docs: 章节节点新增独立索引文件，消除index.md重复链接并补全索引列表-asc-devkit-AtomGit

已合并

docs: 章节节点新增独立索引文件，消除index.md重复链接并补全索引列表 #6714

ai_xin创建于 22 天前从ai_xin/asc-devkit： master 合入到 cann/asc-devkit： master

docs: 章节节点新增独立索引文件，消除index.md重复链接并补全索引列表 #6714

已合并

ai_xin创建于 22 天前

从ai_xin/asc-devkit master合入到cann/asc-devkit master

收起

docs / zh / guide
- cross_gen_migration_guide
  - 3510_arch_migration
    - 2201_to_3510_guide
      - 2201_to_3510_guide.md
    - 3510_arch_migration.md
  - cross_gen_migration_guide.md
  - instructions_for_new_features
    - instructions_for_new_features.md
- getting_started
  - getting_started.md
  - quick_start
    - quick_start.md
    - simd_programming
      - simd_programming.md
    - simt_programming
      - simt_programming.md
- index.md
- operator_practice
  - best_practices
    - best_practices.md
    - matmul_tuning_cases
      - matmul_tuning_cases.md
  - functional_debug
    - functional_debug.md
  - operator_practice.md
  - performance_analysis
    - performance_analysis.md
  - simd_operator_impl
    - fusion_operator_programming
      - cv_fusion
        cv_fusion.md
      - fusion_operator_programming.md
      - general_fusion
        general_fusion.md
    - matrix_advanced_api
      - feature_scenarios
        feature_scenarios.md
      - matrix_advanced_api.md
    - matrix_basic_api
      - matrix_basic_api.md
    - simd_operator_impl.md
    - vector_programming
      - multi_core_tiling
        multi_core_tiling.md
      - vector_programming.md
  - simd_operator_optimization
    - matrix_compute
      - matrix_compute.md
    - memory_access
      - avoid_ub_bank_conflict
        avoid_ub_bank_conflict.md
      - memory_access.md
    - overhead_optimization
      - overhead_optimization.md
    - pipeline_scheduling
      - pipeline_scheduling.md
    - simd_operator_optimization.md
    - tiling_strategy
      - tiling_strategy.md
    - vector_compute
      - vector_compute.md
      - vf_optimization
        vf_optimization.md
  - simd_simt_hybrid_operator_impl
    - simd_simt_hybrid_operator_impl.md
  - simd_simt_hybrid_optimization
    - compute_optimization
      - compute_optimization.md
    - execution_control
      - execution_control.md
    - instruction_optimization
      - instruction_optimization.md
    - memory_access
      - memory_access.md
    - simd_simt_hybrid_optimization.md
  - simt_operator_impl
    - simt_operator_impl.md
  - simt_operator_optimization
    - control_flow
      - control_flow.md
    - execution_config
      - execution_config.md
    - instruction_optimization
      - instruction_optimization.md
    - memory_access
      - memory_access.md
    - simt_operator_optimization.md
- programming_guide
  - advanced_programming
    - aclnn_operator_development
      - aclnn_operator_development.md
      - compilation_and_deployment
        compilation_and_deployment.md
      - design_and_implementation
        design_and_implementation.md
      - invocation
        invocation.md
    - advanced_ai_core_programming_model
      - advanced_ai_core_programming_model.md
      - simd_simt_hybrid_programming
        simd_simt_hybrid_programming.md
    - advanced_kernel_programming
      - advanced_kernel_programming.md
      - intra_core_sync
        intra_core_sync.md
    - advanced_programming.md
    - ai_framework_adaptation
      - ai_framework_adaptation.md
      - onnx_framework
        onnx_framework.md
    - hardware_implementation
      - architecture_spec
        architecture_spec.md
      - hardware_constraints
        hardware_constraints.md
      - hardware_implementation.md
    - inter_core_communication
      - inter_core_communication.md
    - memory_model
      - memory_model.md
    - operator_graph_development
      - operator_graph_development.md
    - super_kernel
      - super_kernel.md
  - appendix
    - appendix.md
    - common_operations
      - common_operations.md
    - faq
      - faq.md
  - compilation_and_execution
    - compilation_and_execution.md
    - operator_compilation
      - operator_compilation.md
  - debug_and_tuning
    - debug_and_tuning.md
    - functional_debug
      - functional_debug.md
  - language_extension
    - language_extension.md
  - library_api
    - advanced_api
      - advanced_api.md
      - quick_reference
        quick_reference.md
    - basic_api
      - basic_api.md
      - interface_classification
        interface_classification.md
      - quick_reference
        quick_reference.md
    - library_api.md
  - programming_guide.md
  - programming_model
    - ai_core_simd_programming
      - ai_core_simd_programming.md
      - c_pointer_programming
        c_pointer_programming.md
      - cpp_tensor_programming
        cpp_tensor_programming.md
      - tpipe_tque_programming
        tpipe_tque_programming.md
    - ai_core_simt_programming
      - ai_core_simt_programming.md
    - programming_model.md
- technical_appendix
  - concepts_and_terms
    - concepts_and_terms.md
    - memory_access
      - memory_access.md
    - neural_networks_and_operators
      - neural_networks_and_operators.md
    - performance_optimization
      - performance_optimization.md
  - cpp_standard_support
    - cpp_standard_support.md
    - syntax_restrictions
      - syntax_restrictions.md
  - technical_appendix.md

共 90 个文件变更+1162-83

Adocs/zh/guide/cross_gen_migration_guide/3510_arch_migration/2201_to_3510_guide/2201_to_3510_guide.md+5-0

@@ -0,0 +1,5 @@
	1	+# 2201迁移3510指导
	2	+
	3	+- [基础API迁移指导](./basic_api_migration.md)
	4	+- [高阶API迁移指导](./advanced_api_migration.md)
	5	+- [算子编译迁移指导](./op_compilation_migration.md)

Adocs/zh/guide/cross_gen_migration_guide/3510_arch_migration/3510_arch_migration.md+7-0

@@ -0,0 +1,7 @@
	1	+# 3510架构迁移指导
	2	+
	3	+- [2201到3510架构变更](./2201_to_3510_arch_changes.md)
	4	+- [2201迁移3510指导](./2201_to_3510_guide/2201_to_3510_guide.md)
	5	+ - [基础API迁移指导](./2201_to_3510_guide/basic_api_migration.md)
	6	+ - [高阶API迁移指导](./2201_to_3510_guide/advanced_api_migration.md)
	7	+ - [算子编译迁移指导](./2201_to_3510_guide/op_compilation_migration.md)

