expand=true -std=c++17 -fstack-protector-all437 --cce-aicore-jump-expand=true -std=c++17 -fstack-protector-all
@@ -458,8 +458,8 @@ add_custom_command(OUTPUT ${CMAKE_CURRENT_BINARY_DIR}/hccl_ag_superkernel_mesh_1
458 -I${ASCEND_CANN_PACKAGE_PATH}/tools/tikcpp/tikcfw/interface/458 -I${ASCEND_CANN_PACKAGE_PATH}/tools/tikcpp/tikcfw/interface/
459 -I${ASCEND_CANN_PACKAGE_PATH}/tools/tikcpp/tikcfw/impl/459 -I${ASCEND_CANN_PACKAGE_PATH}/tools/tikcpp/tikcfw/impl/
460 -I${ASCEND_CANN_PACKAGE_PATH}/include/hccl/460 -I${ASCEND_CANN_PACKAGE_PATH}/include/hccl/
461- -I${OPS_BASE_DIR}/src/ops/op_common/template/aiv/461+ -I${OPS_BASE_DIR}/src/ops/op_common/algorithm/template/aiv/
462- -I${OPS_BASE_DIR}/src/ops/op_common/template/aiv/aiv_interface/462+ -I${OPS_BASE_DIR}/src/ops/op_common/algorithm/template/aiv/aiv_interface/
463 -I${CMAKE_CURRENT_SOURCE_DIR}/aiv_interface/463 -I${CMAKE_CURRENT_SOURCE_DIR}/aiv_interface/
464 -mllvm -cce-aicore-record-overflow=false -mllvm -cce-aicore-addr-transform -mllvm464 -mllvm -cce-aicore-record-overflow=false -mllvm -cce-aicore-addr-transform -mllvm
465 --cce-aicore-jump-expand=true -std=c++17 -fstack-protector-all465 --cce-aicore-jump-expand=true -std=c++17 -fstack-protector-all
@@ -486,8 +486,8 @@ add_custom_command(OUTPUT ${CMAKE_CURRENT_BINARY_DIR}/hccl_ag_superkernel_mesh_1
486 -I${ASCEND_CANN_PACKAGE_PATH}/tools/tikcpp/tikcfw/interface/486 -I${ASCEND_CANN_PACKAGE_PATH}/tools/tikcpp/tikcfw/interface/
487 -I${ASCEND_CANN_PACKAGE_PATH}/tools/tikcpp/tikcfw/impl/487 -I${ASCEND_CANN_PACKAGE_PATH}/tools/tikcpp/tikcfw/impl/
488 -I${ASCEND_CANN_PACKAGE_PATH}/include/hccl/488 -I${ASCEND_CANN_PACKAGE_PATH}/include/hccl/
489- -I${OPS_BASE_DIR}/src/ops/op_common/template/aiv/489+ -I${OPS_BASE_DIR}/src/ops/op_common/algorithm/template/aiv/
490- -I${OPS_BASE_DIR}/src/ops/op_common/template/aiv/aiv_interface/490+ -I${OPS_BASE_DIR}/src/ops/op_common/algorithm/template/aiv/aiv_interface/
491 -I${CMAKE_CURRENT_SOURCE_DIR}/aiv_interface/491 -I${CMAKE_CURRENT_SOURCE_DIR}/aiv_interface/
492 -mllvm -cce-aicore-record-overflow=false -mllvm -cce-aicore-addr-transform -mllvm492 -mllvm -cce-aicore-record-overflow=false -mllvm -cce-aicore-addr-transform -mllvm
493 --cce-aicore-jump-expand=true -std=c++17 -fstack-protector-all493 --cce-aicore-jump-expand=true -std=c++17 -fstack-protector-all
@@ -514,8 +514,8 @@ add_custom_command(OUTPUT ${CMAKE_CURRENT_BINARY_DIR}/hccl_ag_superkernel_mesh_1
514 -I${ASCEND_CANN_PACKAGE_PATH}/tools/tikcpp/tikcfw/interface/514 -I${ASCEND_CANN_PACKAGE_PATH}/tools/tikcpp/tikcfw/interface/
515 -I${ASCEND_CANN_PACKAGE_PATH}/tools/tikcpp/tikcfw/impl/515 -I${ASCEND_CANN_PACKAGE_PATH}/tools/tikcpp/tikcfw/impl/
516 -I${ASCEND_CANN_PACKAGE_PATH}/include/hccl/516 -I${ASCEND_CANN_PACKAGE_PATH}/include/hccl/
517- -I${OPS_BASE_DIR}/src/ops/op_common/template/aiv/517+ -I${OPS_BASE_DIR}/src/ops/op_common/algorithm/template/aiv/
518- -I${OPS_BASE_DIR}/src/ops/op_common/template/aiv/aiv_interface/518+ -I${OPS_BASE_DIR}/src/ops/op_common/algorithm/template/aiv/aiv_interface/
519 -I${CMAKE_CURRENT_SOURCE_DIR}/aiv_interface/519 -I${CMAKE_CURRENT_SOURCE_DIR}/aiv_interface/
520 -mllvm -cce-aicore-record-overflow=false -mllvm -cce-aicore-addr-transform -mllvm520 -mllvm -cce-aicore-record-overflow=false -mllvm -cce-aicore-addr-transform -mllvm
521 --cce-aicore-jump-expand=true -std=c++17 -fstack-protector-all521 --cce-aicore-jump-expand=true -std=c++17 -fstack-protector-all
@@ -542,8 +542,8 @@ add_custom_command(OUTPUT ${CMAKE_CURRENT_BINARY_DIR}/hccl_ag_superkernel_mesh_1
542 -I${ASCEND_CANN_PACKAGE_PATH}/tools/tikcpp/tikcfw/interface/542 -I${ASCEND_CANN_PACKAGE_PATH}/tools/tikcpp/tikcfw/interface/
543 -I${ASCEND_CANN_PACKAGE_PATH}/tools/tikcpp/tikcfw/impl/543 -I${ASCEND_CANN_PACKAGE_PATH}/tools/tikcpp/tikcfw/impl/
544 -I${ASCEND_CANN_PACKAGE_PATH}/include/hccl/544 -I${ASCEND_CANN_PACKAGE_PATH}/include/hccl/
545- -I${OPS_BASE_DIR}/src/ops/op_common/template/aiv/545+ -I${OPS_BASE_DIR}/src/ops/op_common/algorithm/template/aiv/
546- -I${OPS_BASE_DIR}/src/ops/op_common/template/aiv/aiv_interface/546+ -I${OPS_BASE_DIR}/src/ops/op_common/algorithm/template/aiv/aiv_interface/
547 -I${CMAKE_CURRENT_SOURCE_DIR}/aiv_interface/547 -I${CMAKE_CURRENT_SOURCE_DIR}/aiv_interface/
548 -mllvm -cce-aicore-record-overflow=false -mllvm -cce-aicore-addr-transform -mllvm548 -mllvm -cce-aicore-record-overflow=false -mllvm -cce-aicore-addr-transform -mllvm
549 --cce-aicore-jump-expand=true -std=c++17 -fstack-protector-all549 --cce-aicore-jump-expand=true -std=c++17 -fstack-protector-all
@@ -570,8 +570,8 @@ add_custom_command(OUTPUT ${CMAKE_CURRENT_BINARY_DIR}/hccl_ag_superkernel_mesh_1
570 -I${ASCEND_CANN_PACKAGE_PATH}/tools/tikcpp/tikcfw/interface/570 -I${ASCEND_CANN_PACKAGE_PATH}/tools/tikcpp/tikcfw/interface/
571 -I${ASCEND_CANN_PACKAGE_PATH}/tools/tikcpp/tikcfw/impl/571 -I${ASCEND_CANN_PACKAGE_PATH}/tools/tikcpp/tikcfw/impl/
572 -I${ASCEND_CANN_PACKAGE_PATH}/include/hccl/572 -I${ASCEND_CANN_PACKAGE_PATH}/include/hccl/
573- -I${OPS_BASE_DIR}/src/ops/op_common/template/aiv/573+ -I${OPS_BASE_DIR}/src/ops/op_common/algorithm/template/aiv/
574- -I${OPS_BASE_DIR}/src/ops/op_common/template/aiv/aiv_interface/574+ -I${OPS_BASE_DIR}/src/ops/op_common/algorithm/template/aiv/aiv_interface/
575 -I${CMAKE_CURRENT_SOURCE_DIR}/aiv_interface/575 -I${CMAKE_CURRENT_SOURCE_DIR}/aiv_interface/
576 -mllvm -cce-aicore-record-overflow=false -mllvm -cce-aicore-addr-transform -mllvm576 -mllvm -cce-aicore-record-overflow=false -mllvm -cce-aicore-addr-transform -mllvm
577 --cce-aicore-jump-expand=true -std=c++17 -fstack-protector-all577 --cce-aicore-jump-expand=true -std=c++17 -fstack-protector-all
@@ -598,8 +598,8 @@ add_custom_command(OUTPUT ${CMAKE_CURRENT_BINARY_DIR}/hccl_ag_superkernel_mesh_1
598 -I${ASCEND_CANN_PACKAGE_PATH}/tools/tikcpp/tikcfw/interface/598 -I${ASCEND_CANN_PACKAGE_PATH}/tools/tikcpp/tikcfw/interface/
599 -I${ASCEND_CANN_PACKAGE_PATH}/tools/tikcpp/tikcfw/impl/599 -I${ASCEND_CANN_PACKAGE_PATH}/tools/tikcpp/tikcfw/impl/
600 -I${ASCEND_CANN_PACKAGE_PATH}/include/hccl/600 -I${ASCEND_CANN_PACKAGE_PATH}/include/hccl/
601- -I${OPS_BASE_DIR}/src/ops/op_common/template/aiv/601+ -I${OPS_BASE_DIR}/src/ops/op_common/algorithm/template/aiv/
602- -I${OPS_BASE_DIR}/src/ops/op_common/template/aiv/aiv_interface/602+ -I${OPS_BASE_DIR}/src/ops/op_common/algorithm/template/aiv/aiv_interface/
603 -I${CMAKE_CURRENT_SOURCE_DIR}/aiv_interface/603 -I${CMAKE_CURRENT_SOURCE_DIR}/aiv_interface/
604 -mllvm -cce-aicore-record-overflow=false -mllvm -cce-aicore-addr-transform -mllvm604 -mllvm -cce-aicore-record-overflow=false -mllvm -cce-aicore-addr-transform -mllvm
605 --cce-aicore-jump-expand=true -std=c++17 -fstack-protector-all605 --cce-aicore-jump-expand=true -std=c++17 -fstack-protector-all
@@ -626,8 +626,8 @@ add_custom_command(OUTPUT ${CMAKE_CURRENT_BINARY_DIR}/hccl_ag_superkernel_mesh_1
626 -I${ASCEND_CANN_PACKAGE_PATH}/tools/tikcpp/tikcfw/interface/626 -I${ASCEND_CANN_PACKAGE_PATH}/tools/tikcpp/tikcfw/interface/
627 -I${ASCEND_CANN_PACKAGE_PATH}/tools/tikcpp/tikcfw/impl/627 -I${ASCEND_CANN_PACKAGE_PATH}/tools/tikcpp/tikcfw/impl/
628 -I${ASCEND_CANN_PACKAGE_PATH}/include/hccl/628 -I${ASCEND_CANN_PACKAGE_PATH}/include/hccl/
629- -I${OPS_BASE_DIR}/src/ops/op_common/template/aiv/629+ -I${OPS_BASE_DIR}/src/ops/op_common/algorithm/template/aiv/
630- -I${OPS_BASE_DIR}/src/ops/op_common/template/aiv/aiv_interface/630+ -I${OPS_BASE_DIR}/src/ops/op_common/algorithm/template/aiv/aiv_interface/
631 -I${CMAKE_CURRENT_SOURCE_DIR}/aiv_interface/631 -I${CMAKE_CURRENT_SOURCE_DIR}/aiv_interface/
632 -mllvm -cce-aicore-record-overflow=false -mllvm -cce-aicore-addr-transform -mllvm632 -mllvm -cce-aicore-record-overflow=false -mllvm -cce-aicore-addr-transform -mllvm
633 --cce-aicore-jump-expand=true -std=c++17 -fstack-protector-all633 --cce-aicore-jump-expand=true -std=c++17 -fstack-protector-all
@@ -654,8 +654,8 @@ add_custom_command(OUTPUT ${CMAKE_CURRENT_BINARY_DIR}/hccl_ag_superkernel_mesh_1
654 -I${ASCEND_CANN_PACKAGE_PATH}/tools/tikcpp/tikcfw/interface/654 -I${ASCEND_CANN_PACKAGE_PATH}/tools/tikcpp/tikcfw/interface/
655 -I${ASCEND_CANN_PACKAGE_PATH}/tools/tikcpp/tikcfw/impl/655 -I${ASCEND_CANN_PACKAGE_PATH}/tools/tikcpp/tikcfw/impl/
656 -I${ASCEND_CANN_PACKAGE_PATH}/include/hccl/656 -I${ASCEND_CANN_PACKAGE_PATH}/include/hccl/
657- -I${OPS_BASE_DIR}/src/ops/op_common/template/aiv/657+ -I${OPS_BASE_DIR}/src/ops/op_common/algorithm/template/aiv/
658- -I${OPS_BASE_DIR}/src/ops/op_common/template/aiv/aiv_interface/658+ -I${OPS_BASE_DIR}/src/ops/op_common/algorithm/template/aiv/aiv_interface/
659 -I${CMAKE_CURRENT_SOURCE_DIR}/aiv_interface/659 -I${CMAKE_CURRENT_SOURCE_DIR}/aiv_interface/
660 -mllvm -cce-aicore-record-overflow=false -mllvm -cce-aicore-addr-transform -mllvm660 -mllvm -cce-aicore-record-overflow=false -mllvm -cce-aicore-addr-transform -mllvm
661 --cce-aicore-jump-expand=true -std=c++17 -fstack-protector-all661 --cce-aicore-jump-expand=true -std=c++17 -fstack-protector-all
Rsrc/ops/all_gather/template/aiv/kernel/aiv_ag_mesh_1d_superkernel.h→src/ops/all_gather/algorithm/template/aiv/kernel/aiv_ag_mesh_1d_superkernel.h+0-0
文件重命名但无更改。
Rsrc/ops/all_gather/template/aiv/kernel/aiv_all_gather_mesh_1d.h→src/ops/all_gather/algorithm/template/aiv/kernel/aiv_all_gather_mesh_1d.h+0-0
文件重命名但无更改。
Rsrc/ops/all_gather/template/aiv/kernel/aiv_all_gather_op.h→src/ops/all_gather/algorithm/template/aiv/kernel/aiv_all_gather_op.h+0-0
文件重命名但无更改。
Rsrc/ops/all_gather/template/ccu/CMakeLists.txt→src/ops/all_gather/algorithm/template/ccu/CMakeLists.txt+0-0
文件重命名但无更改。
Rsrc/ops/all_gather/template/ccu/ccu_temp_all_gather_2dies_mesh_1D.cc→src/ops/all_gather/algorithm/template/ccu/ccu_temp_all_gather_2dies_mesh_1D.cc+0-0
文件重命名但无更改。
Rsrc/ops/all_gather/template/ccu/ccu_temp_all_gather_2dies_mesh_1D.h→src/ops/all_gather/algorithm/template/ccu/ccu_temp_all_gather_2dies_mesh_1D.h+0-0
文件重命名但无更改。
Rsrc/ops/all_gather/template/ccu/ccu_temp_all_gather_2dies_mesh_1d_mem2mem.cc→src/ops/all_gather/algorithm/template/ccu/ccu_temp_all_gather_2dies_mesh_1d_mem2mem.cc+0-0
文件重命名但无更改。
Rsrc/ops/all_gather/template/ccu/ccu_temp_all_gather_2dies_mesh_1d_mem2mem.h→src/ops/all_gather/algorithm/template/ccu/ccu_temp_all_gather_2dies_mesh_1d_mem2mem.h+0-0
文件重命名但无更改。
Rsrc/ops/all_gather/template/ccu/ccu_temp_all_gather_concurrent_mesh_mem2mem_nhr.cc→src/ops/all_gather/algorithm/template/ccu/ccu_temp_all_gather_concurrent_mesh_mem2mem_nhr.cc+0-0
文件重命名但无更改。
Rsrc/ops/all_gather/template/ccu/ccu_temp_all_gather_concurrent_mesh_mem2mem_nhr.h→src/ops/all_gather/algorithm/template/ccu/ccu_temp_all_gather_concurrent_mesh_mem2mem_nhr.h+0-0
文件重命名但无更改。
Rsrc/ops/all_gather/template/ccu/ccu_temp_all_gather_mesh_1D.cc→src/ops/all_gather/algorithm/template/ccu/ccu_temp_all_gather_mesh_1D.cc+0-0
文件重命名但无更改。
Rsrc/ops/all_gather/template/ccu/ccu_temp_all_gather_mesh_1D.h→src/ops/all_gather/algorithm/template/ccu/ccu_temp_all_gather_mesh_1D.h+0-0
文件重命名但无更改。
Rsrc/ops/all_gather/template/ccu/ccu_temp_all_gather_mesh_1D_mem2mem.cc→src/ops/all_gather/algorithm/template/ccu/ccu_temp_all_gather_mesh_1D_mem2mem.cc+0-0
文件重命名但无更改。
Rsrc/ops/all_gather/template/ccu/ccu_temp_all_gather_mesh_1D_mem2mem.h→src/ops/all_gather/algorithm/template/ccu/ccu_temp_all_gather_mesh_1D_mem2mem.h+0-0
文件重命名但无更改。
Rsrc/ops/all_gather/template/ccu/ccu_temp_all_gather_nhr_1D_mem2mem.cc→src/ops/all_gather/algorithm/template/ccu/ccu_temp_all_gather_nhr_1D_mem2mem.cc+0-0
文件重命名但无更改。
Rsrc/ops/all_gather/template/ccu/ccu_temp_all_gather_nhr_1D_mem2mem.h→src/ops/all_gather/algorithm/template/ccu/ccu_temp_all_gather_nhr_1D_mem2mem.h+0-0
文件重命名但无更改。
Rsrc/ops/all_gather/template/ccu/ccu_temp_all_gather_nhr_1D_multi_jetty_mem2mem.cc→src/ops/all_gather/algorithm/template/ccu/ccu_temp_all_gather_nhr_1D_multi_jetty_mem2mem.cc+0-0
文件重命名但无更改。
Rsrc/ops/all_gather/template/ccu/ccu_temp_all_gather_nhr_1D_multi_jetty_mem2mem.h→src/ops/all_gather/algorithm/template/ccu/ccu_temp_all_gather_nhr_1D_multi_jetty_mem2mem.h+0-0
文件重命名但无更改。
Rsrc/ops/all_gather/template/ccu/ccu_temp_all_gather_omnipipe_mesh1d_mem2mem.cc→src/ops/all_gather/algorithm/template/ccu/ccu_temp_all_gather_omnipipe_mesh1d_mem2mem.cc+0-0
文件重命名但无更改。
Rsrc/ops/all_gather/template/ccu/ccu_temp_all_gather_omnipipe_mesh1d_mem2mem.h→src/ops/all_gather/algorithm/template/ccu/ccu_temp_all_gather_omnipipe_mesh1d_mem2mem.h+0-0
文件重命名但无更改。
Rsrc/ops/all_gather/template/ccu/ccu_temp_all_gather_omnipipe_nhr1d_mem2mem.cc→src/ops/all_gather/algorithm/template/ccu/ccu_temp_all_gather_omnipipe_nhr1d_mem2mem.cc+0-0
文件重命名但无更改。
Rsrc/ops/all_gather/template/ccu/ccu_temp_all_gather_omnipipe_nhr1d_mem2mem.h→src/ops/all_gather/algorithm/template/ccu/ccu_temp_all_gather_omnipipe_nhr1d_mem2mem.h+0-0
文件重命名但无更改。
Rsrc/ops/all_gather/template/ccu/kernel/CMakeLists.txt→src/ops/all_gather/algorithm/template/ccu/kernel/CMakeLists.txt+0-0
文件重命名但无更改。
Rsrc/ops/all_gather/template/ccu/kernel/ccu_kernel_all_gather_2dies_mesh1d_mem2mem.cc→src/ops/all_gather/algorithm/template/ccu/kernel/ccu_kernel_all_gather_2dies_mesh1d_mem2mem.cc+0-0
文件重命名但无更改。
Rsrc/ops/all_gather/template/ccu/kernel/ccu_kernel_all_gather_2dies_mesh1d_mem2mem.h→src/ops/all_gather/algorithm/template/ccu/kernel/ccu_kernel_all_gather_2dies_mesh1d_mem2mem.h+0-0
文件重命名但无更改。
Rsrc/ops/all_gather/template/ccu/kernel/ccu_kernel_all_gather_2dies_mesh_1D.cc→src/ops/all_gather/algorithm/template/ccu/kernel/ccu_kernel_all_gather_2dies_mesh_1D.cc+0-0
文件重命名但无更改。
Rsrc/ops/all_gather/template/ccu/kernel/ccu_kernel_all_gather_2dies_mesh_1D.h→src/ops/all_gather/algorithm/template/ccu/kernel/ccu_kernel_all_gather_2dies_mesh_1D.h+0-0
文件重命名但无更改。
Rsrc/ops/all_gather/template/ccu/kernel/ccu_kernel_all_gather_mesh1d.cc→src/ops/all_gather/algorithm/template/ccu/kernel/ccu_kernel_all_gather_mesh1d.cc+0-0
文件重命名但无更改。
Rsrc/ops/all_gather/template/ccu/kernel/ccu_kernel_all_gather_mesh1d.h→src/ops/all_gather/algorithm/template/ccu/kernel/ccu_kernel_all_gather_mesh1d.h+0-0
文件重命名但无更改。
Rsrc/ops/all_gather/template/ccu/kernel/ccu_kernel_all_gather_mesh1d_mem2mem.cc→src/ops/all_gather/algorithm/template/ccu/kernel/ccu_kernel_all_gather_mesh1d_mem2mem.cc+0-0
文件重命名但无更改。
Rsrc/ops/all_gather/template/ccu/kernel/ccu_kernel_all_gather_mesh1d_mem2mem.h→src/ops/all_gather/algorithm/template/ccu/kernel/ccu_kernel_all_gather_mesh1d_mem2mem.h+0-0
文件重命名但无更改。
Rsrc/ops/all_gather/template/ccu/kernel/ccu_kernel_all_gather_nhr1d_mem2mem.cc→src/ops/all_gather/algorithm/template/ccu/kernel/ccu_kernel_all_gather_nhr1d_mem2mem.cc+0-0
文件重命名但无更改。
Rsrc/ops/all_gather/template/ccu/kernel/ccu_kernel_all_gather_nhr1d_mem2mem.h→src/ops/all_gather/algorithm/template/ccu/kernel/ccu_kernel_all_gather_nhr1d_mem2mem.h+0-0
文件重命名但无更改。
Rsrc/ops/all_gather/template/ccu/kernel/ccu_kernel_all_gather_nhr1d_multi_jetty_mem2mem.cc→src/ops/all_gather/algorithm/template/ccu/kernel/ccu_kernel_all_gather_nhr1d_multi_jetty_mem2mem.cc+0-0
文件重命名但无更改。
Rsrc/ops/all_gather/template/ccu/kernel/ccu_kernel_all_gather_nhr1d_multi_jetty_mem2mem.h→src/ops/all_gather/algorithm/template/ccu/kernel/ccu_kernel_all_gather_nhr1d_multi_jetty_mem2mem.h+0-0
文件重命名但无更改。
Rsrc/ops/all_gather/template/ccu/kernel/ccu_kernel_all_gather_omnipipe_mesh1d_mem2mem.cc→src/ops/all_gather/algorithm/template/ccu/kernel/ccu_kernel_all_gather_omnipipe_mesh1d_mem2mem.cc+0-0
文件重命名但无更改。
Rsrc/ops/all_gather/template/ccu/kernel/ccu_kernel_all_gather_omnipipe_mesh1d_mem2mem.h→src/ops/all_gather/algorithm/template/ccu/kernel/ccu_kernel_all_gather_omnipipe_mesh1d_mem2mem.h+0-0
文件重命名但无更改。
Rsrc/ops/all_gather/template/ccu/kernel/ccu_kernel_all_gather_omnipipe_nhr1d_mem2mem.cc→src/ops/all_gather/algorithm/template/ccu/kernel/ccu_kernel_all_gather_omnipipe_nhr1d_mem2mem.cc+0-0
文件重命名但无更改。
Rsrc/ops/all_gather/template/ccu/kernel/ccu_kernel_all_gather_omnipipe_nhr1d_mem2mem.h→src/ops/all_gather/algorithm/template/ccu/kernel/ccu_kernel_all_gather_omnipipe_nhr1d_mem2mem.h+0-0
文件重命名但无更改。
Rsrc/ops/all_gather/all_gather_op.cc→src/ops/all_gather/all_gather.cc+1-1
@@ -8,7 +8,7 @@
8 * See LICENSE in the root of the software repository for the full text of the License.8 * See LICENSE in the root of the software repository for the full text of the License.