Adocs/zh/guide/cross_gen_migration_guide/cross_gen_migration_guide.md+12-0

@@ -0,0 +1,12 @@
	1	+# 跨代迁移兼容性指南
	2	+
	3	+- [概述](./overview.md)
	4	+- [Ascend C API兼容策略](./ascend_c_api_compatibility.md)
	5	+- [3510架构迁移指导](./3510_arch_migration/3510_arch_migration.md)
	6	+ - [2201到3510架构变更](./3510_arch_migration/2201_to_3510_arch_changes.md)
	7	+ - [2201迁移3510指导](./3510_arch_migration/2201_to_3510_guide/2201_to_3510_guide.md)
	8	+ - [基础API迁移指导](./3510_arch_migration/2201_to_3510_guide/basic_api_migration.md)
	9	+ - [高阶API迁移指导](./3510_arch_migration/2201_to_3510_guide/advanced_api_migration.md)
	10	+ - [算子编译迁移指导](./3510_arch_migration/2201_to_3510_guide/op_compilation_migration.md)
	11	+- [新特性使用指导](./instructions_for_new_features/instructions_for_new_features.md)
	12	+ - [3510新特性使用说明](./instructions_for_new_features/3510_new_features.md)

Adocs/zh/guide/cross_gen_migration_guide/instructions_for_new_features/instructions_for_new_features.md+3-0

@@ -0,0 +1,3 @@
	1	+# 新特性使用指导
	2	+
	3	+- [3510新特性使用说明](./3510_new_features.md)

Adocs/zh/guide/getting_started/getting_started.md+12-0

@@ -0,0 +1,12 @@
	1	+# 入门教程
	2	+
	3	+- [Ascend C概述与学习路径](./ascend_c_overview_and_learning_path.md)
	4	+- [环境准备](./environment_setup.md)
	5	+- [快速入门](./quick_start/quick_start.md)
	6	+ - [异构系统与编程模型](./quick_start/heterogeneous_system_and_programming_model.md)
	7	+ - [基于SIMD编程](./quick_start/simd_programming/simd_programming.md)
	8	+ - [HelloWorld](./quick_start/simd_programming/hello_world.md)
	9	+ - [Add算子快速入门](./quick_start/simd_programming/add_op_quick_start.md)
	10	+ - [基于SIMT编程](./quick_start/simt_programming/simt_programming.md)
	11	+ - [HelloWorld](./quick_start/simt_programming/hello_world.md)
	12	+ - [Gather算子快速入门](./quick_start/simt_programming/gather_op_quick_start.md)

Adocs/zh/guide/getting_started/quick_start/quick_start.md+9-0

@@ -0,0 +1,9 @@
	1	+# 快速入门
	2	+
	3	+- [异构系统与编程模型](./heterogeneous_system_and_programming_model.md)
	4	+- [基于SIMD编程](./simd_programming/simd_programming.md)
	5	+ - [HelloWorld](./simd_programming/hello_world.md)
	6	+ - [Add算子快速入门](./simd_programming/add_op_quick_start.md)
	7	+- [基于SIMT编程](./simt_programming/simt_programming.md)
	8	+ - [HelloWorld](./simt_programming/hello_world.md)
	9	+ - [Gather算子快速入门](./simt_programming/gather_op_quick_start.md)

Mdocs/zh/guide/getting_started/quick_start/simd_programming/simd_programming.md+2-1

	@@ -1,3 +1,4 @@
1	# 基于SIMD编程	1	# 基于SIMD编程
2		2
3	-本文件为文档发布时所需的索引文件。	3	+- [HelloWorld](./hello_world.md)
		4	+- [Add算子快速入门](./add_op_quick_start.md)

Mdocs/zh/guide/getting_started/quick_start/simt_programming/simt_programming.md+2-1

	@@ -1,3 +1,4 @@
1	# 基于SIMT编程	1	# 基于SIMT编程
2		2
3	-本文件为文档发布时所需的索引文件。	3	+- [HelloWorld](./hello_world.md)
		4	+- [Gather算子快速入门](./gather_op_quick_start.md)

Mdocs/zh/guide/index.md+51-51

Mdocs/zh/guide/operator_practice/best_practices/best_practices.md+19-1

	@@ -1,3 +1,21 @@
1	# 优秀实践	1	# 优秀实践
2		2
3	-本文件为文档发布时所需的索引文件。	3	+- [FlashAttention算子性能调优案例](./flash_attention_tuning.md)
		4	+- [GroupedMatmul算子性能调优案例](./grouped_matmul_tuning.md)
		5	+- [MC²算子性能调优案例](./mc_operator_tuning.md)
		6	+- [Matmul性能调优案例](./matmul_tuning_cases/matmul_tuning_cases.md)
		7	+ - [Matmul性能优化策略总览](./matmul_tuning_cases/matmul_optimization_overview.md)
		8	+ - [Matmul算子优化Tiling策略](./matmul_tuning_cases/matmul_tiling_strategy.md)
		9	+ - [Matmul高阶API开启纯Cube模式](./matmul_tuning_cases/matmul_pure_cube.md)
		10	+ - [Matmul高阶API开启MDL模板](./matmul_tuning_cases/matmul_mdl_template.md)
		11	+ - [Matmul高阶API开启UnitFlag](./matmul_tuning_cases/matmul_unit_flag.md)
		12	+ - [Matmul高阶API开启Tiling全量常量化](./matmul_tuning_cases/matmul_tiling_constant.md)
		13	+ - [Matmul高阶API开启L2 Cache切分](./matmul_tuning_cases/matmul_l2_cache_split.md)
		14	+ - [Matmul高阶API开启多核切K](./matmul_tuning_cases/matmul_multi_core_split_k.md)
		15	+ - [Matmul高阶API开启多核K轴错峰访问内存](./matmul_tuning_cases/matmul_multi_core_k_stagger.md)
		16	+ - [Matmul高阶API开启NBuffer33模板](./matmul_tuning_cases/matmul_nbuffer33.md)
		17	+ - [Matmul高阶API开启IBShare模板共享B矩阵数据](./matmul_tuning_cases/matmul_ibshare_b.md)
		18	+ - [Matmul高阶API开启IBShare模板共享A和B矩阵数据](./matmul_tuning_cases/matmul_ibshare_ab.md)
		19	+ - [AIV核上的ND2NZ格式转换](./matmul_tuning_cases/aiv_nd2nz_conversion.md)
		20	+ - [Matmul高阶API开启MTE2 Preload](./matmul_tuning_cases/matmul_mte2_preload.md)
		21	+- [SIMT实现InsertHashTable算子案例](./simt_insert_hash_table.md)

Adocs/zh/guide/operator_practice/best_practices/matmul_tuning_cases/matmul_tuning_cases.md+16-0