9 */9 */
10 10 
11-#include "all_gather_op.h"11+#include "all_gather.h"
12#include "op_common_ops.h"12#include "op_common_ops.h"
13#include <algorithm>13#include <algorithm>
14#include <future>14#include <future>
Rsrc/ops/all_gather/all_gather_op.h→src/ops/all_gather/all_gather.h+0-0
文件重命名但无更改。
Msrc/ops/all_gather_v/CMakeLists.txt+3-4
@@ -8,13 +8,12 @@
8# See LICENSE in the root of the software repository for the full text of the License.8# See LICENSE in the root of the software repository for the full text of the License.
9# -----------------------------------------------------------------------------------------------------------9# -----------------------------------------------------------------------------------------------------------
10 10 
11-add_subdirectory(executor)11+add_subdirectory(algorithm)
12add_subdirectory(selector)12add_subdirectory(selector)
13-add_subdirectory(template)
14add_subdirectory(op_graph)13add_subdirectory(op_graph)
15- 14+ 
16set(src_list15set(src_list
17- ${CMAKE_CURRENT_SOURCE_DIR}/all_gather_v_op.cc16+ ${CMAKE_CURRENT_SOURCE_DIR}/all_gather_v.cc
18)17)
19 18 
20if(TARGET hccl)19if(TARGET hccl)
Asrc/ops/all_gather_v/algorithm/CMakeLists.txt+12-0
@@ -0,0 +1,12 @@
1+# -----------------------------------------------------------------------------------------------------------
2+# Copyright (c) 2026 Huawei Technologies Co., Ltd.
3+# This program is free software, you can redistribute it and/or modify it under the terms and conditions of
4+# CANN Open Software License Agreement Version 2.0 (the "License").
5+# Please refer to the License for details. You may not use this file except in compliance with the License.
6+# THIS SOFTWARE IS PROVIDED ON AN "AS IS" BASIS, WITHOUT WARRANTIES OF ANY KIND, EITHER EXPRESS OR IMPLIED,
7+# INCLUDING BUT NOT LIMITED TO NON-INFRINGEMENT, MERCHANTABILITY, OR FITNESS FOR A PARTICULAR PURPOSE.
8+# See LICENSE in the root of the software repository for the full text of the License.
9+# -----------------------------------------------------------------------------------------------------------
10+ 
11+add_subdirectory(executor)
12+add_subdirectory(template)
Rsrc/ops/all_gather_v/executor/CMakeLists.txt→src/ops/all_gather_v/algorithm/executor/CMakeLists.txt+0-0
文件重命名但无更改。
Rsrc/ops/all_gather_v/executor/ins_v2_all_gather_v_sole_executor.cc→src/ops/all_gather_v/algorithm/executor/ins_v2_all_gather_v_sole_executor.cc+0-0
文件重命名但无更改。
Rsrc/ops/all_gather_v/executor/ins_v2_all_gather_v_sole_executor.h→src/ops/all_gather_v/algorithm/executor/ins_v2_all_gather_v_sole_executor.h+0-0
文件重命名但无更改。
Rsrc/ops/all_gather_v/template/CMakeLists.txt→src/ops/all_gather_v/algorithm/template/CMakeLists.txt+0-0
文件重命名但无更改。
Rsrc/ops/all_gather_v/template/aicpu/CMakeLists.txt→src/ops/all_gather_v/algorithm/template/aicpu/CMakeLists.txt+0-0
Rsrc/ops/all_gather_v/template/aicpu/ins_temp_all_gather_v_mesh_1D.cc→src/ops/all_gather_v/algorithm/template/aicpu/ins_temp_all_gather_v_mesh_1D.cc+0-0
Rsrc/ops/all_gather_v/template/aicpu/ins_temp_all_gather_v_mesh_1D.h→src/ops/all_gather_v/algorithm/template/aicpu/ins_temp_all_gather_v_mesh_1D.h+0-0
Rsrc/ops/all_gather_v/template/ccu/CMakeLists.txt→src/ops/all_gather_v/algorithm/template/ccu/CMakeLists.txt+0-0
Rsrc/ops/all_gather_v/template/ccu/ccu_temp_all_gather_v_mesh_1D_mem2mem.cc→src/ops/all_gather_v/algorithm/template/ccu/ccu_temp_all_gather_v_mesh_1D_mem2mem.cc+0-0
Rsrc/ops/all_gather_v/template/ccu/ccu_temp_all_gather_v_mesh_1D_mem2mem.h→src/ops/all_gather_v/algorithm/template/ccu/ccu_temp_all_gather_v_mesh_1D_mem2mem.h+0-0
Rsrc/ops/all_gather_v/template/ccu/kernel/CMakeLists.txt→src/ops/all_gather_v/algorithm/template/ccu/kernel/CMakeLists.txt+0-0
Rsrc/ops/all_gather_v/template/ccu/kernel/ccu_kernel_all_gather_v_mesh1d_mem2mem.cc→src/ops/all_gather_v/algorithm/template/ccu/kernel/ccu_kernel_all_gather_v_mesh1d_mem2mem.cc+0-0
Rsrc/ops/all_gather_v/template/ccu/kernel/ccu_kernel_all_gather_v_mesh1d_mem2mem.h→src/ops/all_gather_v/algorithm/template/ccu/kernel/ccu_kernel_all_gather_v_mesh1d_mem2mem.h+0-0
Rsrc/ops/all_gather_v/all_gather_v_op.cc→src/ops/all_gather_v/all_gather_v.cc+1-1
Rsrc/ops/all_gather_v/all_gather_v_op.h→src/ops/all_gather_v/all_gather_v.h+0-0
Rsrc/ops/all_reduce/executor/CMakeLists.txt→src/ops/all_reduce/algorithm/executor/CMakeLists.txt+0-0
Rsrc/ops/all_reduce/executor/ins_v2_all_reduce_concurrent_executor.cc→src/ops/all_reduce/algorithm/executor/ins_v2_all_reduce_concurrent_executor.cc+0-0
Rsrc/ops/all_reduce/executor/ins_v2_all_reduce_concurrent_executor.h→src/ops/all_reduce/algorithm/executor/ins_v2_all_reduce_concurrent_executor.h+0-0
Rsrc/ops/all_reduce/executor/ins_v2_all_reduce_omnipipe_2d_executor.cc→src/ops/all_reduce/algorithm/executor/ins_v2_all_reduce_omnipipe_2d_executor.cc+0-0
Rsrc/ops/all_reduce/executor/ins_v2_all_reduce_omnipipe_2d_executor.h→src/ops/all_reduce/algorithm/executor/ins_v2_all_reduce_omnipipe_2d_executor.h+0-0
Rsrc/ops/all_reduce/executor/ins_v2_all_reduce_omnipipe_executor.cc→src/ops/all_reduce/algorithm/executor/ins_v2_all_reduce_omnipipe_executor.cc+0-0
Rsrc/ops/all_reduce/executor/ins_v2_all_reduce_omnipipe_executor.h→src/ops/all_reduce/algorithm/executor/ins_v2_all_reduce_omnipipe_executor.h+0-0
Rsrc/ops/all_reduce/executor/ins_v2_all_reduce_order_preserved_executor.cc→src/ops/all_reduce/algorithm/executor/ins_v2_all_reduce_order_preserved_executor.cc+0-0
Rsrc/ops/all_reduce/executor/ins_v2_all_reduce_order_preserved_executor.h→src/ops/all_reduce/algorithm/executor/ins_v2_all_reduce_order_preserved_executor.h+0-0
Rsrc/ops/all_reduce/executor/ins_v2_all_reduce_parallel_executor.cc→src/ops/all_reduce/algorithm/executor/ins_v2_all_reduce_parallel_executor.cc+0-0
Rsrc/ops/all_reduce/executor/ins_v2_all_reduce_parallel_executor.h→src/ops/all_reduce/algorithm/executor/ins_v2_all_reduce_parallel_executor.h+0-0
Rsrc/ops/all_reduce/executor/ins_v2_all_reduce_sequence_executor.cc→src/ops/all_reduce/algorithm/executor/ins_v2_all_reduce_sequence_executor.cc+0-0
Rsrc/ops/all_reduce/executor/ins_v2_all_reduce_sequence_executor.h→src/ops/all_reduce/algorithm/executor/ins_v2_all_reduce_sequence_executor.h+0-0
Rsrc/ops/all_reduce/executor/ins_v2_all_reduce_sequence_executor_aicpu.cc→src/ops/all_reduce/algorithm/executor/ins_v2_all_reduce_sequence_executor_aicpu.cc+0-0
Rsrc/ops/all_reduce/executor/ins_v2_all_reduce_sequence_executor_aicpu.h→src/ops/all_reduce/algorithm/executor/ins_v2_all_reduce_sequence_executor_aicpu.h+0-0
Rsrc/ops/all_reduce/executor/ins_v2_all_reduce_sequence_executor_aicpu_3level.cc→src/ops/all_reduce/algorithm/executor/ins_v2_all_reduce_sequence_executor_aicpu_3level.cc+0-0
Rsrc/ops/all_reduce/executor/ins_v2_all_reduce_sequence_executor_aicpu_3level.h→src/ops/all_reduce/algorithm/executor/ins_v2_all_reduce_sequence_executor_aicpu_3level.h+0-0
Rsrc/ops/all_reduce/executor/ins_v2_all_reduce_sole_executor.cc→src/ops/all_reduce/algorithm/executor/ins_v2_all_reduce_sole_executor.cc+0-0
Rsrc/ops/all_reduce/executor/ins_v2_all_reduce_sole_executor.h→src/ops/all_reduce/algorithm/executor/ins_v2_all_reduce_sole_executor.h+0-0
Rsrc/ops/all_reduce/executor/ins_v2_all_reduce_two_shot_sole_executor.cc→src/ops/all_reduce/algorithm/executor/ins_v2_all_reduce_two_shot_sole_executor.cc+0-0
Rsrc/ops/all_reduce/executor/ins_v2_all_reduce_two_shot_sole_executor.h→src/ops/all_reduce/algorithm/executor/ins_v2_all_reduce_two_shot_sole_executor.h+0-0
Rsrc/ops/all_reduce/executor/ins_v2_allreduce_sequence2die_executor.cc→src/ops/all_reduce/algorithm/executor/ins_v2_allreduce_sequence2die_executor.cc+0-0
Rsrc/ops/all_reduce/executor/ins_v2_allreduce_sequence2die_executor.h→src/ops/all_reduce/algorithm/executor/ins_v2_allreduce_sequence2die_executor.h+0-0
Rsrc/ops/all_reduce/template/CMakeLists.txt→src/ops/all_reduce/algorithm/template/CMakeLists.txt+0-0
Rsrc/ops/all_reduce/template/aicpu/CMakeLists.txt→src/ops/all_reduce/algorithm/template/aicpu/CMakeLists.txt+0-0
Rsrc/ops/all_reduce/template/aicpu/ins_temp_all_gather_mesh_1D_intra.cc→src/ops/all_reduce/algorithm/template/aicpu/ins_temp_all_gather_mesh_1D_intra.cc+0-0
Rsrc/ops/all_reduce/template/aicpu/ins_temp_all_gather_mesh_1D_intra.h→src/ops/all_reduce/algorithm/template/aicpu/ins_temp_all_gather_mesh_1D_intra.h+0-0
Rsrc/ops/all_reduce/template/aicpu/ins_temp_all_gather_nhr_dpu_inter.cc→src/ops/all_reduce/algorithm/template/aicpu/ins_temp_all_gather_nhr_dpu_inter.cc+0-0
Rsrc/ops/all_reduce/template/aicpu/ins_temp_all_gather_nhr_dpu_inter.h→src/ops/all_reduce/algorithm/template/aicpu/ins_temp_all_gather_nhr_dpu_inter.h+0-0
Rsrc/ops/all_reduce/template/aicpu/ins_temp_all_reduce_aicpu_reduce_nhr.cc→src/ops/all_reduce/algorithm/template/aicpu/ins_temp_all_reduce_aicpu_reduce_nhr.cc+0-0
Rsrc/ops/all_reduce/template/aicpu/ins_temp_all_reduce_aicpu_reduce_nhr.h→src/ops/all_reduce/algorithm/template/aicpu/ins_temp_all_reduce_aicpu_reduce_nhr.h+0-0
Rsrc/ops/all_reduce/template/aicpu/ins_temp_all_reduce_mesh_1D_one_shot.cc→src/ops/all_reduce/algorithm/template/aicpu/ins_temp_all_reduce_mesh_1D_one_shot.cc+0-0
Rsrc/ops/all_reduce/template/aicpu/ins_temp_all_reduce_mesh_1D_one_shot.h→src/ops/all_reduce/algorithm/template/aicpu/ins_temp_all_reduce_mesh_1D_one_shot.h+0-0
Rsrc/ops/all_reduce/template/aicpu/ins_temp_all_reduce_mesh_1D_two_shot.cc→src/ops/all_reduce/algorithm/template/aicpu/ins_temp_all_reduce_mesh_1D_two_shot.cc+0-0
Rsrc/ops/all_reduce/template/aicpu/ins_temp_all_reduce_mesh_1D_two_shot.h→src/ops/all_reduce/algorithm/template/aicpu/ins_temp_all_reduce_mesh_1D_two_shot.h+0-0
Rsrc/ops/all_reduce/template/aicpu/ins_temp_all_reduce_mesh_1D_two_shot_mesh_chunk.cc→src/ops/all_reduce/algorithm/template/aicpu/ins_temp_all_reduce_mesh_1D_two_shot_mesh_chunk.cc+0-0
Rsrc/ops/all_reduce/template/aicpu/ins_temp_all_reduce_mesh_1D_two_shot_mesh_chunk.h→src/ops/all_reduce/algorithm/template/aicpu/ins_temp_all_reduce_mesh_1D_two_shot_mesh_chunk.h+0-0
Rsrc/ops/all_reduce/template/aicpu/ins_temp_all_reduce_nhr.cc→src/ops/all_reduce/algorithm/template/aicpu/ins_temp_all_reduce_nhr.cc+0-0
Rsrc/ops/all_reduce/template/aicpu/ins_temp_all_reduce_nhr.h→src/ops/all_reduce/algorithm/template/aicpu/ins_temp_all_reduce_nhr.h+0-0
Rsrc/ops/all_reduce/template/aicpu/ins_temp_reduce_scatter_mesh_1D_dpu_inter.cc→src/ops/all_reduce/algorithm/template/aicpu/ins_temp_reduce_scatter_mesh_1D_dpu_inter.cc+0-0
Rsrc/ops/all_reduce/template/aicpu/ins_temp_reduce_scatter_mesh_1D_dpu_inter.h→src/ops/all_reduce/algorithm/template/aicpu/ins_temp_reduce_scatter_mesh_1D_dpu_inter.h+0-0
Rsrc/ops/all_reduce/template/aicpu/ins_temp_reduce_scatter_mesh_1D_intra.cc→src/ops/all_reduce/algorithm/template/aicpu/ins_temp_reduce_scatter_mesh_1D_intra.cc+1-1
Rsrc/ops/all_reduce/template/aicpu/ins_temp_reduce_scatter_mesh_1D_intra.h→src/ops/all_reduce/algorithm/template/aicpu/ins_temp_reduce_scatter_mesh_1D_intra.h+0-0
Rsrc/ops/all_reduce/template/aiv/CMakeLists.txt→src/ops/all_reduce/algorithm/template/aiv/CMakeLists.txt+0-0
Rsrc/ops/all_reduce/template/aiv/aiv_temp_all_reduce_mesh_1D_oneshot.cc→src/ops/all_reduce/algorithm/template/aiv/aiv_temp_all_reduce_mesh_1D_oneshot.cc+0-0
Rsrc/ops/all_reduce/template/aiv/aiv_temp_all_reduce_mesh_1D_oneshot.h→src/ops/all_reduce/algorithm/template/aiv/aiv_temp_all_reduce_mesh_1D_oneshot.h+0-0
Rsrc/ops/all_reduce/template/aiv/aiv_temp_all_reduce_mesh_1D_twoshot.cc→src/ops/all_reduce/algorithm/template/aiv/aiv_temp_all_reduce_mesh_1D_twoshot.cc+0-0
Rsrc/ops/all_reduce/template/aiv/aiv_temp_all_reduce_mesh_1D_twoshot.h→src/ops/all_reduce/algorithm/template/aiv/aiv_temp_all_reduce_mesh_1D_twoshot.h+0-0
Rsrc/ops/all_reduce/template/aiv/kernel/CMakeLists.txt→src/ops/all_reduce/algorithm/template/aiv/kernel/CMakeLists.txt+52-52
Rsrc/ops/all_reduce/template/aiv/kernel/aiv_all_reduce_mesh_1d_oneshot.h→src/ops/all_reduce/algorithm/template/aiv/kernel/aiv_all_reduce_mesh_1d_oneshot.h+0-0
Rsrc/ops/all_reduce/template/aiv/kernel/aiv_all_reduce_mesh_1d_twoshot.h→src/ops/all_reduce/algorithm/template/aiv/kernel/aiv_all_reduce_mesh_1d_twoshot.h+0-0
Rsrc/ops/all_reduce/template/aiv/kernel/aiv_all_reduce_op.h→src/ops/all_reduce/algorithm/template/aiv/kernel/aiv_all_reduce_op.h+0-0
Rsrc/ops/all_reduce/template/aiv/kernel/aiv_ar_mesh_1d_oneshot_superkernel.h→src/ops/all_reduce/algorithm/template/aiv/kernel/aiv_ar_mesh_1d_oneshot_superkernel.h+0-0
Rsrc/ops/all_reduce/template/aiv/kernel/aiv_ar_mesh_1d_twoshot_superkernel.h→src/ops/all_reduce/algorithm/template/aiv/kernel/aiv_ar_mesh_1d_twoshot_superkernel.h+0-0
Rsrc/ops/all_reduce/template/ccu/CMakeLists.txt→src/ops/all_reduce/algorithm/template/ccu/CMakeLists.txt+0-0
Rsrc/ops/all_reduce/template/ccu/ccu_temp_all_reduce_concurrent_mesh_nhr.cc→src/ops/all_reduce/algorithm/template/ccu/ccu_temp_all_reduce_concurrent_mesh_nhr.cc+0-0
Rsrc/ops/all_reduce/template/ccu/ccu_temp_all_reduce_concurrent_mesh_nhr.h→src/ops/all_reduce/algorithm/template/ccu/ccu_temp_all_reduce_concurrent_mesh_nhr.h+0-0
Rsrc/ops/all_reduce/template/ccu/ccu_temp_all_reduce_mesh_1D.cc→src/ops/all_reduce/algorithm/template/ccu/ccu_temp_all_reduce_mesh_1D.cc+0-0