@@ -0,0 +1,16 @@
	1	+# Matmul性能调优案例
	2	+
	3	+- [Matmul性能优化策略总览](./matmul_optimization_overview.md)
	4	+- [Matmul算子优化Tiling策略](./matmul_tiling_strategy.md)
	5	+- [Matmul高阶API开启纯Cube模式](./matmul_pure_cube.md)
	6	+- [Matmul高阶API开启MDL模板](./matmul_mdl_template.md)
	7	+- [Matmul高阶API开启UnitFlag](./matmul_unit_flag.md)
	8	+- [Matmul高阶API开启Tiling全量常量化](./matmul_tiling_constant.md)
	9	+- [Matmul高阶API开启L2 Cache切分](./matmul_l2_cache_split.md)
	10	+- [Matmul高阶API开启多核切K](./matmul_multi_core_split_k.md)
	11	+- [Matmul高阶API开启多核K轴错峰访问内存](./matmul_multi_core_k_stagger.md)
	12	+- [Matmul高阶API开启NBuffer33模板](./matmul_nbuffer33.md)
	13	+- [Matmul高阶API开启IBShare模板共享B矩阵数据](./matmul_ibshare_b.md)
	14	+- [Matmul高阶API开启IBShare模板共享A和B矩阵数据](./matmul_ibshare_ab.md)
	15	+- [AIV核上的ND2NZ格式转换](./aiv_nd2nz_conversion.md)
	16	+- [Matmul高阶API开启MTE2 Preload](./matmul_mte2_preload.md)

Mdocs/zh/guide/operator_practice/functional_debug/functional_debug.md+3-1

	@@ -1,3 +1,5 @@
1	# 功能调试	1	# 功能调试
2		2
3	-本文件为文档发布时所需的索引文件。	3	+- [运行正常](./running_normally.md)
		4	+- [精度正常](./accuracy_normal.md)
		5	+- [算子调试](./operator_debug.md)