Rsrc/ops/all_reduce/template/ccu/ccu_temp_all_reduce_mesh_1D.h→src/ops/all_reduce/algorithm/template/ccu/ccu_temp_all_reduce_mesh_1D.h+0-0
Rsrc/ops/all_reduce/template/ccu/ccu_temp_all_reduce_mesh_1D_2die_oneshot.cc→src/ops/all_reduce/algorithm/template/ccu/ccu_temp_all_reduce_mesh_1D_2die_oneshot.cc+0-0
Rsrc/ops/all_reduce/template/ccu/ccu_temp_all_reduce_mesh_1D_2die_oneshot.h→src/ops/all_reduce/algorithm/template/ccu/ccu_temp_all_reduce_mesh_1D_2die_oneshot.h+0-0
Rsrc/ops/all_reduce/template/ccu/ccu_temp_all_reduce_mesh_1D_mem2mem.cc→src/ops/all_reduce/algorithm/template/ccu/ccu_temp_all_reduce_mesh_1D_mem2mem.cc+0-0
Rsrc/ops/all_reduce/template/ccu/ccu_temp_all_reduce_mesh_1D_mem2mem.h→src/ops/all_reduce/algorithm/template/ccu/ccu_temp_all_reduce_mesh_1D_mem2mem.h+0-0
Rsrc/ops/all_reduce/template/ccu/ccu_temp_all_reduce_mesh_1D_mem2mem_2die_oneshot.cc→src/ops/all_reduce/algorithm/template/ccu/ccu_temp_all_reduce_mesh_1D_mem2mem_2die_oneshot.cc+0-0
Rsrc/ops/all_reduce/template/ccu/ccu_temp_all_reduce_mesh_1D_mem2mem_2die_oneshot.h→src/ops/all_reduce/algorithm/template/ccu/ccu_temp_all_reduce_mesh_1D_mem2mem_2die_oneshot.h+0-0
Rsrc/ops/all_reduce/template/ccu/ccu_temp_all_reduce_mesh_1D_one_shot.cc→src/ops/all_reduce/algorithm/template/ccu/ccu_temp_all_reduce_mesh_1D_one_shot.cc+0-0
Rsrc/ops/all_reduce/template/ccu/ccu_temp_all_reduce_mesh_1D_one_shot.h→src/ops/all_reduce/algorithm/template/ccu/ccu_temp_all_reduce_mesh_1D_one_shot.h+0-0
Rsrc/ops/all_reduce/template/ccu/ccu_temp_all_reduce_nhr_1D_mem2mem.cc→src/ops/all_reduce/algorithm/template/ccu/ccu_temp_all_reduce_nhr_1D_mem2mem.cc+0-0
Rsrc/ops/all_reduce/template/ccu/ccu_temp_all_reduce_nhr_1D_mem2mem.h→src/ops/all_reduce/algorithm/template/ccu/ccu_temp_all_reduce_nhr_1D_mem2mem.h+0-0
Rsrc/ops/all_reduce/template/ccu/ccu_temp_all_reduce_nhr_mem2mem_1D_multi_jetty.cc→src/ops/all_reduce/algorithm/template/ccu/ccu_temp_all_reduce_nhr_mem2mem_1D_multi_jetty.cc+0-0
Rsrc/ops/all_reduce/template/ccu/ccu_temp_all_reduce_nhr_mem2mem_1D_multi_jetty.h→src/ops/all_reduce/algorithm/template/ccu/ccu_temp_all_reduce_nhr_mem2mem_1D_multi_jetty.h+0-0
Rsrc/ops/all_reduce/template/ccu/kernel/CMakeLists.txt→src/ops/all_reduce/algorithm/template/ccu/kernel/CMakeLists.txt+0-0
Rsrc/ops/all_reduce/template/ccu/kernel/ccu_kernel_all_reduce_mesh1d.cc→src/ops/all_reduce/algorithm/template/ccu/kernel/ccu_kernel_all_reduce_mesh1d.cc+0-0
Rsrc/ops/all_reduce/template/ccu/kernel/ccu_kernel_all_reduce_mesh1d.h→src/ops/all_reduce/algorithm/template/ccu/kernel/ccu_kernel_all_reduce_mesh1d.h+0-0
Rsrc/ops/all_reduce/template/ccu/kernel/ccu_kernel_all_reduce_mesh1d_mem2mem.cc→src/ops/all_reduce/algorithm/template/ccu/kernel/ccu_kernel_all_reduce_mesh1d_mem2mem.cc+0-0
Rsrc/ops/all_reduce/template/ccu/kernel/ccu_kernel_all_reduce_mesh1d_mem2mem.h→src/ops/all_reduce/algorithm/template/ccu/kernel/ccu_kernel_all_reduce_mesh1d_mem2mem.h+0-0
Rsrc/ops/all_reduce/template/ccu/kernel/ccu_kernel_all_reduce_mesh1d_mem2mem_2die_oneshot.cc→src/ops/all_reduce/algorithm/template/ccu/kernel/ccu_kernel_all_reduce_mesh1d_mem2mem_2die_oneshot.cc+0-0
Rsrc/ops/all_reduce/template/ccu/kernel/ccu_kernel_all_reduce_mesh1d_mem2mem_2die_oneshot.h→src/ops/all_reduce/algorithm/template/ccu/kernel/ccu_kernel_all_reduce_mesh1d_mem2mem_2die_oneshot.h+0-0
Rsrc/ops/all_reduce/template/ccu/kernel/ccu_kernel_all_reduce_mesh1d_one_shot.cc→src/ops/all_reduce/algorithm/template/ccu/kernel/ccu_kernel_all_reduce_mesh1d_one_shot.cc+0-0
Rsrc/ops/all_reduce/template/ccu/kernel/ccu_kernel_all_reduce_mesh1d_one_shot.h→src/ops/all_reduce/algorithm/template/ccu/kernel/ccu_kernel_all_reduce_mesh1d_one_shot.h+0-0
Rsrc/ops/all_reduce/template/ccu/kernel/ccu_kernel_all_reduce_mesh_1D_2die_oneshot.cc→src/ops/all_reduce/algorithm/template/ccu/kernel/ccu_kernel_all_reduce_mesh_1D_2die_oneshot.cc+0-0
Rsrc/ops/all_reduce/template/ccu/kernel/ccu_kernel_all_reduce_mesh_1D_2die_oneshot.h→src/ops/all_reduce/algorithm/template/ccu/kernel/ccu_kernel_all_reduce_mesh_1D_2die_oneshot.h+0-0
Rsrc/ops/all_reduce/template/ccu/kernel/ccu_kernel_all_reduce_nhr1d_mem2mem.cc→src/ops/all_reduce/algorithm/template/ccu/kernel/ccu_kernel_all_reduce_nhr1d_mem2mem.cc+0-0
Rsrc/ops/all_reduce/template/ccu/kernel/ccu_kernel_all_reduce_nhr1d_mem2mem.h→src/ops/all_reduce/algorithm/template/ccu/kernel/ccu_kernel_all_reduce_nhr1d_mem2mem.h+0-0
Rsrc/ops/all_reduce/template/ccu/kernel/ccu_kernel_all_reduce_nhr_mem2mem_1D_multi_jetty.cc→src/ops/all_reduce/algorithm/template/ccu/kernel/ccu_kernel_all_reduce_nhr_mem2mem_1D_multi_jetty.cc+0-0
Rsrc/ops/all_reduce/template/ccu/kernel/ccu_kernel_all_reduce_nhr_mem2mem_1D_multi_jetty.h→src/ops/all_reduce/algorithm/template/ccu/kernel/ccu_kernel_all_reduce_nhr_mem2mem_1D_multi_jetty.h+0-0
Rsrc/ops/all_reduce/all_reduce_op.cc→src/ops/all_reduce/all_reduce.cc+1-1
Rsrc/ops/all_reduce/all_reduce_op.h→src/ops/all_reduce/all_reduce.h+0-0
Rsrc/ops/all_to_all_v/executor/CMakeLists.txt→src/ops/all_to_all_v/algorithm/executor/CMakeLists.txt+0-0
Rsrc/ops/all_to_all_v/executor/ins_all_to_all_v_sole_executor.cc→src/ops/all_to_all_v/algorithm/executor/ins_all_to_all_v_sole_executor.cc+0-0
Rsrc/ops/all_to_all_v/executor/ins_all_to_all_v_sole_executor.h→src/ops/all_to_all_v/algorithm/executor/ins_all_to_all_v_sole_executor.h+0-0
Rsrc/ops/all_to_all_v/executor/ins_v2_aiv_all_to_all_v_sole_executor.cc→src/ops/all_to_all_v/algorithm/executor/ins_v2_aiv_all_to_all_v_sole_executor.cc+0-0
Rsrc/ops/all_to_all_v/executor/ins_v2_aiv_all_to_all_v_sole_executor.h→src/ops/all_to_all_v/algorithm/executor/ins_v2_aiv_all_to_all_v_sole_executor.h+0-0
Rsrc/ops/all_to_all_v/executor/ins_v2_all_to_all_concurrent_executor.cc→src/ops/all_to_all_v/algorithm/executor/ins_v2_all_to_all_concurrent_executor.cc+0-0
Rsrc/ops/all_to_all_v/executor/ins_v2_all_to_all_concurrent_executor.h→src/ops/all_to_all_v/algorithm/executor/ins_v2_all_to_all_concurrent_executor.h+0-0
Rsrc/ops/all_to_all_v/executor/ins_v2_all_to_all_v_concurrent_executor.cc→src/ops/all_to_all_v/algorithm/executor/ins_v2_all_to_all_v_concurrent_executor.cc+0-0
Rsrc/ops/all_to_all_v/executor/ins_v2_all_to_all_v_concurrent_executor.h→src/ops/all_to_all_v/algorithm/executor/ins_v2_all_to_all_v_concurrent_executor.h+0-0
Rsrc/ops/all_to_all_v/executor/ins_v2_all_to_all_v_sole_executor.cc→src/ops/all_to_all_v/algorithm/executor/ins_v2_all_to_all_v_sole_executor.cc+0-0
Rsrc/ops/all_to_all_v/executor/ins_v2_all_to_all_v_sole_executor.h→src/ops/all_to_all_v/algorithm/executor/ins_v2_all_to_all_v_sole_executor.h+0-0
Rsrc/ops/all_to_all_v/template/CMakeLists.txt→src/ops/all_to_all_v/algorithm/template/CMakeLists.txt+0-0
Rsrc/ops/all_to_all_v/template/aicpu/CMakeLists.txt→src/ops/all_to_all_v/algorithm/template/aicpu/CMakeLists.txt+0-0
Rsrc/ops/all_to_all_v/template/aicpu/ins_temp_all_to_all_v_mesh_1D.cc→src/ops/all_to_all_v/algorithm/template/aicpu/ins_temp_all_to_all_v_mesh_1D.cc+0-0
Rsrc/ops/all_to_all_v/template/aicpu/ins_temp_all_to_all_v_mesh_1D.h→src/ops/all_to_all_v/algorithm/template/aicpu/ins_temp_all_to_all_v_mesh_1D.h+0-0
Rsrc/ops/all_to_all_v/template/aicpu/ins_temp_dpu_alltoall_mesh.cc→src/ops/all_to_all_v/algorithm/template/aicpu/ins_temp_dpu_alltoall_mesh.cc+0-0
Rsrc/ops/all_to_all_v/template/aicpu/ins_temp_dpu_alltoall_mesh.h→src/ops/all_to_all_v/algorithm/template/aicpu/ins_temp_dpu_alltoall_mesh.h+0-0
Rsrc/ops/all_to_all_v/template/aicpu/ins_temp_ubx_all_to_all_v_mesh_1D.cc→src/ops/all_to_all_v/algorithm/template/aicpu/ins_temp_ubx_all_to_all_v_mesh_1D.cc+0-0
Rsrc/ops/all_to_all_v/template/aicpu/ins_temp_ubx_all_to_all_v_mesh_1D.h→src/ops/all_to_all_v/algorithm/template/aicpu/ins_temp_ubx_all_to_all_v_mesh_1D.h+0-0
Rsrc/ops/all_to_all_v/template/aiv/CMakeLists.txt→src/ops/all_to_all_v/algorithm/template/aiv/CMakeLists.txt+0-0
Rsrc/ops/all_to_all_v/template/aiv/aiv_temp_all_to_all_mesh_1D.cc→src/ops/all_to_all_v/algorithm/template/aiv/aiv_temp_all_to_all_mesh_1D.cc+0-0
Rsrc/ops/all_to_all_v/template/aiv/aiv_temp_all_to_all_mesh_1D.h→src/ops/all_to_all_v/algorithm/template/aiv/aiv_temp_all_to_all_mesh_1D.h+0-0
Rsrc/ops/all_to_all_v/template/aiv/aiv_temp_all_to_all_v_mesh_1D.cc→src/ops/all_to_all_v/algorithm/template/aiv/aiv_temp_all_to_all_v_mesh_1D.cc+0-0
Rsrc/ops/all_to_all_v/template/aiv/aiv_temp_all_to_all_v_mesh_1D.h→src/ops/all_to_all_v/algorithm/template/aiv/aiv_temp_all_to_all_v_mesh_1D.h+0-0
Rsrc/ops/all_to_all_v/template/aiv/kernel/CMakeLists.txt→src/ops/all_to_all_v/algorithm/template/aiv/kernel/CMakeLists.txt+52-52
Rsrc/ops/all_to_all_v/template/aiv/kernel/aiv_a2a_mesh_1d_superkernel.h→src/ops/all_to_all_v/algorithm/template/aiv/kernel/aiv_a2a_mesh_1d_superkernel.h+0-0
Rsrc/ops/all_to_all_v/template/aiv/kernel/aiv_all_to_all_mesh_1D.h→src/ops/all_to_all_v/algorithm/template/aiv/kernel/aiv_all_to_all_mesh_1D.h+0-0
Rsrc/ops/all_to_all_v/template/aiv/kernel/aiv_all_to_all_op.h→src/ops/all_to_all_v/algorithm/template/aiv/kernel/aiv_all_to_all_op.h+0-0
Rsrc/ops/all_to_all_v/template/aiv/kernel/aiv_all_to_all_v_mesh_1D.h→src/ops/all_to_all_v/algorithm/template/aiv/kernel/aiv_all_to_all_v_mesh_1D.h+0-0
Rsrc/ops/all_to_all_v/template/aiv/kernel/aiv_all_to_all_v_op.h→src/ops/all_to_all_v/algorithm/template/aiv/kernel/aiv_all_to_all_v_op.h+0-0
Rsrc/ops/all_to_all_v/template/ccu/CMakeLists.txt→src/ops/all_to_all_v/algorithm/template/ccu/CMakeLists.txt+0-0
Rsrc/ops/all_to_all_v/template/ccu/ccu_temp_all_to_all_concurrent_mesh_nhr.cc→src/ops/all_to_all_v/algorithm/template/ccu/ccu_temp_all_to_all_concurrent_mesh_nhr.cc+0-0
Rsrc/ops/all_to_all_v/template/ccu/ccu_temp_all_to_all_concurrent_mesh_nhr.h→src/ops/all_to_all_v/algorithm/template/ccu/ccu_temp_all_to_all_concurrent_mesh_nhr.h+0-0
Rsrc/ops/all_to_all_v/template/ccu/ccu_temp_all_to_all_mesh1d_2Die.cc→src/ops/all_to_all_v/algorithm/template/ccu/ccu_temp_all_to_all_mesh1d_2Die.cc+0-0
Rsrc/ops/all_to_all_v/template/ccu/ccu_temp_all_to_all_mesh1d_2Die.h→src/ops/all_to_all_v/algorithm/template/ccu/ccu_temp_all_to_all_mesh1d_2Die.h+0-0
Rsrc/ops/all_to_all_v/template/ccu/ccu_temp_all_to_all_mesh1d_multi_jetty.cc→src/ops/all_to_all_v/algorithm/template/ccu/ccu_temp_all_to_all_mesh1d_multi_jetty.cc+0-0
Rsrc/ops/all_to_all_v/template/ccu/ccu_temp_all_to_all_mesh1d_multi_jetty.h→src/ops/all_to_all_v/algorithm/template/ccu/ccu_temp_all_to_all_mesh1d_multi_jetty.h+0-0
Rsrc/ops/all_to_all_v/template/ccu/ccu_temp_all_to_all_mesh2die.cc→src/ops/all_to_all_v/algorithm/template/ccu/ccu_temp_all_to_all_mesh2die.cc+0-0
Rsrc/ops/all_to_all_v/template/ccu/ccu_temp_all_to_all_mesh2die.h→src/ops/all_to_all_v/algorithm/template/ccu/ccu_temp_all_to_all_mesh2die.h+0-0
Rsrc/ops/all_to_all_v/template/ccu/ccu_temp_all_to_all_mesh_1D.cc→src/ops/all_to_all_v/algorithm/template/ccu/ccu_temp_all_to_all_mesh_1D.cc+0-0
Rsrc/ops/all_to_all_v/template/ccu/ccu_temp_all_to_all_mesh_1D.h→src/ops/all_to_all_v/algorithm/template/ccu/ccu_temp_all_to_all_mesh_1D.h+0-0
Rsrc/ops/all_to_all_v/template/ccu/ccu_temp_all_to_all_v_mesh1d_2Die.cc→src/ops/all_to_all_v/algorithm/template/ccu/ccu_temp_all_to_all_v_mesh1d_2Die.cc+0-0
Rsrc/ops/all_to_all_v/template/ccu/ccu_temp_all_to_all_v_mesh1d_2Die.h→src/ops/all_to_all_v/algorithm/template/ccu/ccu_temp_all_to_all_v_mesh1d_2Die.h+0-0
Rsrc/ops/all_to_all_v/template/ccu/ccu_temp_all_to_all_v_mesh2die.cc→src/ops/all_to_all_v/algorithm/template/ccu/ccu_temp_all_to_all_v_mesh2die.cc+0-0
Rsrc/ops/all_to_all_v/template/ccu/ccu_temp_all_to_all_v_mesh2die.h→src/ops/all_to_all_v/algorithm/template/ccu/ccu_temp_all_to_all_v_mesh2die.h+0-0
Rsrc/ops/all_to_all_v/template/ccu/ccu_temp_all_to_all_v_mesh_1D.cc→src/ops/all_to_all_v/algorithm/template/ccu/ccu_temp_all_to_all_v_mesh_1D.cc+0-0
Rsrc/ops/all_to_all_v/template/ccu/ccu_temp_all_to_all_v_mesh_1D.h→src/ops/all_to_all_v/algorithm/template/ccu/ccu_temp_all_to_all_v_mesh_1D.h+0-0
Rsrc/ops/all_to_all_v/template/ccu/ccu_temp_all_to_all_v_mesh_1D_multi_jetty.cc→src/ops/all_to_all_v/algorithm/template/ccu/ccu_temp_all_to_all_v_mesh_1D_multi_jetty.cc+0-0
Rsrc/ops/all_to_all_v/template/ccu/ccu_temp_all_to_all_v_mesh_1D_multi_jetty.h→src/ops/all_to_all_v/algorithm/template/ccu/ccu_temp_all_to_all_v_mesh_1D_multi_jetty.h+0-0
Rsrc/ops/all_to_all_v/template/ccu/kernel/ccu_kernel_all_to_all_mesh1d.cc→src/ops/all_to_all_v/algorithm/template/ccu/kernel/ccu_kernel_all_to_all_mesh1d.cc+0-0
Rsrc/ops/all_to_all_v/template/ccu/kernel/ccu_kernel_all_to_all_mesh1d.h→src/ops/all_to_all_v/algorithm/template/ccu/kernel/ccu_kernel_all_to_all_mesh1d.h+0-0
Rsrc/ops/all_to_all_v/template/ccu/kernel/ccu_kernel_all_to_all_mesh1d_multi_jetty.cc→src/ops/all_to_all_v/algorithm/template/ccu/kernel/ccu_kernel_all_to_all_mesh1d_multi_jetty.cc+0-0
Rsrc/ops/all_to_all_v/template/ccu/kernel/ccu_kernel_all_to_all_mesh1d_multi_jetty.h→src/ops/all_to_all_v/algorithm/template/ccu/kernel/ccu_kernel_all_to_all_mesh1d_multi_jetty.h+0-0
Rsrc/ops/all_to_all_v/template/ccu/kernel/ccu_kernel_all_to_all_mesh2die.cc→src/ops/all_to_all_v/algorithm/template/ccu/kernel/ccu_kernel_all_to_all_mesh2die.cc+0-0