Adocs/zh/guide/operator_practice/operator_practice.md+156-0

@@ -0,0 +1,156 @@
	1	+# 算子实践参考
	2	+
	3	+- [本文档组织结构](./document_structure.md)
	4	+- [异构计算](./heterogeneous_computing.md)
	5	+- [SIMD算子实现](./simd_operator_impl/simd_operator_impl.md)
	6	+ - [概述](./simd_operator_impl/overview.md)
	7	+ - [矢量编程](./simd_operator_impl/vector_programming/vector_programming.md)
	8	+ - [概述](./simd_operator_impl/vector_programming/overview.md)
	9	+ - [基础矢量算子](./simd_operator_impl/vector_programming/basic_vector_operator.md)
	10	+ - [TBuf的使用](./simd_operator_impl/vector_programming/tbuf_usage.md)
	11	+ - [多核Tiling切分](./simd_operator_impl/vector_programming/multi_core_tiling/multi_core_tiling.md)
	12	+ - [概述](./simd_operator_impl/vector_programming/multi_core_tiling/overview.md)
	13	+ - [主块均分](./simd_operator_impl/vector_programming/multi_core_tiling/main_block_even_split.md)
	14	+ - [尾块均分](./simd_operator_impl/vector_programming/multi_core_tiling/tail_block_even_split.md)
	15	+ - [尾核切分](./simd_operator_impl/vector_programming/multi_core_tiling/tail_core_split.md)
	16	+ - [尾核尾块切分](./simd_operator_impl/vector_programming/multi_core_tiling/tail_core_tail_block_split.md)
	17	+ - [Double Buffer场景](./simd_operator_impl/vector_programming/double_buffer_scenario.md)
	18	+ - [Broadcast场景](./simd_operator_impl/vector_programming/broadcast_scenario.md)
	19	+ - [非对齐场景](./simd_operator_impl/vector_programming/unaligned_scenario.md)
	20	+ - [矩阵编程（高阶API）](./simd_operator_impl/matrix_advanced_api/matrix_advanced_api.md)
	21	+ - [基础知识](./simd_operator_impl/matrix_advanced_api/basics.md)
	22	+ - [算子实现](./simd_operator_impl/matrix_advanced_api/operator_impl.md)
	23	+ - [特性场景](./simd_operator_impl/matrix_advanced_api/feature_scenarios/feature_scenarios.md)
	24	+ - [Matmul特性介绍](./simd_operator_impl/matrix_advanced_api/feature_scenarios/matmul_features.md)
	25	+ - [多核对齐切分](./simd_operator_impl/matrix_advanced_api/feature_scenarios/multi_core_aligned_split.md)
	26	+ - [多核非对齐切分](./simd_operator_impl/matrix_advanced_api/feature_scenarios/multi_core_unaligned_split.md)
	27	+ - [异步场景处理](./simd_operator_impl/matrix_advanced_api/feature_scenarios/async_scenario.md)
	28	+ - [矩阵乘输出的量化/反量化](./simd_operator_impl/matrix_advanced_api/feature_scenarios/matmul_output_quant_dequant.md)
	29	+ - [矩阵乘输出的Channel拆分](./simd_operator_impl/matrix_advanced_api/feature_scenarios/matmul_output_channel_split.md)
	30	+ - [矩阵向量乘](./simd_operator_impl/matrix_advanced_api/feature_scenarios/matrix_vector_multiply.md)
	31	+ - [4:2稀疏矩阵乘](./simd_operator_impl/matrix_advanced_api/feature_scenarios/4_2_sparse_matmul.md)
	32	+ - [TSCM输入的矩阵乘](./simd_operator_impl/matrix_advanced_api/feature_scenarios/tscm_input_matmul.md)
	33	+ - [矩阵乘输出的N方向对齐](./simd_operator_impl/matrix_advanced_api/feature_scenarios/matmul_output_n_alignment.md)
	34	+ - [单次矩阵乘局部输出](./simd_operator_impl/matrix_advanced_api/feature_scenarios/single_matmul_local_output.md)
	35	+ - [AIC和AIV独立运行机制](./simd_operator_impl/matrix_advanced_api/feature_scenarios/aic_aiv_independent_execution.md)
	36	+ - [MxMatmul场景](./simd_operator_impl/matrix_advanced_api/feature_scenarios/mx_matmul_scenario.md)
	37	+ - [Batch Matmul基础功能](./simd_operator_impl/matrix_advanced_api/feature_scenarios/batch_matmul_basics.md)
	38	+ - [Batch Matmul复用Bias矩阵](./simd_operator_impl/matrix_advanced_api/feature_scenarios/batch_matmul_reuse_bias.md)
	39	+ - [矩阵编程（基础API）](./simd_operator_impl/matrix_basic_api/matrix_basic_api.md)
	40	+ - [耦合模式](./simd_operator_impl/matrix_basic_api/coupled_mode.md)
	41	+ - [分离模式](./simd_operator_impl/matrix_basic_api/decoupled_mode.md)
	42	+ - [融合算子编程](./simd_operator_impl/fusion_operator_programming/fusion_operator_programming.md)
	43	+ - [CV融合](./simd_operator_impl/fusion_operator_programming/cv_fusion/cv_fusion.md)
	44	+ - [基础知识](./simd_operator_impl/fusion_operator_programming/cv_fusion/basics.md)
	45	+ - [算子实现](./simd_operator_impl/fusion_operator_programming/cv_fusion/operator_impl.md)
	46	+ - [基于SSBuffer的核间通信特性](./simd_operator_impl/fusion_operator_programming/cv_fusion/ssbuffer_feature_scenarios.md)
	47	+ - [通算融合](./simd_operator_impl/fusion_operator_programming/general_fusion/general_fusion.md)
	48	+ - [基础知识](./simd_operator_impl/fusion_operator_programming/general_fusion/basics.md)
	49	+ - [算子实现](./simd_operator_impl/fusion_operator_programming/general_fusion/operator_impl.md)
	50	+ - [特性场景](./simd_operator_impl/fusion_operator_programming/general_fusion/feature_scenarios.md)
	51	+- [SIMT算子实现](./simt_operator_impl/simt_operator_impl.md)
	52	+ - [基础知识](./simt_operator_impl/basics.md)
	53	+ - [算子实现](./simt_operator_impl/operator_impl.md)
	54	+- [SIMD与SIMT混合算子实现](./simd_simt_hybrid_operator_impl/simd_simt_hybrid_operator_impl.md)
	55	+ - [基础知识](./simd_simt_hybrid_operator_impl/basics.md)
	56	+ - [算子实现](./simd_simt_hybrid_operator_impl/operator_impl.md)
	57	+- [功能调试](./functional_debug/functional_debug.md)
	58	+ - [运行正常](./functional_debug/running_normally.md)
	59	+ - [精度正常](./functional_debug/accuracy_normal.md)
	60	+ - [算子调试](./functional_debug/operator_debug.md)
	61	+- [性能分析](./performance_analysis/performance_analysis.md)
	62	+ - [获取性能数据](./performance_analysis/collect_performance_data.md)
	63	+ - [分析性能数据](./performance_analysis/analyze_performance_data.md)
	64	+- [SIMD算子性能优化](./simd_operator_optimization/simd_operator_optimization.md)
	65	+ - [优化建议总览表](./simd_operator_optimization/optimization_summary.md)
	66	+ - [Tiling策略](./simd_operator_optimization/tiling_strategy/tiling_strategy.md)
	67	+ - [核间负载均衡](./simd_operator_optimization/tiling_strategy/inter_core_load_balance.md)
	68	+ - [头尾开销优化](./simd_operator_optimization/overhead_optimization/overhead_optimization.md)
	69	+ - [设置合适的核数和算子Kernel类型](./simd_operator_optimization/overhead_optimization/core_count_kernel_type.md)
	70	+ - [限制TilingData结构大小](./simd_operator_optimization/overhead_optimization/limit_tiling_data_size.md)
	71	+ - [避免TPipe在对象内创建和初始化](./simd_operator_optimization/overhead_optimization/avoid_tpipe_init_in_object.md)
	72	+ - [核函数内删除Workspace相关冗余操作](./simd_operator_optimization/overhead_optimization/remove_workspace_redundancy.md)
	73	+ - [设置DCI编译选项来减少算子尾开销](./simd_operator_optimization/overhead_optimization/dci_compile_option.md)
	74	+ - [流水编排](./simd_operator_optimization/pipeline_scheduling/pipeline_scheduling.md)
	75	+ - [开启Double Buffer](./simd_operator_optimization/pipeline_scheduling/enable_double_buffer.md)
	76	+ - [启用Iterate或IterateAll异步接口避免AIC/AIV同步依赖](./simd_operator_optimization/pipeline_scheduling/enable_iterate_avoid_sync.md)
	77	+ - [内存访问](./simd_operator_optimization/memory_access/memory_access.md)
	78	+ - [尽量一次搬运较大的数据块](./simd_operator_optimization/memory_access/transfer_larger_data_blocks.md)
	79	+ - [GM地址尽量512B对齐](./simd_operator_optimization/memory_access/gm_address_512b_alignment.md)
	80	+ - [高效的使用搬运API](./simd_operator_optimization/memory_access/efficient_transfer_api.md)
	81	+ - [非对齐场景减少无效数据的搬运](./simd_operator_optimization/memory_access/reduce_invalid_data_transfer.md)
	82	+ - [非连续搬运场景减少搬运次数](./simd_operator_optimization/memory_access/reduce_non_contiguous_transfer.md)
	83	+ - [避免同地址访问](./simd_operator_optimization/memory_access/avoid_same_address_access.md)
	84	+ - [设置合理的L2 CacheMode](./simd_operator_optimization/memory_access/l2_cache_mode_config.md)
	85	+ - [算子与高阶API共享临时Buffer](./simd_operator_optimization/memory_access/shared_temp_buffer.md)
	86	+ - [纯搬运类算子VECIN和VECOUT建议复用](./simd_operator_optimization/memory_access/vecin_vecout_reuse.md)
	87	+ - [通过缩减Tensor ShapeInfo维度，优化栈空间](./simd_operator_optimization/memory_access/reduce_shapeinfo_optimize_stack.md)
	88	+ - [避免UB的bank冲突](./simd_operator_optimization/memory_access/avoid_ub_bank_conflict/avoid_ub_bank_conflict.md)
	89	+ - [概述](./simd_operator_optimization/memory_access/avoid_ub_bank_conflict/overview.md)
	90	+ - [避免bank冲突（NPU架构版本2201）](./simd_operator_optimization/memory_access/avoid_ub_bank_conflict/avoid_bank_conflict_npu_arch_2201.md)
	91	+ - [避免bank冲突（NPU架构版本3510）](./simd_operator_optimization/memory_access/avoid_ub_bank_conflict/avoid_bank_conflict_npu_arch_3510.md)
	92	+ - [L2 Cache切分](./simd_operator_optimization/memory_access/l2_cache_split.md)
	93	+ - [矢量计算](./simd_operator_optimization/vector_compute/vector_compute.md)
	94	+ - [通过Unified Buffer融合实现连续vector计算](./simd_operator_optimization/vector_compute/ub_fusion_vector_compute.md)
	95	+ - [Vector算子灵活运用Counter模式](./simd_operator_optimization/vector_compute/vector_counter_mode.md)
	96	+ - [选择低延迟指令，优化归约操作性能](./simd_operator_optimization/vector_compute/low_latency_instruction.md)
	97	+ - [VF性能优化](./simd_operator_optimization/vector_compute/vf_optimization/vf_optimization.md)
	98	+ - [VF循环优化](./simd_operator_optimization/vector_compute/vf_optimization/vf_loop_optimization.md)
	99	+ - [指令双发优化](./simd_operator_optimization/vector_compute/vf_optimization/dual_issue_optimization.md)
	100	+ - [连续非对齐场景优化](./simd_operator_optimization/vector_compute/vf_optimization/continuous_unaligned_optimization.md)
	101	+ - [VF融合优化](./simd_operator_optimization/vector_compute/vf_optimization/vf_fusion_optimization.md)
	102	+ - [基于全局掩码复用的计算性能优化](./simd_operator_optimization/vector_compute/mask_reuse_optimization.md)
	103	+ - [矩阵计算](./simd_operator_optimization/matrix_compute/matrix_compute.md)
	104	+ - [使用hifloat8_t实现Matmul性能优化](./simd_operator_optimization/matrix_compute/hifloat8_matmul_optimization.md)
	105	+ - [通过BT Buffer实现高效的bias计算](./simd_operator_optimization/matrix_compute/bt_buffer_bias_compute.md)
	106	+ - [通过FP Buffer存放量化参数实现高效随路量化](./simd_operator_optimization/matrix_compute/fp_buffer_quantization.md)
	107	+ - [通过L0C Buffer数据暂存实现高效的矩阵乘结果累加](./simd_operator_optimization/matrix_compute/l0c_buffer_matmul_accumulate.md)
	108	+ - [较小矩阵长驻L1 Buffer，仅分次搬运较大矩阵](./simd_operator_optimization/matrix_compute/small_matrix_l1_buffer.md)
	109	+ - [Matmul开启AtomicAdd选项](./simd_operator_optimization/matrix_compute/matmul_atomic_add.md)
	110	+- [SIMT算子性能优化](./simt_operator_optimization/simt_operator_optimization.md)
	111	+ - [概述](./simt_operator_optimization/overview.md)
	112	+ - [内存访问](./simt_operator_optimization/memory_access/memory_access.md)
	113	+ - [访存合并](./simt_operator_optimization/memory_access/memory_access_merge.md)
	114	+ - [避免UB的Bank冲突](./simt_operator_optimization/memory_access/avoid_ub_bank_conflict_simt.md)
	115	+ - [DCache访问优化](./simt_operator_optimization/memory_access/dcache_access.md)
	116	+ - [通过类型对齐提升搬运效率](./simt_operator_optimization/memory_access/type_alignment_transfer.md)
	117	+ - [使用短向量类型提升效率](./simt_operator_optimization/memory_access/short_vector_efficiency.md)
	118	+ - [执行配置](./simt_operator_optimization/execution_config/execution_config.md)
	119	+ - [线程块数量配置优化](./simt_operator_optimization/execution_config/thread_block_config.md)
	120	+ - [合理配置线程数避免寄存器溢出](./simt_operator_optimization/execution_config/thread_count_register_overflow.md)
	121	+ - [控制流](./simt_operator_optimization/control_flow/control_flow.md)
	122	+ - [同一Warp内避免分支跳转](./simt_operator_optimization/control_flow/avoid_warp_branch.md)
	123	+ - [分支处理](./simt_operator_optimization/control_flow/branch_handing.md)
	124	+ - [指令优化](./simt_operator_optimization/instruction_optimization/instruction_optimization.md)
	125	+ - [原子操作指令优化](./simt_operator_optimization/instruction_optimization/atomic_instruction_optimization.md)
	126	+ - [原地加法指令优化](./simt_operator_optimization/instruction_optimization/inplace_add_instruction_optimization.md)
	127	+- [SIMD与SIMT混合算子性能优化](./simd_simt_hybrid_optimization/simd_simt_hybrid_optimization.md)
	128	+ - [概述](./simd_simt_hybrid_optimization/overview.md)
	129	+ - [内存访问](./simd_simt_hybrid_optimization/memory_access/memory_access.md)
	130	+ - [使用Unified Buffer提升内存访问效率](./simd_simt_hybrid_optimization/memory_access/ub_memory_access.md)
	131	+ - [指令优化](./simd_simt_hybrid_optimization/instruction_optimization/instruction_optimization.md)
	132	+ - [整数除法的快速算法](./simd_simt_hybrid_optimization/instruction_optimization/fast_integer_division.md)
	133	+ - [计算优化](./simd_simt_hybrid_optimization/compute_optimization/compute_optimization.md)
	134	+ - [通过SIMT实现分支判断](./simd_simt_hybrid_optimization/compute_optimization/simt_branch_judgment.md)
	135	+ - [执行控制](./simd_simt_hybrid_optimization/execution_control/execution_control.md)
	136	+ - [线程块数量配置优化](./simd_simt_hybrid_optimization/execution_control/thread_block_config.md)
	137	+- [优秀实践](./best_practices/best_practices.md)
	138	+ - [FlashAttention算子性能调优案例](./best_practices/flash_attention_tuning.md)
	139	+ - [GroupedMatmul算子性能调优案例](./best_practices/grouped_matmul_tuning.md)
	140	+ - [MC²算子性能调优案例](./best_practices/mc_operator_tuning.md)
	141	+ - [Matmul性能调优案例](./best_practices/matmul_tuning_cases/matmul_tuning_cases.md)
	142	+ - [Matmul性能优化策略总览](./best_practices/matmul_tuning_cases/matmul_optimization_overview.md)
	143	+ - [Matmul算子优化Tiling策略](./best_practices/matmul_tuning_cases/matmul_tiling_strategy.md)
	144	+ - [Matmul高阶API开启纯Cube模式](./best_practices/matmul_tuning_cases/matmul_pure_cube.md)
	145	+ - [Matmul高阶API开启MDL模板](./best_practices/matmul_tuning_cases/matmul_mdl_template.md)
	146	+ - [Matmul高阶API开启UnitFlag](./best_practices/matmul_tuning_cases/matmul_unit_flag.md)
	147	+ - [Matmul高阶API开启Tiling全量常量化](./best_practices/matmul_tuning_cases/matmul_tiling_constant.md)
	148	+ - [Matmul高阶API开启L2 Cache切分](./best_practices/matmul_tuning_cases/matmul_l2_cache_split.md)
	149	+ - [Matmul高阶API开启多核切K](./best_practices/matmul_tuning_cases/matmul_multi_core_split_k.md)
	150	+ - [Matmul高阶API开启多核K轴错峰访问内存](./best_practices/matmul_tuning_cases/matmul_multi_core_k_stagger.md)
	151	+ - [Matmul高阶API开启NBuffer33模板](./best_practices/matmul_tuning_cases/matmul_nbuffer33.md)
	152	+ - [Matmul高阶API开启IBShare模板共享B矩阵数据](./best_practices/matmul_tuning_cases/matmul_ibshare_b.md)
	153	+ - [Matmul高阶API开启IBShare模板共享A和B矩阵数据](./best_practices/matmul_tuning_cases/matmul_ibshare_ab.md)
	154	+ - [AIV核上的ND2NZ格式转换](./best_practices/matmul_tuning_cases/aiv_nd2nz_conversion.md)
	155	+ - [Matmul高阶API开启MTE2 Preload](./best_practices/matmul_tuning_cases/matmul_mte2_preload.md)
	156	+ - [SIMT实现InsertHashTable算子案例](./best_practices/simt_insert_hash_table.md)