Rsrc/ops/all_to_all_v/template/ccu/kernel/ccu_kernel_all_to_all_mesh2die.h→src/ops/all_to_all_v/algorithm/template/ccu/kernel/ccu_kernel_all_to_all_mesh2die.h+0-0
Rsrc/ops/all_to_all_v/template/ccu/kernel/ccu_kernel_all_to_all_v_mesh1d.cc→src/ops/all_to_all_v/algorithm/template/ccu/kernel/ccu_kernel_all_to_all_v_mesh1d.cc+0-0
Rsrc/ops/all_to_all_v/template/ccu/kernel/ccu_kernel_all_to_all_v_mesh1d.h→src/ops/all_to_all_v/algorithm/template/ccu/kernel/ccu_kernel_all_to_all_v_mesh1d.h+0-0
Rsrc/ops/all_to_all_v/template/ccu/kernel/ccu_kernel_all_to_all_v_mesh1d_2die.cc→src/ops/all_to_all_v/algorithm/template/ccu/kernel/ccu_kernel_all_to_all_v_mesh1d_2die.cc+0-0
Rsrc/ops/all_to_all_v/template/ccu/kernel/ccu_kernel_all_to_all_v_mesh1d_2die.h→src/ops/all_to_all_v/algorithm/template/ccu/kernel/ccu_kernel_all_to_all_v_mesh1d_2die.h+0-0
Rsrc/ops/all_to_all_v/template/ccu/kernel/ccu_kernel_all_to_all_v_mesh1d_multi_jetty.cc→src/ops/all_to_all_v/algorithm/template/ccu/kernel/ccu_kernel_all_to_all_v_mesh1d_multi_jetty.cc+0-0
Rsrc/ops/all_to_all_v/template/ccu/kernel/ccu_kernel_all_to_all_v_mesh1d_multi_jetty.h→src/ops/all_to_all_v/algorithm/template/ccu/kernel/ccu_kernel_all_to_all_v_mesh1d_multi_jetty.h+0-0
Rsrc/ops/all_to_all_v/template/ccu/kernel/ccu_kernel_all_to_all_v_mesh2die.cc→src/ops/all_to_all_v/algorithm/template/ccu/kernel/ccu_kernel_all_to_all_v_mesh2die.cc+0-0
Rsrc/ops/all_to_all_v/template/ccu/kernel/ccu_kernel_all_to_all_v_mesh2die.h→src/ops/all_to_all_v/algorithm/template/ccu/kernel/ccu_kernel_all_to_all_v_mesh2die.h+0-0
Rsrc/ops/all_to_all_v/all_to_all_v_op.cc→src/ops/all_to_all_v/all_to_all_v.cc+1-1
Rsrc/ops/all_to_all_v/all_to_all_v_op.h→src/ops/all_to_all_v/all_to_all_v.h+0-0
Msrc/ops/barrier/CMakeLists.txt+2-3
Rsrc/ops/barrier/executor/CMakeLists.txt→src/ops/barrier/algorithm/executor/CMakeLists.txt+0-0
Rsrc/ops/barrier/executor/ins_v2_barrier_sequence_executor.cc→src/ops/barrier/algorithm/executor/ins_v2_barrier_sequence_executor.cc+0-0
Rsrc/ops/barrier/executor/ins_v2_barrier_sequence_executor.h→src/ops/barrier/algorithm/executor/ins_v2_barrier_sequence_executor.h+0-0
Rsrc/ops/barrier/executor/ins_v2_barrier_sole_executor.cc→src/ops/barrier/algorithm/executor/ins_v2_barrier_sole_executor.cc+0-0
Rsrc/ops/barrier/executor/ins_v2_barrier_sole_executor.h→src/ops/barrier/algorithm/executor/ins_v2_barrier_sole_executor.h+0-0
Rsrc/ops/barrier/template/CMakeLists.txt→src/ops/barrier/algorithm/template/CMakeLists.txt+0-0
Rsrc/ops/barrier/template/aicpu/CMakeLists.txt→src/ops/barrier/algorithm/template/aicpu/CMakeLists.txt+0-0
Rsrc/ops/barrier/template/aicpu/ins_temp_barrier_mesh_1D.cc→src/ops/barrier/algorithm/template/aicpu/ins_temp_barrier_mesh_1D.cc+0-0
Rsrc/ops/barrier/template/aicpu/ins_temp_barrier_mesh_1D.h→src/ops/barrier/algorithm/template/aicpu/ins_temp_barrier_mesh_1D.h+0-0
Rsrc/ops/barrier/template/aicpu/ins_temp_barrier_nhr_aicpu.cc→src/ops/barrier/algorithm/template/aicpu/ins_temp_barrier_nhr_aicpu.cc+0-0
Rsrc/ops/barrier/template/aicpu/ins_temp_barrier_nhr_aicpu.h→src/ops/barrier/algorithm/template/aicpu/ins_temp_barrier_nhr_aicpu.h+0-0
Rsrc/ops/barrier/template/aicpu/ins_temp_barrier_nhr_dpu.cc→src/ops/barrier/algorithm/template/aicpu/ins_temp_barrier_nhr_dpu.cc+0-0
Rsrc/ops/barrier/template/aicpu/ins_temp_barrier_nhr_dpu.h→src/ops/barrier/algorithm/template/aicpu/ins_temp_barrier_nhr_dpu.h+0-0
Rsrc/ops/barrier/barrier_op.cc→src/ops/barrier/barrier.cc+1-1
Rsrc/ops/barrier/barrier_op.h→src/ops/barrier/barrier.h+0-0
Rsrc/ops/batch_send_recv/executor/CMakeLists.txt→src/ops/batch_send_recv/algorithm/executor/CMakeLists.txt+0-0
Rsrc/ops/batch_send_recv/executor/ins_v2_batch_send_recv_executor.cc→src/ops/batch_send_recv/algorithm/executor/ins_v2_batch_send_recv_executor.cc+0-0
Rsrc/ops/batch_send_recv/executor/ins_v2_batch_send_recv_executor.h→src/ops/batch_send_recv/algorithm/executor/ins_v2_batch_send_recv_executor.h+0-0
Rsrc/ops/batch_send_recv/executor/ins_v2_batch_send_recv_sole_executor.cc→src/ops/batch_send_recv/algorithm/executor/ins_v2_batch_send_recv_sole_executor.cc+0-0
Rsrc/ops/batch_send_recv/executor/ins_v2_batch_send_recv_sole_executor.h→src/ops/batch_send_recv/algorithm/executor/ins_v2_batch_send_recv_sole_executor.h+0-0
Rsrc/ops/batch_send_recv/template/CMakeLists.txt→src/ops/batch_send_recv/algorithm/template/CMakeLists.txt+0-0
Rsrc/ops/batch_send_recv/template/ins_temp_batch_send_recv_dpu.cc→src/ops/batch_send_recv/algorithm/template/ins_temp_batch_send_recv_dpu.cc+0-0
Rsrc/ops/batch_send_recv/template/ins_temp_batch_send_recv_dpu.h→src/ops/batch_send_recv/algorithm/template/ins_temp_batch_send_recv_dpu.h+0-0
Rsrc/ops/batch_send_recv/batch_send_recv_op.cc→src/ops/batch_send_recv/batch_send_recv.cc+1-1
Rsrc/ops/batch_send_recv/batch_send_recv_op.h→src/ops/batch_send_recv/batch_send_recv.h+0-0
Rsrc/ops/broadcast/executor/CMakeLists.txt→src/ops/broadcast/algorithm/executor/CMakeLists.txt+0-0
Rsrc/ops/broadcast/executor/ins_v2_broadcast_concurrent_executor.cc→src/ops/broadcast/algorithm/executor/ins_v2_broadcast_concurrent_executor.cc+0-0
Rsrc/ops/broadcast/executor/ins_v2_broadcast_concurrent_executor.h→src/ops/broadcast/algorithm/executor/ins_v2_broadcast_concurrent_executor.h+0-0
Rsrc/ops/broadcast/executor/ins_v2_broadcast_omnipipe_2d_executor.cc→src/ops/broadcast/algorithm/executor/ins_v2_broadcast_omnipipe_2d_executor.cc+0-0
Rsrc/ops/broadcast/executor/ins_v2_broadcast_omnipipe_2d_executor.h→src/ops/broadcast/algorithm/executor/ins_v2_broadcast_omnipipe_2d_executor.h+0-0
Rsrc/ops/broadcast/executor/ins_v2_broadcast_parallel_executor.cc→src/ops/broadcast/algorithm/executor/ins_v2_broadcast_parallel_executor.cc+0-0
Rsrc/ops/broadcast/executor/ins_v2_broadcast_parallel_executor.h→src/ops/broadcast/algorithm/executor/ins_v2_broadcast_parallel_executor.h+0-0
Rsrc/ops/broadcast/executor/ins_v2_broadcast_sequence_executor.cc→src/ops/broadcast/algorithm/executor/ins_v2_broadcast_sequence_executor.cc+0-0
Rsrc/ops/broadcast/executor/ins_v2_broadcast_sequence_executor.h→src/ops/broadcast/algorithm/executor/ins_v2_broadcast_sequence_executor.h+0-0
Rsrc/ops/broadcast/executor/ins_v2_broadcast_sequence_executor_aicpu_3level.cc→src/ops/broadcast/algorithm/executor/ins_v2_broadcast_sequence_executor_aicpu_3level.cc+0-0
Rsrc/ops/broadcast/executor/ins_v2_broadcast_sequence_executor_aicpu_3level.h→src/ops/broadcast/algorithm/executor/ins_v2_broadcast_sequence_executor_aicpu_3level.h+0-0
Rsrc/ops/broadcast/executor/ins_v2_broadcast_sole_executor.cc→src/ops/broadcast/algorithm/executor/ins_v2_broadcast_sole_executor.cc+0-0
Rsrc/ops/broadcast/executor/ins_v2_broadcast_sole_executor.h→src/ops/broadcast/algorithm/executor/ins_v2_broadcast_sole_executor.h+0-0
Rsrc/ops/broadcast/template/CMakeLists.txt→src/ops/broadcast/algorithm/template/CMakeLists.txt+0-0
Rsrc/ops/broadcast/template/aicpu/CMakeLists.txt→src/ops/broadcast/algorithm/template/aicpu/CMakeLists.txt+0-0
Rsrc/ops/broadcast/template/aicpu/ins_temp_allgather_mesh_1D_intra.cc→src/ops/broadcast/algorithm/template/aicpu/ins_temp_allgather_mesh_1D_intra.cc+0-0
Rsrc/ops/broadcast/template/aicpu/ins_temp_allgather_mesh_1D_intra.h→src/ops/broadcast/algorithm/template/aicpu/ins_temp_allgather_mesh_1D_intra.h+0-0
Rsrc/ops/broadcast/template/aicpu/ins_temp_allgather_nhr_dpu_inter.cc→src/ops/broadcast/algorithm/template/aicpu/ins_temp_allgather_nhr_dpu_inter.cc+0-0
Rsrc/ops/broadcast/template/aicpu/ins_temp_allgather_nhr_dpu_inter.h→src/ops/broadcast/algorithm/template/aicpu/ins_temp_allgather_nhr_dpu_inter.h+0-0
Rsrc/ops/broadcast/template/aicpu/ins_temp_broadcast_mesh_1D_two_shot.cc→src/ops/broadcast/algorithm/template/aicpu/ins_temp_broadcast_mesh_1D_two_shot.cc+0-0
Rsrc/ops/broadcast/template/aicpu/ins_temp_broadcast_mesh_1D_two_shot.h→src/ops/broadcast/algorithm/template/aicpu/ins_temp_broadcast_mesh_1D_two_shot.h+0-0
Rsrc/ops/broadcast/template/aicpu/ins_temp_broadcast_nhr.cc→src/ops/broadcast/algorithm/template/aicpu/ins_temp_broadcast_nhr.cc+0-0
Rsrc/ops/broadcast/template/aicpu/ins_temp_broadcast_nhr.h→src/ops/broadcast/algorithm/template/aicpu/ins_temp_broadcast_nhr.h+0-0
Rsrc/ops/broadcast/template/aicpu/ins_temp_scatter_mesh_1D_intra.cc→src/ops/broadcast/algorithm/template/aicpu/ins_temp_scatter_mesh_1D_intra.cc+0-0
Rsrc/ops/broadcast/template/aicpu/ins_temp_scatter_mesh_1D_intra.h→src/ops/broadcast/algorithm/template/aicpu/ins_temp_scatter_mesh_1D_intra.h+0-0
Rsrc/ops/broadcast/template/aicpu/ins_temp_scatter_nhr_dpu_inter.cc→src/ops/broadcast/algorithm/template/aicpu/ins_temp_scatter_nhr_dpu_inter.cc+0-0
Rsrc/ops/broadcast/template/aicpu/ins_temp_scatter_nhr_dpu_inter.h→src/ops/broadcast/algorithm/template/aicpu/ins_temp_scatter_nhr_dpu_inter.h+0-0
Rsrc/ops/broadcast/template/aiv/CMakeLists.txt→src/ops/broadcast/algorithm/template/aiv/CMakeLists.txt+0-0
Rsrc/ops/broadcast/template/aiv/aiv_temp_broadcast_mesh_1D.cc→src/ops/broadcast/algorithm/template/aiv/aiv_temp_broadcast_mesh_1D.cc+0-0
Rsrc/ops/broadcast/template/aiv/aiv_temp_broadcast_mesh_1D.h→src/ops/broadcast/algorithm/template/aiv/aiv_temp_broadcast_mesh_1D.h+0-0
Rsrc/ops/broadcast/template/aiv/kernel/CMakeLists.txt→src/ops/broadcast/algorithm/template/aiv/kernel/CMakeLists.txt+4-4
Rsrc/ops/broadcast/template/aiv/kernel/aiv_broadcast_mesh_1d.h→src/ops/broadcast/algorithm/template/aiv/kernel/aiv_broadcast_mesh_1d.h+0-0
Rsrc/ops/broadcast/template/aiv/kernel/aiv_broadcast_op.h→src/ops/broadcast/algorithm/template/aiv/kernel/aiv_broadcast_op.h+0-0
Rsrc/ops/broadcast/template/ccu/CMakeLists.txt→src/ops/broadcast/algorithm/template/ccu/CMakeLists.txt+0-0
Rsrc/ops/broadcast/template/ccu/ccu_temp_broadcast_mesh_1D.cc→src/ops/broadcast/algorithm/template/ccu/ccu_temp_broadcast_mesh_1D.cc+0-0
Rsrc/ops/broadcast/template/ccu/ccu_temp_broadcast_mesh_1D.h→src/ops/broadcast/algorithm/template/ccu/ccu_temp_broadcast_mesh_1D.h+0-0
Rsrc/ops/broadcast/template/ccu/ccu_temp_broadcast_mesh_1D_mem2mem.cc→src/ops/broadcast/algorithm/template/ccu/ccu_temp_broadcast_mesh_1D_mem2mem.cc+0-0
Rsrc/ops/broadcast/template/ccu/ccu_temp_broadcast_mesh_1D_mem2mem.h→src/ops/broadcast/algorithm/template/ccu/ccu_temp_broadcast_mesh_1D_mem2mem.h+0-0
Rsrc/ops/broadcast/template/ccu/ccu_temp_broadcast_nhr_1D_mem2mem.cc→src/ops/broadcast/algorithm/template/ccu/ccu_temp_broadcast_nhr_1D_mem2mem.cc+0-0
Rsrc/ops/broadcast/template/ccu/ccu_temp_broadcast_nhr_1D_mem2mem.h→src/ops/broadcast/algorithm/template/ccu/ccu_temp_broadcast_nhr_1D_mem2mem.h+0-0
Rsrc/ops/broadcast/template/ccu/kernel/CMakeLists.txt→src/ops/broadcast/algorithm/template/ccu/kernel/CMakeLists.txt+0-0
Rsrc/ops/broadcast/template/ccu/kernel/ccu_kernel_broadcast_mesh1d.cc→src/ops/broadcast/algorithm/template/ccu/kernel/ccu_kernel_broadcast_mesh1d.cc+0-0
Rsrc/ops/broadcast/template/ccu/kernel/ccu_kernel_broadcast_mesh1d.h→src/ops/broadcast/algorithm/template/ccu/kernel/ccu_kernel_broadcast_mesh1d.h+0-0
Rsrc/ops/broadcast/template/ccu/kernel/ccu_kernel_broadcast_mesh1d_mem2mem.cc→src/ops/broadcast/algorithm/template/ccu/kernel/ccu_kernel_broadcast_mesh1d_mem2mem.cc+0-0
Rsrc/ops/broadcast/template/ccu/kernel/ccu_kernel_broadcast_mesh1d_mem2mem.h→src/ops/broadcast/algorithm/template/ccu/kernel/ccu_kernel_broadcast_mesh1d_mem2mem.h+0-0
Rsrc/ops/broadcast/template/ccu/kernel/ccu_kernel_broadcast_nhr1d_mem2mem.cc→src/ops/broadcast/algorithm/template/ccu/kernel/ccu_kernel_broadcast_nhr1d_mem2mem.cc+0-0
Rsrc/ops/broadcast/template/ccu/kernel/ccu_kernel_broadcast_nhr1d_mem2mem.h→src/ops/broadcast/algorithm/template/ccu/kernel/ccu_kernel_broadcast_nhr1d_mem2mem.h+0-0
Rsrc/ops/broadcast/broadcast_op.cc→src/ops/broadcast/broadcast.cc+1-1
Rsrc/ops/broadcast/broadcast_op.h→src/ops/broadcast/broadcast.h+0-0
Rsrc/ops/op_common/executor/CMakeLists.txt→src/ops/op_common/algorithm/executor/CMakeLists.txt+0-0
Rsrc/ops/op_common/executor/channel/CMakeLists.txt→src/ops/op_common/algorithm/executor/channel/CMakeLists.txt+0-0
Rsrc/ops/op_common/executor/channel/channel.cc→src/ops/op_common/algorithm/executor/channel/channel.cc+0-0
Rsrc/ops/op_common/executor/channel/channel.h→src/ops/op_common/algorithm/executor/channel/channel.h+0-0
Rsrc/ops/op_common/executor/channel/channel_request.cc→src/ops/op_common/algorithm/executor/channel/channel_request.cc+0-0
Rsrc/ops/op_common/executor/channel/channel_request.h→src/ops/op_common/algorithm/executor/channel/channel_request.h+0-0
Rsrc/ops/op_common/executor/executor_base.cc→src/ops/op_common/algorithm/executor/executor_base.cc+0-0
Rsrc/ops/op_common/executor/executor_base.h→src/ops/op_common/algorithm/executor/executor_base.h+0-0
Rsrc/ops/op_common/executor/executor_common_ops.h→src/ops/op_common/algorithm/executor/executor_common_ops.h+0-0
Rsrc/ops/op_common/executor/executor_v2_base.cc→src/ops/op_common/algorithm/executor/executor_v2_base.cc+0-0
Rsrc/ops/op_common/executor/executor_v2_base.h→src/ops/op_common/algorithm/executor/executor_v2_base.h+0-0
Rsrc/ops/op_common/executor/registry/CMakeLists.txt→src/ops/op_common/algorithm/executor/registry/CMakeLists.txt+0-0