Mdocs/zh/guide/operator_practice/performance_analysis/performance_analysis.md+2-1

	@@ -1,3 +1,4 @@
1	# 性能分析	1	# 性能分析
2		2
3	-本文件为文档发布时所需的索引文件。	3	+- [获取性能数据](./collect_performance_data.md)
		4	+- [分析性能数据](./analyze_performance_data.md)

Adocs/zh/guide/operator_practice/simd_operator_impl/fusion_operator_programming/cv_fusion/cv_fusion.md+5-0

@@ -0,0 +1,5 @@
	1	+# CV融合
	2	+
	3	+- [基础知识](./basics.md)
	4	+- [算子实现](./operator_impl.md)
	5	+- [基于SSBuffer的核间通信特性](./ssbuffer_feature_scenarios.md)

Mdocs/zh/guide/operator_practice/simd_operator_impl/fusion_operator_programming/fusion_operator_programming.md+10-0

	@@ -1 +1,11 @@
1	# 融合算子编程	1	# 融合算子编程
		2	+
		3	+- [CV融合](./cv_fusion/basics.md)
		4	+ - [基础知识](./cv_fusion/basics.md)
		5	+ - [算子实现](./cv_fusion/operator_impl.md)
		6	+ - [基于SSBuffer的核间通信特性](./cv_fusion/ssbuffer_feature_scenarios.md)
		7	+
		8	+- [通算融合](./general_fusion/basics.md)
		9	+ - [基础知识](./general_fusion/basics.md)
		10	+ - [算子实现](./general_fusion/operator_impl.md)
		11	+ - [特性场景](./general_fusion/feature_scenarios.md)

Adocs/zh/guide/operator_practice/simd_operator_impl/fusion_operator_programming/general_fusion/general_fusion.md+5-0

@@ -0,0 +1,5 @@
	1	+# 通算融合
	2	+
	3	+- [基础知识](./basics.md)
	4	+- [算子实现](./operator_impl.md)
	5	+- [特性场景](./feature_scenarios.md)

Adocs/zh/guide/operator_practice/simd_operator_impl/matrix_advanced_api/feature_scenarios/feature_scenarios.md+17-0

@@ -0,0 +1,17 @@
	1	+# 特性场景
	2	+
	3	+- [Matmul特性介绍](./matmul_features.md)
	4	+- [多核对齐切分](./multi_core_aligned_split.md)
	5	+- [多核非对齐切分](./multi_core_unaligned_split.md)
	6	+- [异步场景处理](./async_scenario.md)
	7	+- [矩阵乘输出的量化/反量化](./matmul_output_quant_dequant.md)
	8	+- [矩阵乘输出的Channel拆分](./matmul_output_channel_split.md)
	9	+- [矩阵向量乘](./matrix_vector_multiply.md)
	10	+- [4:2稀疏矩阵乘](./4_2_sparse_matmul.md)
	11	+- [TSCM输入的矩阵乘](./tscm_input_matmul.md)
	12	+- [矩阵乘输出的N方向对齐](./matmul_output_n_alignment.md)
	13	+- [单次矩阵乘局部输出](./single_matmul_local_output.md)
	14	+- [AIC和AIV独立运行机制](./aic_aiv_independent_execution.md)
	15	+- [MxMatmul场景](./mx_matmul_scenario.md)
	16	+- [Batch Matmul基础功能](./batch_matmul_basics.md)
	17	+- [Batch Matmul复用Bias矩阵](./batch_matmul_reuse_bias.md)

Mdocs/zh/guide/operator_practice/simd_operator_impl/matrix_advanced_api/matrix_advanced_api.md+19-0

	@@ -1 +1,20 @@
1	# 矩阵编程（高阶API）	1	# 矩阵编程（高阶API）
		2	+
		3	+- [基础知识](basics.md)
		4	+- [算子实现](operator_impl.md)
		5	+- [特性场景](./feature_scenarios/matmul_features.md)
		6	+ - [Matmul特性介绍](./feature_scenarios/matmul_features.md)
		7	+ - [多核对齐切分](./feature_scenarios/multi_core_aligned_split.md)
		8	+ - [多核非对齐切分](./feature_scenarios/multi_core_unaligned_split.md)
		9	+ - [异步场景处理](./feature_scenarios/async_scenario.md)
		10	+ - [矩阵乘输出的量化/反量化](./feature_scenarios/matmul_output_quant_dequant.md)
		11	+ - [矩阵乘输出的Channel拆分](./feature_scenarios/matmul_output_channel_split.md)
		12	+ - [矩阵向量乘](./feature_scenarios/matrix_vector_multiply.md)
		13	+ - [4:2稀疏矩阵乘](./feature_scenarios/4_2_sparse_matmul.md)
		14	+ - [TSCM输入的矩阵乘](./feature_scenarios/tscm_input_matmul.md)
		15	+ - [矩阵乘输出的N方向对齐](./feature_scenarios/matmul_output_n_alignment.md)
		16	+ - [单次矩阵乘局部输出](./feature_scenarios/single_matmul_local_output.md)
		17	+ - [AIC和AIV独立运行机制](./feature_scenarios/aic_aiv_independent_execution.md)
		18	+ - [MxMatmul场景](./feature_scenarios/mx_matmul_scenario.md)
		19	+ - [Batch Matmul基础功能](./feature_scenarios/batch_matmul_basics.md)
		20	+ - [Batch Matmul复用Bias矩阵](./feature_scenarios/batch_matmul_reuse_bias.md)

Adocs/zh/guide/operator_practice/simd_operator_impl/matrix_basic_api/matrix_basic_api.md+4-0

@@ -0,0 +1,4 @@
	1	+# 矩阵编程（基础API）
	2	+
	3	+- [耦合模式](./coupled_mode.md)
	4	+- [分离模式](./decoupled_mode.md)

Mdocs/zh/guide/operator_practice/simd_operator_impl/simd_operator_impl.md+46-0

	@@ -1 +1,47 @@
1	# SIMD算子实现	1	# SIMD算子实现
		2	+
		3	+- [概述](./overview.md)
		4	+- [矢量编程](./vector_programming/vector_programming.md)
		5	+ - [概述](./vector_programming/overview.md)
		6	+ - [基础矢量算子](./vector_programming/basic_vector_operator.md)
		7	+ - [TBuf的使用](./vector_programming/tbuf_usage.md)
		8	+ - [多核Tiling切分](./vector_programming/multi_core_tiling/multi_core_tiling.md)
		9	+ - [概述](./vector_programming/multi_core_tiling/overview.md)
		10	+ - [主块均分](./vector_programming/multi_core_tiling/main_block_even_split.md)
		11	+ - [尾块均分](./vector_programming/multi_core_tiling/tail_block_even_split.md)
		12	+ - [尾核切分](./vector_programming/multi_core_tiling/tail_core_split.md)
		13	+ - [尾核尾块切分](./vector_programming/multi_core_tiling/tail_core_tail_block_split.md)
		14	+ - [Double Buffer场景](./vector_programming/double_buffer_scenario.md)
		15	+ - [Broadcast场景](./vector_programming/broadcast_scenario.md)
		16	+ - [非对齐场景](./vector_programming/unaligned_scenario.md)
		17	+- [矩阵编程（高阶API）](./matrix_advanced_api/matrix_advanced_api.md)
		18	+ - [基础知识](./matrix_advanced_api/basics.md)
		19	+ - [算子实现](./matrix_advanced_api/operator_impl.md)
		20	+ - [特性场景](./matrix_advanced_api/feature_scenarios/feature_scenarios.md)
		21	+ - [Matmul特性介绍](./matrix_advanced_api/feature_scenarios/matmul_features.md)
		22	+ - [多核对齐切分](./matrix_advanced_api/feature_scenarios/multi_core_aligned_split.md)
		23	+ - [多核非对齐切分](./matrix_advanced_api/feature_scenarios/multi_core_unaligned_split.md)
		24	+ - [异步场景处理](./matrix_advanced_api/feature_scenarios/async_scenario.md)
		25	+ - [矩阵乘输出的量化/反量化](./matrix_advanced_api/feature_scenarios/matmul_output_quant_dequant.md)
		26	+ - [矩阵乘输出的Channel拆分](./matrix_advanced_api/feature_scenarios/matmul_output_channel_split.md)
		27	+ - [矩阵向量乘](./matrix_advanced_api/feature_scenarios/matrix_vector_multiply.md)
		28	+ - [4:2稀疏矩阵乘](./matrix_advanced_api/feature_scenarios/4_2_sparse_matmul.md)
		29	+ - [TSCM输入的矩阵乘](./matrix_advanced_api/feature_scenarios/tscm_input_matmul.md)
		30	+ - [矩阵乘输出的N方向对齐](./matrix_advanced_api/feature_scenarios/matmul_output_n_alignment.md)
		31	+ - [单次矩阵乘局部输出](./matrix_advanced_api/feature_scenarios/single_matmul_local_output.md)
		32	+ - [AIC和AIV独立运行机制](./matrix_advanced_api/feature_scenarios/aic_aiv_independent_execution.md)
		33	+ - [MxMatmul场景](./matrix_advanced_api/feature_scenarios/mx_matmul_scenario.md)
		34	+ - [Batch Matmul基础功能](./matrix_advanced_api/feature_scenarios/batch_matmul_basics.md)
		35	+ - [Batch Matmul复用Bias矩阵](./matrix_advanced_api/feature_scenarios/batch_matmul_reuse_bias.md)
		36	+- [矩阵编程（基础API）](./matrix_basic_api/matrix_basic_api.md)
		37	+ - [耦合模式](./matrix_basic_api/coupled_mode.md)
		38	+ - [分离模式](./matrix_basic_api/decoupled_mode.md)
		39	+- [融合算子编程](./fusion_operator_programming/fusion_operator_programming.md)
		40	+ - [CV融合](./fusion_operator_programming/cv_fusion/cv_fusion.md)
		41	+ - [基础知识](./fusion_operator_programming/cv_fusion/basics.md)
		42	+ - [算子实现](./fusion_operator_programming/cv_fusion/operator_impl.md)
		43	+ - [基于SSBuffer的核间通信特性](./fusion_operator_programming/cv_fusion/ssbuffer_feature_scenarios.md)
		44	+ - [通算融合](./fusion_operator_programming/general_fusion/general_fusion.md)
		45	+ - [基础知识](./fusion_operator_programming/general_fusion/basics.md)
		46	+ - [算子实现](./fusion_operator_programming/general_fusion/operator_impl.md)
		47	+ - [特性场景](./fusion_operator_programming/general_fusion/feature_scenarios.md)

Adocs/zh/guide/operator_practice/simd_operator_impl/vector_programming/multi_core_tiling/multi_core_tiling.md+7-0

@@ -0,0 +1,7 @@
	1	+# 多核Tiling切分
	2	+
	3	+- [概述](./overview.md)
	4	+- [主块均分](./main_block_even_split.md)
	5	+- [尾块均分](./tail_block_even_split.md)
	6	+- [尾核切分](./tail_core_split.md)
	7	+- [尾核尾块切分](./tail_core_tail_block_split.md)

Mdocs/zh/guide/operator_practice/simd_operator_impl/vector_programming/vector_programming.md+12-1

	@@ -1,3 +1,14 @@
1	# 矢量编程	1	# 矢量编程
2		2
3	-本文件为文档发布时所需的索引文件。	3	+- [概述](./overview.md)
		4	+- [基础矢量算子](./basic_vector_operator.md)
		5	+- [TBuf的使用](./tbuf_usage.md)
		6	+- [多核Tiling切分](./multi_core_tiling/multi_core_tiling.md)
		7	+ - [概述](./multi_core_tiling/overview.md)
		8	+ - [主块均分](./multi_core_tiling/main_block_even_split.md)
		9	+ - [尾块均分](./multi_core_tiling/tail_block_even_split.md)
		10	+ - [尾核切分](./multi_core_tiling/tail_core_split.md)
		11	+ - [尾核尾块切分](./multi_core_tiling/tail_core_tail_block_split.md)
		12	+- [Double Buffer场景](./double_buffer_scenario.md)
		13	+- [Broadcast场景](./broadcast_scenario.md)
		14	+- [非对齐场景](./unaligned_scenario.md)

Mdocs/zh/guide/operator_practice/simd_operator_optimization/matrix_compute/matrix_compute.md+6-1

	@@ -1,3 +1,8 @@
1	# 矩阵计算	1	# 矩阵计算
2		2
3	-本文件为文档发布时所需的索引文件。	3	+- [使用hifloat8_t实现Matmul性能优化](./hifloat8_matmul_optimization.md)
		4	+- [通过BT Buffer实现高效的bias计算](./bt_buffer_bias_compute.md)
		5	+- [通过FP Buffer存放量化参数实现高效随路量化](./fp_buffer_quantization.md)
		6	+- [通过L0C Buffer数据暂存实现高效的矩阵乘结果累加](./l0c_buffer_matmul_accumulate.md)
		7	+- [较小矩阵长驻L1 Buffer，仅分次搬运较大矩阵](./small_matrix_l1_buffer.md)
		8	+- [Matmul开启AtomicAdd选项](./matmul_atomic_add.md)