Rsrc/ops/op_common/executor/registry/coll_alg_exec_registry.cc→src/ops/op_common/algorithm/executor/registry/coll_alg_exec_registry.cc+0-0
Rsrc/ops/op_common/executor/registry/coll_alg_exec_registry.h→src/ops/op_common/algorithm/executor/registry/coll_alg_exec_registry.h+0-0
Rsrc/ops/op_common/executor/registry/coll_alg_v2_exec_registry.cc→src/ops/op_common/algorithm/executor/registry/coll_alg_v2_exec_registry.cc+0-0
Rsrc/ops/op_common/executor/registry/coll_alg_v2_exec_registry.h→src/ops/op_common/algorithm/executor/registry/coll_alg_v2_exec_registry.h+0-0
Rsrc/ops/op_common/template/CMakeLists.txt→src/ops/op_common/algorithm/template/CMakeLists.txt+0-0
Rsrc/ops/op_common/template/aicpu/CMakeLists.txt→src/ops/op_common/algorithm/template/aicpu/CMakeLists.txt+0-0
Rsrc/ops/op_common/template/aicpu/dfx/task_exception_fun.cc→src/ops/op_common/algorithm/template/aicpu/dfx/task_exception_fun.cc+0-0
Rsrc/ops/op_common/template/aicpu/dfx/task_exception_fun.h→src/ops/op_common/algorithm/template/aicpu/dfx/task_exception_fun.h+0-0
Rsrc/ops/op_common/template/aicpu/kernel_launch.cc→src/ops/op_common/algorithm/template/aicpu/kernel_launch.cc+0-0
Rsrc/ops/op_common/template/aicpu/kernel_launch.h→src/ops/op_common/algorithm/template/aicpu/kernel_launch.h+0-0
Rsrc/ops/op_common/template/aicpu/load_kernel.cc→src/ops/op_common/algorithm/template/aicpu/load_kernel.cc+0-0
Rsrc/ops/op_common/template/aicpu/load_kernel.h→src/ops/op_common/algorithm/template/aicpu/load_kernel.h+0-0
Rsrc/ops/op_common/template/aicpu/task_cache/aicpu_task_cache_comm_manager.cc→src/ops/op_common/algorithm/template/aicpu/task_cache/aicpu_task_cache_comm_manager.cc+0-0
Rsrc/ops/op_common/template/aicpu/task_cache/aicpu_task_cache_comm_manager.h→src/ops/op_common/algorithm/template/aicpu/task_cache/aicpu_task_cache_comm_manager.h+0-0
Rsrc/ops/op_common/template/aicpu/task_cache/aicpu_task_cache_key.cc→src/ops/op_common/algorithm/template/aicpu/task_cache/aicpu_task_cache_key.cc+0-0
Rsrc/ops/op_common/template/aicpu/task_cache/aicpu_task_cache_key.h→src/ops/op_common/algorithm/template/aicpu/task_cache/aicpu_task_cache_key.h+0-0
Rsrc/ops/op_common/template/aicpu/task_cache/aicpu_task_cache_policy.cc→src/ops/op_common/algorithm/template/aicpu/task_cache/aicpu_task_cache_policy.cc+0-0
Rsrc/ops/op_common/template/aicpu/task_cache/aicpu_task_cache_policy.h→src/ops/op_common/algorithm/template/aicpu/task_cache/aicpu_task_cache_policy.h+0-0
Rsrc/ops/op_common/template/aicpu/task_cache/aicpu_task_cache_utils.cc→src/ops/op_common/algorithm/template/aicpu/task_cache/aicpu_task_cache_utils.cc+0-0
Rsrc/ops/op_common/template/aicpu/task_cache/aicpu_task_cache_utils.h→src/ops/op_common/algorithm/template/aicpu/task_cache/aicpu_task_cache_utils.h+0-0
Rsrc/ops/op_common/template/aicpu_task_cache_clear.cc→src/ops/op_common/algorithm/template/aicpu_task_cache_clear.cc+0-0
Rsrc/ops/op_common/template/aiv/CMakeLists.txt→src/ops/op_common/algorithm/template/aiv/CMakeLists.txt+0-0
Rsrc/ops/op_common/template/aiv/aiv_communication_base_v2.h→src/ops/op_common/algorithm/template/aiv/aiv_communication_base_v2.h+0-0
Rsrc/ops/op_common/template/aiv/aiv_defines.h→src/ops/op_common/algorithm/template/aiv/aiv_defines.h+0-0
Rsrc/ops/op_common/template/aiv/aiv_interface/sync_interface.h→src/ops/op_common/algorithm/template/aiv/aiv_interface/sync_interface.h+0-0
Rsrc/ops/op_common/template/aiv/aiv_kernel_def.h→src/ops/op_common/algorithm/template/aiv/aiv_kernel_def.h+0-0
Rsrc/ops/op_common/template/aiv/hccl_aiv_utils.cc→src/ops/op_common/algorithm/template/aiv/hccl_aiv_utils.cc+0-0
Rsrc/ops/op_common/template/aiv/hccl_aiv_utils.h→src/ops/op_common/algorithm/template/aiv/hccl_aiv_utils.h+0-0
Rsrc/ops/op_common/template/aiv_alg_template_base.cc→src/ops/op_common/algorithm/template/aiv_alg_template_base.cc+0-0
Rsrc/ops/op_common/template/aiv_alg_template_base.h→src/ops/op_common/algorithm/template/aiv_alg_template_base.h+0-0
Rsrc/ops/op_common/template/alg_template_base.cc→src/ops/op_common/algorithm/template/alg_template_base.cc+0-0
Rsrc/ops/op_common/template/alg_template_base.h→src/ops/op_common/algorithm/template/alg_template_base.h+0-0
Rsrc/ops/op_common/template/alg_v2_template_base.cc→src/ops/op_common/algorithm/template/alg_v2_template_base.cc+0-0
Rsrc/ops/op_common/template/alg_v2_template_base.h→src/ops/op_common/algorithm/template/alg_v2_template_base.h+0-0
Rsrc/ops/op_common/template/ccu/CMakeLists.txt→src/ops/op_common/algorithm/template/ccu/CMakeLists.txt+0-0
Rsrc/ops/op_common/template/ccu/ccu_kernel_alg_base.cc→src/ops/op_common/algorithm/template/ccu/ccu_kernel_alg_base.cc+0-0
Rsrc/ops/op_common/template/ccu/ccu_kernel_alg_base.h→src/ops/op_common/algorithm/template/ccu/ccu_kernel_alg_base.h+0-0
Rsrc/ops/op_common/template/ccu/ccu_kernel_utils.cc→src/ops/op_common/algorithm/template/ccu/ccu_kernel_utils.cc+0-0
Rsrc/ops/op_common/template/ccu/ccu_kernel_utils.h→src/ops/op_common/algorithm/template/ccu/ccu_kernel_utils.h+0-0
Rsrc/ops/op_common/template/ccu_alg_template_base.cc→src/ops/op_common/algorithm/template/ccu_alg_template_base.cc+0-0
Rsrc/ops/op_common/template/ccu_alg_template_base.h→src/ops/op_common/algorithm/template/ccu_alg_template_base.h+0-0
Rsrc/ops/op_common/template/common_alg_template_base.h→src/ops/op_common/algorithm/template/common_alg_template_base.h+0-0
Rsrc/ops/op_common/template/dpu/CMakeLists.txt→src/ops/op_common/algorithm/template/dpu/CMakeLists.txt+0-0
Rsrc/ops/op_common/template/dpu/kernel_launch.cc→src/ops/op_common/algorithm/template/dpu/kernel_launch.cc+0-0
Rsrc/ops/op_common/template/dpu/kernel_launch.h→src/ops/op_common/algorithm/template/dpu/kernel_launch.h+0-0
Rsrc/ops/op_common/template/omnipipe_template_utils.cc→src/ops/op_common/algorithm/template/omnipipe_template_utils.cc+0-0
Rsrc/ops/op_common/template/omnipipe_template_utils.h→src/ops/op_common/algorithm/template/omnipipe_template_utils.h+0-0
Rsrc/ops/op_common/template/registry/CMakeLists.txt→src/ops/op_common/algorithm/template/registry/CMakeLists.txt+0-0
Rsrc/ops/op_common/template/registry/alg_template_register.cc→src/ops/op_common/algorithm/template/registry/alg_template_register.cc+0-0
Rsrc/ops/op_common/template/registry/alg_template_register.h→src/ops/op_common/algorithm/template/registry/alg_template_register.h+0-0
Rsrc/ops/op_common/template/registry/alg_v2_template_register.cc→src/ops/op_common/algorithm/template/registry/alg_v2_template_register.cc+0-0
Rsrc/ops/op_common/template/registry/alg_v2_template_register.h→src/ops/op_common/algorithm/template/registry/alg_v2_template_register.h+0-0
Rsrc/ops/op_common/template/template_utils.cc→src/ops/op_common/algorithm/template/template_utils.cc+0-0
Rsrc/ops/op_common/template/template_utils.h→src/ops/op_common/algorithm/template/template_utils.h+0-0
Rsrc/ops/op_common/template/wrapper/CMakeLists.txt→src/ops/op_common/algorithm/template/wrapper/CMakeLists.txt+0-0
Rsrc/ops/op_common/template/wrapper/alg_data_trans_wrapper.cc→src/ops/op_common/algorithm/template/wrapper/alg_data_trans_wrapper.cc+0-0
Rsrc/ops/op_common/template/wrapper/alg_data_trans_wrapper.h→src/ops/op_common/algorithm/template/wrapper/alg_data_trans_wrapper.h+0-0
Rsrc/ops/op_common/template/wrapper/dpu_alg_data_trans_wrapper.cc→src/ops/op_common/algorithm/template/wrapper/dpu_alg_data_trans_wrapper.cc+0-0
Rsrc/ops/op_common/template/wrapper/dpu_alg_data_trans_wrapper.h→src/ops/op_common/algorithm/template/wrapper/dpu_alg_data_trans_wrapper.h+0-0
Rsrc/ops/op_common/template/wrapper/dpu_alg_nhr_opt_wrapper.cc→src/ops/op_common/algorithm/template/wrapper/dpu_alg_nhr_opt_wrapper.cc+0-0
Rsrc/ops/op_common/template/wrapper/dpu_alg_nhr_opt_wrapper.h→src/ops/op_common/algorithm/template/wrapper/dpu_alg_nhr_opt_wrapper.h+0-0
Rsrc/ops/op_common/topo/CMakeLists.txt→src/ops/op_common/algorithm/topo_match/CMakeLists.txt+0-2
Rsrc/ops/op_common/topo/physical_level.h→src/ops/op_common/algorithm/topo_match/physical_level.h+0-0
Rsrc/ops/op_common/topo/physical_level_build.cc→src/ops/op_common/algorithm/topo_match/physical_level_build.cc+0-0
Rsrc/ops/op_common/topo/physical_level_normalize.cc→src/ops/op_common/algorithm/topo_match/physical_level_normalize.cc+0-0
Rsrc/ops/op_common/topo/topo_match_1d.cc→src/ops/op_common/algorithm/topo_match/topo_match_1d.cc+0-0
Rsrc/ops/op_common/topo/topo_match_1d.h→src/ops/op_common/algorithm/topo_match/topo_match_1d.h+0-0
Rsrc/ops/op_common/topo/topo_match_3_level.cc→src/ops/op_common/algorithm/topo_match/topo_match_3_level.cc+0-0
Rsrc/ops/op_common/topo/topo_match_3_level.h→src/ops/op_common/algorithm/topo_match/topo_match_3_level.h+0-0
Rsrc/ops/op_common/topo/topo_match_base.cc→src/ops/op_common/algorithm/topo_match/topo_match_base.cc+0-0
Rsrc/ops/op_common/topo/topo_match_base.h→src/ops/op_common/algorithm/topo_match/topo_match_base.h+0-0
Rsrc/ops/op_common/topo/topo_match_base_v2.cc→src/ops/op_common/algorithm/topo_match/topo_match_base_v2.cc+0-0
Rsrc/ops/op_common/topo/topo_match_base_v2.h→src/ops/op_common/algorithm/topo_match/topo_match_base_v2.h+0-0
Rsrc/ops/op_common/topo/topo_match_concurrent.cc→src/ops/op_common/algorithm/topo_match/topo_match_concurrent.cc+0-0
Rsrc/ops/op_common/topo/topo_match_concurrent.h→src/ops/op_common/algorithm/topo_match/topo_match_concurrent.h+0-0
Rsrc/ops/op_common/topo/topo_match_concurrent_v2.cc→src/ops/op_common/algorithm/topo_match/topo_match_concurrent_v2.cc+0-0
Rsrc/ops/op_common/topo/topo_match_concurrent_v2.h→src/ops/op_common/algorithm/topo_match/topo_match_concurrent_v2.h+0-0
Rsrc/ops/op_common/topo/topo_match_multilevel.cc→src/ops/op_common/algorithm/topo_match/topo_match_multilevel.cc+0-0
Rsrc/ops/op_common/topo/topo_match_multilevel.h→src/ops/op_common/algorithm/topo_match/topo_match_multilevel.h+0-0
Rsrc/ops/op_common/topo/topo_match_one_level.cc→src/ops/op_common/algorithm/topo_match/topo_match_one_level.cc+0-0
Rsrc/ops/op_common/topo/topo_match_one_level.h→src/ops/op_common/algorithm/topo_match/topo_match_one_level.h+0-0
Rsrc/ops/op_common/topo/topo_match_pcie_mix.cc→src/ops/op_common/algorithm/topo_match/topo_match_pcie_mix.cc+0-0
Rsrc/ops/op_common/topo/topo_match_pcie_mix.h→src/ops/op_common/algorithm/topo_match/topo_match_pcie_mix.h+0-0
Rsrc/ops/op_common/topo/topo_match_squeeze_2d.cc→src/ops/op_common/algorithm/topo_match/topo_match_squeeze_2d.cc+0-0
Rsrc/ops/op_common/topo/topo_match_squeeze_2d.h→src/ops/op_common/algorithm/topo_match/topo_match_squeeze_2d.h+0-0
Rsrc/ops/op_common/topo/topo_match_three_level.cc→src/ops/op_common/algorithm/topo_match/topo_match_three_level.cc+0-0
Rsrc/ops/op_common/topo/topo_match_three_level.h→src/ops/op_common/algorithm/topo_match/topo_match_three_level.h+0-0
Rsrc/ops/op_common/topo/topo_match_two_level.cc→src/ops/op_common/algorithm/topo_match/topo_match_two_level.cc+0-0
Rsrc/ops/op_common/topo/topo_match_two_level.h→src/ops/op_common/algorithm/topo_match/topo_match_two_level.h+0-0
Rsrc/ops/op_common/topo/topo_match_ubx.cc→src/ops/op_common/algorithm/topo_match/topo_match_ubx.cc+0-0
Rsrc/ops/op_common/topo/topo_match_ubx.h→src/ops/op_common/algorithm/topo_match/topo_match_ubx.h+0-0
Rsrc/ops/op_common/topo/topo_match_ubx_1d.cc→src/ops/op_common/algorithm/topo_match/topo_match_ubx_1d.cc+0-0
Rsrc/ops/op_common/topo/topo_match_ubx_1d.h→src/ops/op_common/algorithm/topo_match/topo_match_ubx_1d.h+0-0
Rsrc/ops/op_common/topo/topo.cc→src/ops/op_common/topo_info/topo.cc+0-0
Rsrc/ops/op_common/topo/topo.h→src/ops/op_common/topo_info/topo.h+0-0
Rsrc/ops/op_common/topo/topo_host.cc→src/ops/op_common/topo_info/topo_host.cc+0-0
Rsrc/ops/op_common/topo/topo_host.h→src/ops/op_common/topo_info/topo_host.h+0-0
Msrc/ops/recv/CMakeLists.txt+2-3
Rsrc/ops/recv/executor/CMakeLists.txt→src/ops/recv/algorithm/executor/CMakeLists.txt+0-0
Rsrc/ops/recv/executor/ins_recv_dpu_executor.cc→src/ops/recv/algorithm/executor/ins_recv_dpu_executor.cc+1-1
Rsrc/ops/recv/executor/ins_recv_dpu_executor.h→src/ops/recv/algorithm/executor/ins_recv_dpu_executor.h+0-0
Rsrc/ops/recv/executor/ins_recv_executor.cc→src/ops/recv/algorithm/executor/ins_recv_executor.cc+0-0
Rsrc/ops/recv/executor/ins_recv_executor.h→src/ops/recv/algorithm/executor/ins_recv_executor.h+0-0
Rsrc/ops/recv/executor/ins_v2_recv_executor.cc→src/ops/recv/algorithm/executor/ins_v2_recv_executor.cc+0-0
Rsrc/ops/recv/executor/ins_v2_recv_executor.h→src/ops/recv/algorithm/executor/ins_v2_recv_executor.h+0-0
Rsrc/ops/recv/executor/ins_v2_recv_sole_executor.cc→src/ops/recv/algorithm/executor/ins_v2_recv_sole_executor.cc+0-0
Rsrc/ops/recv/executor/ins_v2_recv_sole_executor.h→src/ops/recv/algorithm/executor/ins_v2_recv_sole_executor.h+0-0
Rsrc/ops/recv/template/CMakeLists.txt→src/ops/recv/algorithm/template/CMakeLists.txt+0-0
Rsrc/ops/recv/template/aiv/CMakeLists.txt→src/ops/recv/algorithm/template/aiv/CMakeLists.txt+0-0
Rsrc/ops/recv/template/aiv/kernel/CMakeLists.txt→src/ops/recv/algorithm/template/aiv/kernel/CMakeLists.txt+4-4
Rsrc/ops/recv/template/aiv/kernel/aiv_recv_mesh_1D.h→src/ops/recv/algorithm/template/aiv/kernel/aiv_recv_mesh_1D.h+0-0
Rsrc/ops/recv/template/aiv/kernel/aiv_recv_op.h→src/ops/recv/algorithm/template/aiv/kernel/aiv_recv_op.h+0-0
Rsrc/ops/recv/template/host_nic/CMakeLists.txt→src/ops/recv/algorithm/template/host_nic/CMakeLists.txt+0-0
Rsrc/ops/recv/template/host_nic/ins_temp_recv_host_nic_dpu.cc→src/ops/recv/algorithm/template/host_nic/ins_temp_recv_host_nic_dpu.cc+0-0
Rsrc/ops/recv/template/host_nic/ins_temp_recv_host_nic_dpu.h→src/ops/recv/algorithm/template/host_nic/ins_temp_recv_host_nic_dpu.h+0-0
Rsrc/ops/recv/template/ins_temp_recv_dpu.cc→src/ops/recv/algorithm/template/ins_temp_recv_dpu.cc+0-0
Rsrc/ops/recv/template/ins_temp_recv_dpu.h→src/ops/recv/algorithm/template/ins_temp_recv_dpu.h+0-0
Rsrc/ops/recv/recv_op.cc→src/ops/recv/recv.cc+1-1
Rsrc/ops/recv/recv_op.h→src/ops/recv/recv.h+0-0
Msrc/ops/reduce/CMakeLists.txt+2-3
Rsrc/ops/reduce/executor/CMakeLists.txt→src/ops/reduce/algorithm/executor/CMakeLists.txt+0-0
Rsrc/ops/reduce/executor/ins_v2_reduce_omnipipe_3d_executor.cc→src/ops/reduce/algorithm/executor/ins_v2_reduce_omnipipe_3d_executor.cc+0-0
Rsrc/ops/reduce/executor/ins_v2_reduce_omnipipe_3d_executor.h→src/ops/reduce/algorithm/executor/ins_v2_reduce_omnipipe_3d_executor.h+0-0
Rsrc/ops/reduce/executor/ins_v2_reduce_omnipipe_executor.cc→src/ops/reduce/algorithm/executor/ins_v2_reduce_omnipipe_executor.cc+0-0
Rsrc/ops/reduce/executor/ins_v2_reduce_omnipipe_executor.h→src/ops/reduce/algorithm/executor/ins_v2_reduce_omnipipe_executor.h+0-0
Rsrc/ops/reduce/executor/ins_v2_reduce_sequence_executor.cc→src/ops/reduce/algorithm/executor/ins_v2_reduce_sequence_executor.cc+0-0
Rsrc/ops/reduce/executor/ins_v2_reduce_sequence_executor.h→src/ops/reduce/algorithm/executor/ins_v2_reduce_sequence_executor.h+0-0
Rsrc/ops/reduce/executor/reduce_parallel_executor.cc→src/ops/reduce/algorithm/executor/reduce_parallel_executor.cc+0-0
Rsrc/ops/reduce/executor/reduce_parallel_executor.h→src/ops/reduce/algorithm/executor/reduce_parallel_executor.h+0-0
Rsrc/ops/reduce/executor/reduce_sequence_executor_aicpu_3level.cc→src/ops/reduce/algorithm/executor/reduce_sequence_executor_aicpu_3level.cc+0-0
Rsrc/ops/reduce/executor/reduce_sequence_executor_aicpu_3level.h→src/ops/reduce/algorithm/executor/reduce_sequence_executor_aicpu_3level.h+0-0
Rsrc/ops/reduce/executor/reduce_sole_executor.cc→src/ops/reduce/algorithm/executor/reduce_sole_executor.cc+0-0
Rsrc/ops/reduce/executor/reduce_sole_executor.h→src/ops/reduce/algorithm/executor/reduce_sole_executor.h+0-0
Rsrc/ops/reduce/template/CMakeLists.txt→src/ops/reduce/algorithm/template/CMakeLists.txt+0-0
Rsrc/ops/reduce/template/aicpu/CMakeLists.txt→src/ops/reduce/algorithm/template/aicpu/CMakeLists.txt+0-0
Rsrc/ops/reduce/template/aicpu/ins_temp_gather_dpu_inter.cc→src/ops/reduce/algorithm/template/aicpu/ins_temp_gather_dpu_inter.cc+0-0
Rsrc/ops/reduce/template/aicpu/ins_temp_gather_dpu_inter.h→src/ops/reduce/algorithm/template/aicpu/ins_temp_gather_dpu_inter.h+0-0
Rsrc/ops/reduce/template/aicpu/ins_temp_gather_mesh_1D_intra.cc→src/ops/reduce/algorithm/template/aicpu/ins_temp_gather_mesh_1D_intra.cc+0-0
Rsrc/ops/reduce/template/aicpu/ins_temp_gather_mesh_1D_intra.h→src/ops/reduce/algorithm/template/aicpu/ins_temp_gather_mesh_1D_intra.h+0-0
Rsrc/ops/reduce/template/aicpu/reduce_aicpu_reduce_nhr.cc→src/ops/reduce/algorithm/template/aicpu/reduce_aicpu_reduce_nhr.cc+0-0
Rsrc/ops/reduce/template/aicpu/reduce_aicpu_reduce_nhr.h→src/ops/reduce/algorithm/template/aicpu/reduce_aicpu_reduce_nhr.h+0-0
Rsrc/ops/reduce/template/aicpu/reduce_mesh_1D.cc→src/ops/reduce/algorithm/template/aicpu/reduce_mesh_1D.cc+0-0
Rsrc/ops/reduce/template/aicpu/reduce_mesh_1D.h→src/ops/reduce/algorithm/template/aicpu/reduce_mesh_1D.h+0-0
Rsrc/ops/reduce/template/aicpu/reduce_mesh_1D_two_shot.cc→src/ops/reduce/algorithm/template/aicpu/reduce_mesh_1D_two_shot.cc+0-0
Rsrc/ops/reduce/template/aicpu/reduce_mesh_1D_two_shot.h→src/ops/reduce/algorithm/template/aicpu/reduce_mesh_1D_two_shot.h+0-0
Rsrc/ops/reduce/template/aicpu/reduce_nhr.cc→src/ops/reduce/algorithm/template/aicpu/reduce_nhr.cc+0-0
Rsrc/ops/reduce/template/aicpu/reduce_nhr.h→src/ops/reduce/algorithm/template/aicpu/reduce_nhr.h+0-0
Rsrc/ops/reduce/template/aiv/CMakeLists.txt→src/ops/reduce/algorithm/template/aiv/CMakeLists.txt+0-0
Rsrc/ops/reduce/template/aiv/aiv_temp_reduce_mesh_1D.cc→src/ops/reduce/algorithm/template/aiv/aiv_temp_reduce_mesh_1D.cc+0-0
Rsrc/ops/reduce/template/aiv/aiv_temp_reduce_mesh_1D.h→src/ops/reduce/algorithm/template/aiv/aiv_temp_reduce_mesh_1D.h+0-0
Rsrc/ops/reduce/template/aiv/kernel/CMakeLists.txt→src/ops/reduce/algorithm/template/aiv/kernel/CMakeLists.txt+4-4
Rsrc/ops/reduce/template/aiv/kernel/aiv_reduce_mesh_1d.h→src/ops/reduce/algorithm/template/aiv/kernel/aiv_reduce_mesh_1d.h+0-0
Rsrc/ops/reduce/template/aiv/kernel/aiv_reduce_op.h→src/ops/reduce/algorithm/template/aiv/kernel/aiv_reduce_op.h+0-0
Rsrc/ops/reduce/template/ccu/CMakeLists.txt→src/ops/reduce/algorithm/template/ccu/CMakeLists.txt+0-0
Rsrc/ops/reduce/template/ccu/ccu_temp_gather_omnipipe_mesh_1d_mem2mem.cc→src/ops/reduce/algorithm/template/ccu/ccu_temp_gather_omnipipe_mesh_1d_mem2mem.cc+0-0
Rsrc/ops/reduce/template/ccu/ccu_temp_gather_omnipipe_mesh_1d_mem2mem.h→src/ops/reduce/algorithm/template/ccu/ccu_temp_gather_omnipipe_mesh_1d_mem2mem.h+0-0
Rsrc/ops/reduce/template/ccu/ccu_temp_gather_omnipipe_nhr1d_mem2mem.cc→src/ops/reduce/algorithm/template/ccu/ccu_temp_gather_omnipipe_nhr1d_mem2mem.cc+0-0
Rsrc/ops/reduce/template/ccu/ccu_temp_gather_omnipipe_nhr1d_mem2mem.h→src/ops/reduce/algorithm/template/ccu/ccu_temp_gather_omnipipe_nhr1d_mem2mem.h+0-0
Rsrc/ops/reduce/template/ccu/ccu_temp_reduce_mesh_1D.cc→src/ops/reduce/algorithm/template/ccu/ccu_temp_reduce_mesh_1D.cc+0-0
Rsrc/ops/reduce/template/ccu/ccu_temp_reduce_mesh_1D.h→src/ops/reduce/algorithm/template/ccu/ccu_temp_reduce_mesh_1D.h+0-0
Rsrc/ops/reduce/template/ccu/ccu_temp_reduce_mesh_1D_mem2mem.cc→src/ops/reduce/algorithm/template/ccu/ccu_temp_reduce_mesh_1D_mem2mem.cc+0-0
Rsrc/ops/reduce/template/ccu/ccu_temp_reduce_mesh_1D_mem2mem.h→src/ops/reduce/algorithm/template/ccu/ccu_temp_reduce_mesh_1D_mem2mem.h+0-0
Rsrc/ops/reduce/template/ccu/ccu_temp_reduce_mesh_1D_twoshot_mem2mem.cc→src/ops/reduce/algorithm/template/ccu/ccu_temp_reduce_mesh_1D_twoshot_mem2mem.cc+0-0
Rsrc/ops/reduce/template/ccu/ccu_temp_reduce_mesh_1D_twoshot_mem2mem.h→src/ops/reduce/algorithm/template/ccu/ccu_temp_reduce_mesh_1D_twoshot_mem2mem.h+0-0
Rsrc/ops/reduce/template/ccu/ccu_temp_reduce_nhr_1D_mem2mem.cc→src/ops/reduce/algorithm/template/ccu/ccu_temp_reduce_nhr_1D_mem2mem.cc+0-0
Rsrc/ops/reduce/template/ccu/ccu_temp_reduce_nhr_1D_mem2mem.h→src/ops/reduce/algorithm/template/ccu/ccu_temp_reduce_nhr_1D_mem2mem.h+0-0
Rsrc/ops/reduce/template/ccu/kernel/CMakeLists.txt→src/ops/reduce/algorithm/template/ccu/kernel/CMakeLists.txt+0-0
Rsrc/ops/reduce/template/ccu/kernel/ccu_kernel_gather_omnipipe_mesh_1d_mem2mem.cc→src/ops/reduce/algorithm/template/ccu/kernel/ccu_kernel_gather_omnipipe_mesh_1d_mem2mem.cc+0-0
Rsrc/ops/reduce/template/ccu/kernel/ccu_kernel_gather_omnipipe_mesh_1d_mem2mem.h→src/ops/reduce/algorithm/template/ccu/kernel/ccu_kernel_gather_omnipipe_mesh_1d_mem2mem.h+0-0
Rsrc/ops/reduce/template/ccu/kernel/ccu_kernel_gather_omnipipe_nhr1d_mem2mem.cc→src/ops/reduce/algorithm/template/ccu/kernel/ccu_kernel_gather_omnipipe_nhr1d_mem2mem.cc+0-0
Rsrc/ops/reduce/template/ccu/kernel/ccu_kernel_gather_omnipipe_nhr1d_mem2mem.h→src/ops/reduce/algorithm/template/ccu/kernel/ccu_kernel_gather_omnipipe_nhr1d_mem2mem.h+0-0
Rsrc/ops/reduce/template/ccu/kernel/ccu_kernel_reduce_mesh1d.cc→src/ops/reduce/algorithm/template/ccu/kernel/ccu_kernel_reduce_mesh1d.cc+0-0
Rsrc/ops/reduce/template/ccu/kernel/ccu_kernel_reduce_mesh1d.h→src/ops/reduce/algorithm/template/ccu/kernel/ccu_kernel_reduce_mesh1d.h+0-0
Rsrc/ops/reduce/template/ccu/kernel/ccu_kernel_reduce_mesh1d_mem2mem.cc→src/ops/reduce/algorithm/template/ccu/kernel/ccu_kernel_reduce_mesh1d_mem2mem.cc+0-0
Rsrc/ops/reduce/template/ccu/kernel/ccu_kernel_reduce_mesh1d_mem2mem.h→src/ops/reduce/algorithm/template/ccu/kernel/ccu_kernel_reduce_mesh1d_mem2mem.h+0-0
Rsrc/ops/reduce/template/ccu/kernel/ccu_kernel_reduce_mesh1d_twoshot_mem2mem.cc→src/ops/reduce/algorithm/template/ccu/kernel/ccu_kernel_reduce_mesh1d_twoshot_mem2mem.cc+0-0
Rsrc/ops/reduce/template/ccu/kernel/ccu_kernel_reduce_mesh1d_twoshot_mem2mem.h→src/ops/reduce/algorithm/template/ccu/kernel/ccu_kernel_reduce_mesh1d_twoshot_mem2mem.h+0-0
Rsrc/ops/reduce/template/ccu/kernel/ccu_kernel_reduce_nhr1d_mem2mem.cc→src/ops/reduce/algorithm/template/ccu/kernel/ccu_kernel_reduce_nhr1d_mem2mem.cc+0-0
Rsrc/ops/reduce/template/ccu/kernel/ccu_kernel_reduce_nhr1d_mem2mem.h→src/ops/reduce/algorithm/template/ccu/kernel/ccu_kernel_reduce_nhr1d_mem2mem.h+0-0
Rsrc/ops/reduce/reduce_op.cc→src/ops/reduce/reduce.cc+1-1
Rsrc/ops/reduce/reduce_op.h→src/ops/reduce/reduce.h+0-0
Rsrc/ops/reduce_scatter/executor/CMakeLists.txt→src/ops/reduce_scatter/algorithm/executor/CMakeLists.txt+0-0
Rsrc/ops/reduce_scatter/executor/ins_reduce_scatter_concurrent_executor.cc→src/ops/reduce_scatter/algorithm/executor/ins_reduce_scatter_concurrent_executor.cc+0-0
Rsrc/ops/reduce_scatter/executor/ins_reduce_scatter_concurrent_executor.h→src/ops/reduce_scatter/algorithm/executor/ins_reduce_scatter_concurrent_executor.h+0-0
Rsrc/ops/reduce_scatter/executor/ins_reduce_scatter_parallel_executor.cc→src/ops/reduce_scatter/algorithm/executor/ins_reduce_scatter_parallel_executor.cc+0-0
Rsrc/ops/reduce_scatter/executor/ins_reduce_scatter_parallel_executor.h→src/ops/reduce_scatter/algorithm/executor/ins_reduce_scatter_parallel_executor.h+0-0
Rsrc/ops/reduce_scatter/executor/ins_v2_reduce_scatter_omnipipe_2d_executor.cc→src/ops/reduce_scatter/algorithm/executor/ins_v2_reduce_scatter_omnipipe_2d_executor.cc+0-0
Rsrc/ops/reduce_scatter/executor/ins_v2_reduce_scatter_omnipipe_2d_executor.h→src/ops/reduce_scatter/algorithm/executor/ins_v2_reduce_scatter_omnipipe_2d_executor.h+0-0
Rsrc/ops/reduce_scatter/executor/ins_v2_reduce_scatter_omnipipe_executor.cc→src/ops/reduce_scatter/algorithm/executor/ins_v2_reduce_scatter_omnipipe_executor.cc+0-0
Rsrc/ops/reduce_scatter/executor/ins_v2_reduce_scatter_omnipipe_executor.h→src/ops/reduce_scatter/algorithm/executor/ins_v2_reduce_scatter_omnipipe_executor.h+0-0
Rsrc/ops/reduce_scatter/executor/ins_v2_reduce_scatter_order_preserved_executor.cc→src/ops/reduce_scatter/algorithm/executor/ins_v2_reduce_scatter_order_preserved_executor.cc+0-0
Rsrc/ops/reduce_scatter/executor/ins_v2_reduce_scatter_order_preserved_executor.h→src/ops/reduce_scatter/algorithm/executor/ins_v2_reduce_scatter_order_preserved_executor.h+0-0
Rsrc/ops/reduce_scatter/executor/ins_v2_reduce_scatter_sequence_executor.cc→src/ops/reduce_scatter/algorithm/executor/ins_v2_reduce_scatter_sequence_executor.cc+0-0
Rsrc/ops/reduce_scatter/executor/ins_v2_reduce_scatter_sequence_executor.h→src/ops/reduce_scatter/algorithm/executor/ins_v2_reduce_scatter_sequence_executor.h+0-0
Rsrc/ops/reduce_scatter/executor/ins_v2_reduce_scatter_sequence_executor_3level.cc→src/ops/reduce_scatter/algorithm/executor/ins_v2_reduce_scatter_sequence_executor_3level.cc+0-0
Rsrc/ops/reduce_scatter/executor/ins_v2_reduce_scatter_sequence_executor_3level.h→src/ops/reduce_scatter/algorithm/executor/ins_v2_reduce_scatter_sequence_executor_3level.h+0-0
Rsrc/ops/reduce_scatter/executor/ins_v2_reduce_scatter_sequence_executor_aicpu.cc→src/ops/reduce_scatter/algorithm/executor/ins_v2_reduce_scatter_sequence_executor_aicpu.cc+0-0
Rsrc/ops/reduce_scatter/executor/ins_v2_reduce_scatter_sequence_executor_aicpu.h→src/ops/reduce_scatter/algorithm/executor/ins_v2_reduce_scatter_sequence_executor_aicpu.h+0-0
Rsrc/ops/reduce_scatter/executor/ins_v2_reduce_scatter_sole_executor.cc→src/ops/reduce_scatter/algorithm/executor/ins_v2_reduce_scatter_sole_executor.cc+0-0
Rsrc/ops/reduce_scatter/executor/ins_v2_reduce_scatter_sole_executor.h→src/ops/reduce_scatter/algorithm/executor/ins_v2_reduce_scatter_sole_executor.h+0-0
Rsrc/ops/reduce_scatter/template/CMakeLists.txt→src/ops/reduce_scatter/algorithm/template/CMakeLists.txt+0-0
Rsrc/ops/reduce_scatter/template/aicpu/CMakeLists.txt→src/ops/reduce_scatter/algorithm/template/aicpu/CMakeLists.txt+0-0
Rsrc/ops/reduce_scatter/template/aicpu/ins_temp_reduce_scatter_aicpu_reduce_nhr.cc→src/ops/reduce_scatter/algorithm/template/aicpu/ins_temp_reduce_scatter_aicpu_reduce_nhr.cc+0-0
Rsrc/ops/reduce_scatter/template/aicpu/ins_temp_reduce_scatter_aicpu_reduce_nhr.h→src/ops/reduce_scatter/algorithm/template/aicpu/ins_temp_reduce_scatter_aicpu_reduce_nhr.h+0-0
Rsrc/ops/reduce_scatter/template/aicpu/ins_temp_reduce_scatter_aicpu_reduce_nhr_pcie.cc→src/ops/reduce_scatter/algorithm/template/aicpu/ins_temp_reduce_scatter_aicpu_reduce_nhr_pcie.cc+0-0
Rsrc/ops/reduce_scatter/template/aicpu/ins_temp_reduce_scatter_aicpu_reduce_nhr_pcie.h→src/ops/reduce_scatter/algorithm/template/aicpu/ins_temp_reduce_scatter_aicpu_reduce_nhr_pcie.h+0-0
Rsrc/ops/reduce_scatter/template/aicpu/ins_temp_reduce_scatter_mesh_1D.cc→src/ops/reduce_scatter/algorithm/template/aicpu/ins_temp_reduce_scatter_mesh_1D.cc+1-1
Rsrc/ops/reduce_scatter/template/aicpu/ins_temp_reduce_scatter_mesh_1D.h→src/ops/reduce_scatter/algorithm/template/aicpu/ins_temp_reduce_scatter_mesh_1D.h+0-0
Rsrc/ops/reduce_scatter/template/aicpu/ins_temp_reduce_scatter_mesh_1D_Z_axis_detour.cc→src/ops/reduce_scatter/algorithm/template/aicpu/ins_temp_reduce_scatter_mesh_1D_Z_axis_detour.cc+0-0
Rsrc/ops/reduce_scatter/template/aicpu/ins_temp_reduce_scatter_mesh_1D_Z_axis_detour.h→src/ops/reduce_scatter/algorithm/template/aicpu/ins_temp_reduce_scatter_mesh_1D_Z_axis_detour.h+0-0
Rsrc/ops/reduce_scatter/template/aicpu/ins_temp_reduce_scatter_mesh_1D_meshchunk.cc→src/ops/reduce_scatter/algorithm/template/aicpu/ins_temp_reduce_scatter_mesh_1D_meshchunk.cc+0-0
Rsrc/ops/reduce_scatter/template/aicpu/ins_temp_reduce_scatter_mesh_1D_meshchunk.h→src/ops/reduce_scatter/algorithm/template/aicpu/ins_temp_reduce_scatter_mesh_1D_meshchunk.h+0-0
Rsrc/ops/reduce_scatter/template/aicpu/ins_temp_reduce_scatter_mesh_1d_dpu.cc→src/ops/reduce_scatter/algorithm/template/aicpu/ins_temp_reduce_scatter_mesh_1d_dpu.cc+0-0
Rsrc/ops/reduce_scatter/template/aicpu/ins_temp_reduce_scatter_mesh_1d_dpu.h→src/ops/reduce_scatter/algorithm/template/aicpu/ins_temp_reduce_scatter_mesh_1d_dpu.h+0-0
Rsrc/ops/reduce_scatter/template/aicpu/ins_temp_reduce_scatter_nhr.cc→src/ops/reduce_scatter/algorithm/template/aicpu/ins_temp_reduce_scatter_nhr.cc+0-0
Rsrc/ops/reduce_scatter/template/aicpu/ins_temp_reduce_scatter_nhr.h→src/ops/reduce_scatter/algorithm/template/aicpu/ins_temp_reduce_scatter_nhr.h+0-0
Rsrc/ops/reduce_scatter/template/aicpu/ins_temp_reduce_scatter_omnipipe_mesh_1D.cc→src/ops/reduce_scatter/algorithm/template/aicpu/ins_temp_reduce_scatter_omnipipe_mesh_1D.cc+0-0
Rsrc/ops/reduce_scatter/template/aicpu/ins_temp_reduce_scatter_omnipipe_mesh_1D.h→src/ops/reduce_scatter/algorithm/template/aicpu/ins_temp_reduce_scatter_omnipipe_mesh_1D.h+0-0
Rsrc/ops/reduce_scatter/template/aicpu/ins_temp_reduce_scatter_omnipipe_mesh_1d_dpu.cc→src/ops/reduce_scatter/algorithm/template/aicpu/ins_temp_reduce_scatter_omnipipe_mesh_1d_dpu.cc+0-0
Rsrc/ops/reduce_scatter/template/aicpu/ins_temp_reduce_scatter_omnipipe_mesh_1d_dpu.h→src/ops/reduce_scatter/algorithm/template/aicpu/ins_temp_reduce_scatter_omnipipe_mesh_1d_dpu.h+0-0
Rsrc/ops/reduce_scatter/template/aicpu/ins_temp_reduce_scatter_omnipipe_nhr.cc→src/ops/reduce_scatter/algorithm/template/aicpu/ins_temp_reduce_scatter_omnipipe_nhr.cc+0-0
Rsrc/ops/reduce_scatter/template/aicpu/ins_temp_reduce_scatter_omnipipe_nhr.h→src/ops/reduce_scatter/algorithm/template/aicpu/ins_temp_reduce_scatter_omnipipe_nhr.h+0-0
Rsrc/ops/reduce_scatter/template/aicpu/ins_temp_reduce_scatter_order_preserved_group.cc→src/ops/reduce_scatter/algorithm/template/aicpu/ins_temp_reduce_scatter_order_preserved_group.cc+0-0
Rsrc/ops/reduce_scatter/template/aicpu/ins_temp_reduce_scatter_order_preserved_group.h→src/ops/reduce_scatter/algorithm/template/aicpu/ins_temp_reduce_scatter_order_preserved_group.h+0-0
Rsrc/ops/reduce_scatter/template/aicpu/ins_temp_reduce_scatter_order_preserved_level1.cc→src/ops/reduce_scatter/algorithm/template/aicpu/ins_temp_reduce_scatter_order_preserved_level1.cc+0-0
Rsrc/ops/reduce_scatter/template/aicpu/ins_temp_reduce_scatter_order_preserved_level1.h→src/ops/reduce_scatter/algorithm/template/aicpu/ins_temp_reduce_scatter_order_preserved_level1.h+0-0
Rsrc/ops/reduce_scatter/template/aiv/CMakeLists.txt→src/ops/reduce_scatter/algorithm/template/aiv/CMakeLists.txt+0-0
Rsrc/ops/reduce_scatter/template/aiv/aiv_temp_reduce_scatter_mesh_1D.cc→src/ops/reduce_scatter/algorithm/template/aiv/aiv_temp_reduce_scatter_mesh_1D.cc+0-0
Rsrc/ops/reduce_scatter/template/aiv/aiv_temp_reduce_scatter_mesh_1D.h→src/ops/reduce_scatter/algorithm/template/aiv/aiv_temp_reduce_scatter_mesh_1D.h+0-0
Rsrc/ops/reduce_scatter/template/aiv/kernel/CMakeLists.txt→src/ops/reduce_scatter/algorithm/template/aiv/kernel/CMakeLists.txt+28-28
Rsrc/ops/reduce_scatter/template/aiv/kernel/aiv_reduce_scatter_local_tree.h→src/ops/reduce_scatter/algorithm/template/aiv/kernel/aiv_reduce_scatter_local_tree.h+0-0
Rsrc/ops/reduce_scatter/template/aiv/kernel/aiv_reduce_scatter_local_tree_corectrl.h→src/ops/reduce_scatter/algorithm/template/aiv/kernel/aiv_reduce_scatter_local_tree_corectrl.h+0-0
Rsrc/ops/reduce_scatter/template/aiv/kernel/aiv_reduce_scatter_mesh_1d.h→src/ops/reduce_scatter/algorithm/template/aiv/kernel/aiv_reduce_scatter_mesh_1d.h+0-0
Rsrc/ops/reduce_scatter/template/aiv/kernel/aiv_reduce_scatter_mesh_1d_bigdata.h→src/ops/reduce_scatter/algorithm/template/aiv/kernel/aiv_reduce_scatter_mesh_1d_bigdata.h+0-0
Rsrc/ops/reduce_scatter/template/aiv/kernel/aiv_reduce_scatter_mesh_1d_corectrl.h→src/ops/reduce_scatter/algorithm/template/aiv/kernel/aiv_reduce_scatter_mesh_1d_corectrl.h+0-0
Rsrc/ops/reduce_scatter/template/aiv/kernel/aiv_reduce_scatter_op.h→src/ops/reduce_scatter/algorithm/template/aiv/kernel/aiv_reduce_scatter_op.h+0-0
Rsrc/ops/reduce_scatter/template/aiv/kernel/aiv_rs_mesh_1d_superkernel.h→src/ops/reduce_scatter/algorithm/template/aiv/kernel/aiv_rs_mesh_1d_superkernel.h+0-0
Rsrc/ops/reduce_scatter/template/ccu/CMakeLists.txt→src/ops/reduce_scatter/algorithm/template/ccu/CMakeLists.txt+0-0
Rsrc/ops/reduce_scatter/template/ccu/ccu_temp_reduce_scatter_concurrent_mesh_nhr.cc→src/ops/reduce_scatter/algorithm/template/ccu/ccu_temp_reduce_scatter_concurrent_mesh_nhr.cc+0-0
Rsrc/ops/reduce_scatter/template/ccu/ccu_temp_reduce_scatter_concurrent_mesh_nhr.h→src/ops/reduce_scatter/algorithm/template/ccu/ccu_temp_reduce_scatter_concurrent_mesh_nhr.h+0-0
Rsrc/ops/reduce_scatter/template/ccu/ccu_temp_reduce_scatter_mesh2die.cc→src/ops/reduce_scatter/algorithm/template/ccu/ccu_temp_reduce_scatter_mesh2die.cc+0-0
Rsrc/ops/reduce_scatter/template/ccu/ccu_temp_reduce_scatter_mesh2die.h→src/ops/reduce_scatter/algorithm/template/ccu/ccu_temp_reduce_scatter_mesh2die.h+0-0
Rsrc/ops/reduce_scatter/template/ccu/ccu_temp_reduce_scatter_mesh_1D.cc→src/ops/reduce_scatter/algorithm/template/ccu/ccu_temp_reduce_scatter_mesh_1D.cc+0-0
Rsrc/ops/reduce_scatter/template/ccu/ccu_temp_reduce_scatter_mesh_1D.h→src/ops/reduce_scatter/algorithm/template/ccu/ccu_temp_reduce_scatter_mesh_1D.h+0-0
Rsrc/ops/reduce_scatter/template/ccu/ccu_temp_reduce_scatter_mesh_1D_2die_mem2mem.cc→src/ops/reduce_scatter/algorithm/template/ccu/ccu_temp_reduce_scatter_mesh_1D_2die_mem2mem.cc+0-0
Rsrc/ops/reduce_scatter/template/ccu/ccu_temp_reduce_scatter_mesh_1D_2die_mem2mem.h→src/ops/reduce_scatter/algorithm/template/ccu/ccu_temp_reduce_scatter_mesh_1D_2die_mem2mem.h+0-0
Rsrc/ops/reduce_scatter/template/ccu/ccu_temp_reduce_scatter_mesh_1D_mem2mem.cc→src/ops/reduce_scatter/algorithm/template/ccu/ccu_temp_reduce_scatter_mesh_1D_mem2mem.cc+0-0
Rsrc/ops/reduce_scatter/template/ccu/ccu_temp_reduce_scatter_mesh_1D_mem2mem.h→src/ops/reduce_scatter/algorithm/template/ccu/ccu_temp_reduce_scatter_mesh_1D_mem2mem.h+0-0
Rsrc/ops/reduce_scatter/template/ccu/ccu_temp_reduce_scatter_nhr_1D_mem2mem.cc→src/ops/reduce_scatter/algorithm/template/ccu/ccu_temp_reduce_scatter_nhr_1D_mem2mem.cc+0-0
Rsrc/ops/reduce_scatter/template/ccu/ccu_temp_reduce_scatter_nhr_1D_mem2mem.h→src/ops/reduce_scatter/algorithm/template/ccu/ccu_temp_reduce_scatter_nhr_1D_mem2mem.h+0-0
Rsrc/ops/reduce_scatter/template/ccu/ccu_temp_reduce_scatter_nhr_1D_multi_jetty_mem2mem.cc→src/ops/reduce_scatter/algorithm/template/ccu/ccu_temp_reduce_scatter_nhr_1D_multi_jetty_mem2mem.cc+0-0
Rsrc/ops/reduce_scatter/template/ccu/ccu_temp_reduce_scatter_nhr_1D_multi_jetty_mem2mem.h→src/ops/reduce_scatter/algorithm/template/ccu/ccu_temp_reduce_scatter_nhr_1D_multi_jetty_mem2mem.h+0-0
Rsrc/ops/reduce_scatter/template/ccu/ccu_temp_reduce_scatter_omnipipe_mesh1d.cc→src/ops/reduce_scatter/algorithm/template/ccu/ccu_temp_reduce_scatter_omnipipe_mesh1d.cc+0-0
Rsrc/ops/reduce_scatter/template/ccu/ccu_temp_reduce_scatter_omnipipe_mesh1d.h→src/ops/reduce_scatter/algorithm/template/ccu/ccu_temp_reduce_scatter_omnipipe_mesh1d.h+0-0
Rsrc/ops/reduce_scatter/template/ccu/ccu_temp_reduce_scatter_omnipipe_mesh1d_mem2mem.cc→src/ops/reduce_scatter/algorithm/template/ccu/ccu_temp_reduce_scatter_omnipipe_mesh1d_mem2mem.cc+0-0
Rsrc/ops/reduce_scatter/template/ccu/ccu_temp_reduce_scatter_omnipipe_mesh1d_mem2mem.h→src/ops/reduce_scatter/algorithm/template/ccu/ccu_temp_reduce_scatter_omnipipe_mesh1d_mem2mem.h+0-0
Rsrc/ops/reduce_scatter/template/ccu/ccu_temp_reduce_scatter_omnipipe_nhr1d_mem2mem.cc→src/ops/reduce_scatter/algorithm/template/ccu/ccu_temp_reduce_scatter_omnipipe_nhr1d_mem2mem.cc+0-0
Rsrc/ops/reduce_scatter/template/ccu/ccu_temp_reduce_scatter_omnipipe_nhr1d_mem2mem.h→src/ops/reduce_scatter/algorithm/template/ccu/ccu_temp_reduce_scatter_omnipipe_nhr1d_mem2mem.h+0-0
Rsrc/ops/reduce_scatter/template/ccu/kernel/CMakeLists.txt→src/ops/reduce_scatter/algorithm/template/ccu/kernel/CMakeLists.txt+0-0
Rsrc/ops/reduce_scatter/template/ccu/kernel/ccu_kernel_reduce_scatter_mesh1d.cc→src/ops/reduce_scatter/algorithm/template/ccu/kernel/ccu_kernel_reduce_scatter_mesh1d.cc+0-0
Rsrc/ops/reduce_scatter/template/ccu/kernel/ccu_kernel_reduce_scatter_mesh1d.h→src/ops/reduce_scatter/algorithm/template/ccu/kernel/ccu_kernel_reduce_scatter_mesh1d.h+0-0
Rsrc/ops/reduce_scatter/template/ccu/kernel/ccu_kernel_reduce_scatter_mesh1d_2die_mem2mem.cc→src/ops/reduce_scatter/algorithm/template/ccu/kernel/ccu_kernel_reduce_scatter_mesh1d_2die_mem2mem.cc+0-0
Rsrc/ops/reduce_scatter/template/ccu/kernel/ccu_kernel_reduce_scatter_mesh1d_2die_mem2mem.h→src/ops/reduce_scatter/algorithm/template/ccu/kernel/ccu_kernel_reduce_scatter_mesh1d_2die_mem2mem.h+0-0
Rsrc/ops/reduce_scatter/template/ccu/kernel/ccu_kernel_reduce_scatter_mesh1d_mem2mem.cc→src/ops/reduce_scatter/algorithm/template/ccu/kernel/ccu_kernel_reduce_scatter_mesh1d_mem2mem.cc+0-0
Rsrc/ops/reduce_scatter/template/ccu/kernel/ccu_kernel_reduce_scatter_mesh1d_mem2mem.h→src/ops/reduce_scatter/algorithm/template/ccu/kernel/ccu_kernel_reduce_scatter_mesh1d_mem2mem.h+0-0
Rsrc/ops/reduce_scatter/template/ccu/kernel/ccu_kernel_reduce_scatter_mesh2die.cc→src/ops/reduce_scatter/algorithm/template/ccu/kernel/ccu_kernel_reduce_scatter_mesh2die.cc+0-0
Rsrc/ops/reduce_scatter/template/ccu/kernel/ccu_kernel_reduce_scatter_mesh2die.h→src/ops/reduce_scatter/algorithm/template/ccu/kernel/ccu_kernel_reduce_scatter_mesh2die.h+0-0
Rsrc/ops/reduce_scatter/template/ccu/kernel/ccu_kernel_reduce_scatter_nhr1d_mem2mem.cc→src/ops/reduce_scatter/algorithm/template/ccu/kernel/ccu_kernel_reduce_scatter_nhr1d_mem2mem.cc+0-0
Rsrc/ops/reduce_scatter/template/ccu/kernel/ccu_kernel_reduce_scatter_nhr1d_mem2mem.h→src/ops/reduce_scatter/algorithm/template/ccu/kernel/ccu_kernel_reduce_scatter_nhr1d_mem2mem.h+0-0
Rsrc/ops/reduce_scatter/template/ccu/kernel/ccu_kernel_reduce_scatter_nhr1d_multi_jetty_mem2mem.cc→src/ops/reduce_scatter/algorithm/template/ccu/kernel/ccu_kernel_reduce_scatter_nhr1d_multi_jetty_mem2mem.cc+0-0
Rsrc/ops/reduce_scatter/template/ccu/kernel/ccu_kernel_reduce_scatter_nhr1d_multi_jetty_mem2mem.h→src/ops/reduce_scatter/algorithm/template/ccu/kernel/ccu_kernel_reduce_scatter_nhr1d_multi_jetty_mem2mem.h+0-0
Rsrc/ops/reduce_scatter/template/ccu/kernel/ccu_kernel_reduce_scatter_omnipipe_mesh1d.cc→src/ops/reduce_scatter/algorithm/template/ccu/kernel/ccu_kernel_reduce_scatter_omnipipe_mesh1d.cc+0-0
Rsrc/ops/reduce_scatter/template/ccu/kernel/ccu_kernel_reduce_scatter_omnipipe_mesh1d.h→src/ops/reduce_scatter/algorithm/template/ccu/kernel/ccu_kernel_reduce_scatter_omnipipe_mesh1d.h+0-0
Rsrc/ops/reduce_scatter/template/ccu/kernel/ccu_kernel_reduce_scatter_omnipipe_mesh1d_mem2mem.cc→src/ops/reduce_scatter/algorithm/template/ccu/kernel/ccu_kernel_reduce_scatter_omnipipe_mesh1d_mem2mem.cc+0-0
Rsrc/ops/reduce_scatter/template/ccu/kernel/ccu_kernel_reduce_scatter_omnipipe_mesh1d_mem2mem.h→src/ops/reduce_scatter/algorithm/template/ccu/kernel/ccu_kernel_reduce_scatter_omnipipe_mesh1d_mem2mem.h+0-0
Rsrc/ops/reduce_scatter/template/ccu/kernel/ccu_kernel_reduce_scatter_omnipipe_nhr1d_mem2mem.cc→src/ops/reduce_scatter/algorithm/template/ccu/kernel/ccu_kernel_reduce_scatter_omnipipe_nhr1d_mem2mem.cc+0-0
Rsrc/ops/reduce_scatter/template/ccu/kernel/ccu_kernel_reduce_scatter_omnipipe_nhr1d_mem2mem.h→src/ops/reduce_scatter/algorithm/template/ccu/kernel/ccu_kernel_reduce_scatter_omnipipe_nhr1d_mem2mem.h+0-0
Rsrc/ops/reduce_scatter/reduce_scatter_op.cc→src/ops/reduce_scatter/reduce_scatter.cc+1-1
Rsrc/ops/reduce_scatter/reduce_scatter_op.h→src/ops/reduce_scatter/reduce_scatter.h+0-0
Rsrc/ops/reduce_scatter_v/executor/CMakeLists.txt→src/ops/reduce_scatter_v/algorithm/executor/CMakeLists.txt+0-0
Rsrc/ops/reduce_scatter_v/executor/ins_v2_reduce_scatter_v_sole_executor.cc→src/ops/reduce_scatter_v/algorithm/executor/ins_v2_reduce_scatter_v_sole_executor.cc+0-0
Rsrc/ops/reduce_scatter_v/executor/ins_v2_reduce_scatter_v_sole_executor.h→src/ops/reduce_scatter_v/algorithm/executor/ins_v2_reduce_scatter_v_sole_executor.h+0-0
Rsrc/ops/reduce_scatter_v/template/CMakeLists.txt→src/ops/reduce_scatter_v/algorithm/template/CMakeLists.txt+0-0
Rsrc/ops/reduce_scatter_v/template/aicpu/CMakeLists.txt→src/ops/reduce_scatter_v/algorithm/template/aicpu/CMakeLists.txt+0-0
Rsrc/ops/reduce_scatter_v/template/aicpu/ins_temp_reduce_scatter_v_mesh_1D.cc→src/ops/reduce_scatter_v/algorithm/template/aicpu/ins_temp_reduce_scatter_v_mesh_1D.cc+1-1
Rsrc/ops/reduce_scatter_v/template/aicpu/ins_temp_reduce_scatter_v_mesh_1D.h→src/ops/reduce_scatter_v/algorithm/template/aicpu/ins_temp_reduce_scatter_v_mesh_1D.h+0-0
Rsrc/ops/reduce_scatter_v/template/ccu/CMakeLists.txt→src/ops/reduce_scatter_v/algorithm/template/ccu/CMakeLists.txt+0-0
Rsrc/ops/reduce_scatter_v/template/ccu/ccu_temp_reduce_scatter_v_mesh_1D_mem2mem.cc→src/ops/reduce_scatter_v/algorithm/template/ccu/ccu_temp_reduce_scatter_v_mesh_1D_mem2mem.cc+0-0
Rsrc/ops/reduce_scatter_v/template/ccu/ccu_temp_reduce_scatter_v_mesh_1D_mem2mem.h→src/ops/reduce_scatter_v/algorithm/template/ccu/ccu_temp_reduce_scatter_v_mesh_1D_mem2mem.h+0-0
Rsrc/ops/reduce_scatter_v/template/ccu/kernel/CMakeLists.txt→src/ops/reduce_scatter_v/algorithm/template/ccu/kernel/CMakeLists.txt+0-0
Rsrc/ops/reduce_scatter_v/template/ccu/kernel/ccu_kernel_reduce_scatter_v_mesh1d_mem2mem.cc→src/ops/reduce_scatter_v/algorithm/template/ccu/kernel/ccu_kernel_reduce_scatter_v_mesh1d_mem2mem.cc+0-0
Rsrc/ops/reduce_scatter_v/template/ccu/kernel/ccu_kernel_reduce_scatter_v_mesh1d_mem2mem.h→src/ops/reduce_scatter_v/algorithm/template/ccu/kernel/ccu_kernel_reduce_scatter_v_mesh1d_mem2mem.h+0-0
Rsrc/ops/reduce_scatter_v/reduce_scatter_v_op.cc→src/ops/reduce_scatter_v/reduce_scatter_v.cc+1-1
Rsrc/ops/reduce_scatter_v/reduce_scatter_v_op.h→src/ops/reduce_scatter_v/reduce_scatter_v.h+0-0
Msrc/ops/scatter/CMakeLists.txt+2-4
Rsrc/ops/scatter/algo/CMakeLists.txt→src/ops/scatter/algorithm/CMakeLists.txt+2-1
Rsrc/ops/scatter/executor/CMakeLists.txt→src/ops/scatter/algorithm/executor/CMakeLists.txt+0-0
Rsrc/ops/scatter/executor/ins_v2_scatter_concurrent_executor.cc→src/ops/scatter/algorithm/executor/ins_v2_scatter_concurrent_executor.cc+0-0
Rsrc/ops/scatter/executor/ins_v2_scatter_concurrent_executor.h→src/ops/scatter/algorithm/executor/ins_v2_scatter_concurrent_executor.h+0-0
Rsrc/ops/scatter/executor/ins_v2_scatter_omnipipe_2d_executor.cc→src/ops/scatter/algorithm/executor/ins_v2_scatter_omnipipe_2d_executor.cc+0-0
Rsrc/ops/scatter/executor/ins_v2_scatter_omnipipe_2d_executor.h→src/ops/scatter/algorithm/executor/ins_v2_scatter_omnipipe_2d_executor.h+0-0
Rsrc/ops/scatter/executor/ins_v2_scatter_parallel_executor.cc→src/ops/scatter/algorithm/executor/ins_v2_scatter_parallel_executor.cc+0-0
Rsrc/ops/scatter/executor/ins_v2_scatter_parallel_executor.h→src/ops/scatter/algorithm/executor/ins_v2_scatter_parallel_executor.h+0-0
Rsrc/ops/scatter/executor/ins_v2_scatter_sequence_executor.cc→src/ops/scatter/algorithm/executor/ins_v2_scatter_sequence_executor.cc+0-0
Rsrc/ops/scatter/executor/ins_v2_scatter_sequence_executor.h→src/ops/scatter/algorithm/executor/ins_v2_scatter_sequence_executor.h+0-0
Rsrc/ops/scatter/executor/ins_v2_scatter_sole_executor.cc→src/ops/scatter/algorithm/executor/ins_v2_scatter_sole_executor.cc+0-0
Rsrc/ops/scatter/executor/ins_v2_scatter_sole_executor.h→src/ops/scatter/algorithm/executor/ins_v2_scatter_sole_executor.h+0-0
Rsrc/ops/scatter/executor/scatter_sequence_aicpu_executor_3level.cc→src/ops/scatter/algorithm/executor/scatter_sequence_aicpu_executor_3level.cc+0-0
Rsrc/ops/scatter/executor/scatter_sequence_aicpu_executor_3level.h→src/ops/scatter/algorithm/executor/scatter_sequence_aicpu_executor_3level.h+0-0
Rsrc/ops/scatter/algo/scatter_comm_executor.cc→src/ops/scatter/algorithm/scatter_comm_executor.cc+0-0
Rsrc/ops/scatter/algo/scatter_comm_executor.h→src/ops/scatter/algorithm/scatter_comm_executor.h+0-0
Rsrc/ops/scatter/algo/scatter_executor_base.cc→src/ops/scatter/algorithm/scatter_executor_base.cc+0-0
Rsrc/ops/scatter/algo/scatter_executor_base.h→src/ops/scatter/algorithm/scatter_executor_base.h+0-0
Rsrc/ops/scatter/algo/scatter_mesh_executor.cc→src/ops/scatter/algorithm/scatter_mesh_executor.cc+0-0
Rsrc/ops/scatter/algo/scatter_mesh_executor.h→src/ops/scatter/algorithm/scatter_mesh_executor.h+0-0
Rsrc/ops/scatter/algo/scatter_ring_executor.cc→src/ops/scatter/algorithm/scatter_ring_executor.cc+0-0
Rsrc/ops/scatter/algo/scatter_ring_executor.h→src/ops/scatter/algorithm/scatter_ring_executor.h+0-0
Rsrc/ops/scatter/algo/scatter_single_executor.cc→src/ops/scatter/algorithm/scatter_single_executor.cc+0-0
Rsrc/ops/scatter/algo/scatter_single_executor.h→src/ops/scatter/algorithm/scatter_single_executor.h+0-0
Rsrc/ops/scatter/algo/template/CMakeLists.txt→src/ops/scatter/algorithm/template/CMakeLists.txt+6-0
Rsrc/ops/scatter/template/aicpu/CMakeLists.txt→src/ops/scatter/algorithm/template/aicpu/CMakeLists.txt+0-0
Rsrc/ops/scatter/template/aicpu/aicpu_temp_scatter_mesh_1D_Z_axis_detour.cc→src/ops/scatter/algorithm/template/aicpu/aicpu_temp_scatter_mesh_1D_Z_axis_detour.cc+0-0
Rsrc/ops/scatter/template/aicpu/aicpu_temp_scatter_mesh_1D_Z_axis_detour.h→src/ops/scatter/algorithm/template/aicpu/aicpu_temp_scatter_mesh_1D_Z_axis_detour.h+0-0
Rsrc/ops/scatter/template/aicpu/ins_temp_scatter_mesh_1D.cc→src/ops/scatter/algorithm/template/aicpu/ins_temp_scatter_mesh_1D.cc+0-0
Rsrc/ops/scatter/template/aicpu/ins_temp_scatter_mesh_1D.h→src/ops/scatter/algorithm/template/aicpu/ins_temp_scatter_mesh_1D.h+0-0
Rsrc/ops/scatter/template/aicpu/ins_temp_scatter_nhr.cc→src/ops/scatter/algorithm/template/aicpu/ins_temp_scatter_nhr.cc+0-0
Rsrc/ops/scatter/template/aicpu/ins_temp_scatter_nhr.h→src/ops/scatter/algorithm/template/aicpu/ins_temp_scatter_nhr.h+0-0
Rsrc/ops/scatter/template/aicpu/ins_temp_scatter_nhr_dpu_inter_node.cc→src/ops/scatter/algorithm/template/aicpu/ins_temp_scatter_nhr_dpu_inter_node.cc+0-0
Rsrc/ops/scatter/template/aicpu/ins_temp_scatter_nhr_dpu_inter_node.h→src/ops/scatter/algorithm/template/aicpu/ins_temp_scatter_nhr_dpu_inter_node.h+0-0
Rsrc/ops/scatter/template/aiv/CMakeLists.txt→src/ops/scatter/algorithm/template/aiv/CMakeLists.txt+0-0
Rsrc/ops/scatter/template/aiv/aiv_temp_scatter_mesh_1D.cc→src/ops/scatter/algorithm/template/aiv/aiv_temp_scatter_mesh_1D.cc+0-0
Rsrc/ops/scatter/template/aiv/aiv_temp_scatter_mesh_1D.h→src/ops/scatter/algorithm/template/aiv/aiv_temp_scatter_mesh_1D.h+0-0
Rsrc/ops/scatter/template/aiv/kernel/CMakeLists.txt→src/ops/scatter/algorithm/template/aiv/kernel/CMakeLists.txt+4-4
Rsrc/ops/scatter/template/aiv/kernel/aiv_scatter_mesh_1d.h→src/ops/scatter/algorithm/template/aiv/kernel/aiv_scatter_mesh_1d.h+0-0
Rsrc/ops/scatter/template/aiv/kernel/aiv_scatter_op.h→src/ops/scatter/algorithm/template/aiv/kernel/aiv_scatter_op.h+0-0
Rsrc/ops/scatter/template/ccu/CMakeLists.txt→src/ops/scatter/algorithm/template/ccu/CMakeLists.txt+0-0
Rsrc/ops/scatter/template/ccu/ccu_temp_scatter_mesh1d.cc→src/ops/scatter/algorithm/template/ccu/ccu_temp_scatter_mesh1d.cc+0-0
Rsrc/ops/scatter/template/ccu/ccu_temp_scatter_mesh1d.h→src/ops/scatter/algorithm/template/ccu/ccu_temp_scatter_mesh1d.h+0-0
Rsrc/ops/scatter/template/ccu/ccu_temp_scatter_nhr1d_mem2mem.cc→src/ops/scatter/algorithm/template/ccu/ccu_temp_scatter_nhr1d_mem2mem.cc+0-0
Rsrc/ops/scatter/template/ccu/ccu_temp_scatter_nhr1d_mem2mem.h→src/ops/scatter/algorithm/template/ccu/ccu_temp_scatter_nhr1d_mem2mem.h+0-0
Rsrc/ops/scatter/template/ccu/ccu_temp_scatter_omnipipe_mesh1d_mem2mem.cc→src/ops/scatter/algorithm/template/ccu/ccu_temp_scatter_omnipipe_mesh1d_mem2mem.cc+0-0
Rsrc/ops/scatter/template/ccu/ccu_temp_scatter_omnipipe_mesh1d_mem2mem.h→src/ops/scatter/algorithm/template/ccu/ccu_temp_scatter_omnipipe_mesh1d_mem2mem.h+0-0
Rsrc/ops/scatter/template/ccu/ccu_temp_scatter_omnipipe_nhr1d_mem2mem.cc→src/ops/scatter/algorithm/template/ccu/ccu_temp_scatter_omnipipe_nhr1d_mem2mem.cc+0-0
Rsrc/ops/scatter/template/ccu/ccu_temp_scatter_omnipipe_nhr1d_mem2mem.h→src/ops/scatter/algorithm/template/ccu/ccu_temp_scatter_omnipipe_nhr1d_mem2mem.h+0-0
Rsrc/ops/scatter/template/ccu/kernel/CMakeLists.txt→src/ops/scatter/algorithm/template/ccu/kernel/CMakeLists.txt+0-0
Rsrc/ops/scatter/template/ccu/kernel/ccu_kernel_scatter_mesh1d.cc→src/ops/scatter/algorithm/template/ccu/kernel/ccu_kernel_scatter_mesh1d.cc+0-0
Rsrc/ops/scatter/template/ccu/kernel/ccu_kernel_scatter_mesh1d.h→src/ops/scatter/algorithm/template/ccu/kernel/ccu_kernel_scatter_mesh1d.h+0-0
Rsrc/ops/scatter/template/ccu/kernel/ccu_kernel_scatter_nhr1d_mem2mem.cc→src/ops/scatter/algorithm/template/ccu/kernel/ccu_kernel_scatter_nhr1d_mem2mem.cc+0-0
Rsrc/ops/scatter/template/ccu/kernel/ccu_kernel_scatter_nhr1d_mem2mem.h→src/ops/scatter/algorithm/template/ccu/kernel/ccu_kernel_scatter_nhr1d_mem2mem.h+0-0
Rsrc/ops/scatter/template/ccu/kernel/ccu_kernel_scatter_omnipipe_mesh1d_mem2mem.cc→src/ops/scatter/algorithm/template/ccu/kernel/ccu_kernel_scatter_omnipipe_mesh1d_mem2mem.cc+0-0
Rsrc/ops/scatter/template/ccu/kernel/ccu_kernel_scatter_omnipipe_mesh1d_mem2mem.h→src/ops/scatter/algorithm/template/ccu/kernel/ccu_kernel_scatter_omnipipe_mesh1d_mem2mem.h+0-0
Rsrc/ops/scatter/template/ccu/kernel/ccu_kernel_scatter_omnipipe_nhr1d_mem2mem.cc→src/ops/scatter/algorithm/template/ccu/kernel/ccu_kernel_scatter_omnipipe_nhr1d_mem2mem.cc+0-0
Rsrc/ops/scatter/template/ccu/kernel/ccu_kernel_scatter_omnipipe_nhr1d_mem2mem.h→src/ops/scatter/algorithm/template/ccu/kernel/ccu_kernel_scatter_omnipipe_nhr1d_mem2mem.h+0-0
Rsrc/ops/scatter/algo/template/nhr_base.cc→src/ops/scatter/algorithm/template/nhr_base.cc+0-0
Rsrc/ops/scatter/algo/template/nhr_base.h→src/ops/scatter/algorithm/template/nhr_base.h+0-0
Rsrc/ops/scatter/algo/template/scatter_mesh.cc→src/ops/scatter/algorithm/template/scatter_mesh.cc+0-0
Rsrc/ops/scatter/algo/template/scatter_mesh.h→src/ops/scatter/algorithm/template/scatter_mesh.h+0-0
Rsrc/ops/scatter/algo/template/scatter_nb.cc→src/ops/scatter/algorithm/template/scatter_nb.cc+0-0
Rsrc/ops/scatter/algo/template/scatter_nb.h→src/ops/scatter/algorithm/template/scatter_nb.h+0-0
Rsrc/ops/scatter/algo/template/scatter_nhr.cc→src/ops/scatter/algorithm/template/scatter_nhr.cc+0-0
Rsrc/ops/scatter/algo/template/scatter_nhr.h→src/ops/scatter/algorithm/template/scatter_nhr.h+0-0
Rsrc/ops/scatter/algo/template/scatter_ring.cc→src/ops/scatter/algorithm/template/scatter_ring.cc+0-0
Rsrc/ops/scatter/algo/template/scatter_ring.h→src/ops/scatter/algorithm/template/scatter_ring.h+0-0
Rsrc/ops/scatter/algo/template/scatter_ring_direct.cc→src/ops/scatter/algorithm/template/scatter_ring_direct.cc+0-0
Rsrc/ops/scatter/algo/template/scatter_ring_direct.h→src/ops/scatter/algorithm/template/scatter_ring_direct.h+0-0
Rsrc/ops/scatter/scatter_op.cc→src/ops/scatter/scatter.cc+1-1
Rsrc/ops/scatter/scatter_op.h→src/ops/scatter/scatter.h+0-0
Msrc/ops/send/CMakeLists.txt+2-3
Rsrc/ops/send/executor/CMakeLists.txt→src/ops/send/algorithm/executor/CMakeLists.txt+0-0
Rsrc/ops/send/executor/ins_send_dpu_executor.cc→src/ops/send/algorithm/executor/ins_send_dpu_executor.cc+1-1
Rsrc/ops/send/executor/ins_send_dpu_executor.h→src/ops/send/algorithm/executor/ins_send_dpu_executor.h+0-0
Rsrc/ops/send/executor/ins_send_executor.cc→src/ops/send/algorithm/executor/ins_send_executor.cc+0-0
Rsrc/ops/send/executor/ins_send_executor.h→src/ops/send/algorithm/executor/ins_send_executor.h+0-0
Rsrc/ops/send/executor/ins_v2_send_executor.cc→src/ops/send/algorithm/executor/ins_v2_send_executor.cc+0-0
Rsrc/ops/send/executor/ins_v2_send_executor.h→src/ops/send/algorithm/executor/ins_v2_send_executor.h+0-0
Rsrc/ops/send/executor/ins_v2_send_sole_executor.cc→src/ops/send/algorithm/executor/ins_v2_send_sole_executor.cc+0-0
Rsrc/ops/send/executor/ins_v2_send_sole_executor.h→src/ops/send/algorithm/executor/ins_v2_send_sole_executor.h+0-0
Rsrc/ops/send/template/CMakeLists.txt→src/ops/send/algorithm/template/CMakeLists.txt+0-0
Rsrc/ops/send/template/aiv/CMakeLists.txt→src/ops/send/algorithm/template/aiv/CMakeLists.txt+0-0
Rsrc/ops/send/template/aiv/kernel/CMakeLists.txt→src/ops/send/algorithm/template/aiv/kernel/CMakeLists.txt+4-4
Rsrc/ops/send/template/aiv/kernel/aiv_send_mesh_1D.h→src/ops/send/algorithm/template/aiv/kernel/aiv_send_mesh_1D.h+0-0
Rsrc/ops/send/template/aiv/kernel/aiv_send_op.h→src/ops/send/algorithm/template/aiv/kernel/aiv_send_op.h+0-0
Rsrc/ops/send/template/host_nic/CMakeLists.txt→src/ops/send/algorithm/template/host_nic/CMakeLists.txt+0-0
Rsrc/ops/send/template/host_nic/ins_temp_send_host_nic_dpu.cc→src/ops/send/algorithm/template/host_nic/ins_temp_send_host_nic_dpu.cc+0-0
Rsrc/ops/send/template/host_nic/ins_temp_send_host_nic_dpu.h→src/ops/send/algorithm/template/host_nic/ins_temp_send_host_nic_dpu.h+0-0
Rsrc/ops/send/template/ins_temp_send_dpu.cc→src/ops/send/algorithm/template/ins_temp_send_dpu.cc+0-0
Rsrc/ops/send/template/ins_temp_send_dpu.h→src/ops/send/algorithm/template/ins_temp_send_dpu.h+0-0
Rsrc/ops/send/send_op.cc→src/ops/send/send.cc+1-1
Rsrc/ops/send/send_op.h→src/ops/send/send.h+0-0
Msrc/scatter_aicpu_kernel.cmake+157-157