Mdocs/zh/guide/operator_practice/simd_operator_optimization/memory_access/avoid_ub_bank_conflict/avoid_ub_bank_conflict.md+3-1

	@@ -1,3 +1,5 @@
1	# 避免UB的bank冲突	1	# 避免UB的bank冲突
2		2
3	-本文件为文档发布时所需的索引文件。	3	+- [概述](./overview.md)
		4	+- [避免bank冲突（NPU架构版本2201）](./avoid_bank_conflict_npu_arch_2201.md)
		5	+- [避免bank冲突（NPU架构版本3510）](./avoid_bank_conflict_npu_arch_3510.md)

Mdocs/zh/guide/operator_practice/simd_operator_optimization/memory_access/memory_access.md+15-1

	@@ -1,3 +1,17 @@
1	# 内存访问	1	# 内存访问
2		2
3	-本文件为文档发布时所需的索引文件。	3	+- [尽量一次搬运较大的数据块](./transfer_larger_data_blocks.md)
		4	+- [GM地址尽量512B对齐](./gm_address_512b_alignment.md)
		5	+- [高效的使用搬运API](./efficient_transfer_api.md)
		6	+- [非对齐场景减少无效数据的搬运](./reduce_invalid_data_transfer.md)
		7	+- [非连续搬运场景减少搬运次数](./reduce_non_contiguous_transfer.md)
		8	+- [避免同地址访问](./avoid_same_address_access.md)
		9	+- [设置合理的L2 CacheMode](./l2_cache_mode_config.md)
		10	+- [算子与高阶API共享临时Buffer](./shared_temp_buffer.md)
		11	+- [纯搬运类算子VECIN和VECOUT建议复用](./vecin_vecout_reuse.md)
		12	+- [通过缩减Tensor ShapeInfo维度，优化栈空间](./reduce_shapeinfo_optimize_stack.md)
		13	+- [避免UB的bank冲突](./avoid_ub_bank_conflict/avoid_ub_bank_conflict.md)
		14	+ - [概述](./avoid_ub_bank_conflict/overview.md)
		15	+ - [避免bank冲突（NPU架构版本2201）](./avoid_ub_bank_conflict/avoid_bank_conflict_npu_arch_2201.md)
		16	+ - [避免bank冲突（NPU架构版本3510）](./avoid_ub_bank_conflict/avoid_bank_conflict_npu_arch_3510.md)
		17	+- [L2 Cache切分](./l2_cache_split.md)

Mdocs/zh/guide/operator_practice/simd_operator_optimization/overhead_optimization/overhead_optimization.md+5-1

	@@ -1,3 +1,7 @@
1	# 头尾开销优化	1	# 头尾开销优化
2		2
3	-本文件为文档发布时所需的索引文件。	3	+- [设置合适的核数和算子Kernel类型](./core_count_kernel_type.md)
		4	+- [限制TilingData结构大小](./limit_tiling_data_size.md)
		5	+- [避免TPipe在对象内创建和初始化](./avoid_tpipe_init_in_object.md)
		6	+- [核函数内删除Workspace相关冗余操作](./remove_workspace_redundancy.md)
		7	+- [设置DCI编译选项来减少算子尾开销](./dci_compile_option.md)

Mdocs/zh/guide/operator_practice/simd_operator_optimization/pipeline_scheduling/pipeline_scheduling.md+2-1

	@@ -1,3 +1,4 @@
1	# 流水编排	1	# 流水编排
2		2
3	-本文件为文档发布时所需的索引文件。	3	+- [开启Double Buffer](./enable_double_buffer.md)
		4	+- [启用Iterate或IterateAll异步接口避免AIC/AIV同步依赖](./enable_iterate_avoid_sync.md)

Mdocs/zh/guide/operator_practice/simd_operator_optimization/simd_operator_optimization.md+45-1

	@@ -1,3 +1,47 @@
1	# SIMD算子性能优化	1	# SIMD算子性能优化
2		2
3	-本文件为文档发布时所需的索引文件。	3	+- [优化建议总览表](./optimization_summary.md)
		4	+- [Tiling策略](./tiling_strategy/tiling_strategy.md)
		5	+ - [核间负载均衡](./tiling_strategy/inter_core_load_balance.md)
		6	+- [头尾开销优化](./overhead_optimization/overhead_optimization.md)
		7	+ - [设置合适的核数和算子Kernel类型](./overhead_optimization/core_count_kernel_type.md)
		8	+ - [限制TilingData结构大小](./overhead_optimization/limit_tiling_data_size.md)
		9	+ - [避免TPipe在对象内创建和初始化](./overhead_optimization/avoid_tpipe_init_in_object.md)
		10	+ - [核函数内删除Workspace相关冗余操作](./overhead_optimization/remove_workspace_redundancy.md)
		11	+ - [设置DCI编译选项来减少算子尾开销](./overhead_optimization/dci_compile_option.md)
		12	+- [流水编排](./pipeline_scheduling/pipeline_scheduling.md)
		13	+ - [开启Double Buffer](./pipeline_scheduling/enable_double_buffer.md)
		14	+ - [启用Iterate或IterateAll异步接口避免AIC/AIV同步依赖](./pipeline_scheduling/enable_iterate_avoid_sync.md)
		15	+- [内存访问](./memory_access/memory_access.md)
		16	+ - [尽量一次搬运较大的数据块](./memory_access/transfer_larger_data_blocks.md)
		17	+ - [GM地址尽量512B对齐](./memory_access/gm_address_512b_alignment.md)
		18	+ - [高效的使用搬运API](./memory_access/efficient_transfer_api.md)
		19	+ - [非对齐场景减少无效数据的搬运](./memory_access/reduce_invalid_data_transfer.md)
		20	+ - [非连续搬运场景减少搬运次数](./memory_access/reduce_non_contiguous_transfer.md)
		21	+ - [避免同地址访问](./memory_access/avoid_same_address_access.md)
		22	+ - [设置合理的L2 CacheMode](./memory_access/l2_cache_mode_config.md)
		23	+ - [算子与高阶API共享临时Buffer](./memory_access/shared_temp_buffer.md)
		24	+ - [纯搬运类算子VECIN和VECOUT建议复用](./memory_access/vecin_vecout_reuse.md)
		25	+ - [通过缩减Tensor ShapeInfo维度，优化栈空间](./memory_access/reduce_shapeinfo_optimize_stack.md)
		26	+ - [避免UB的bank冲突](./memory_access/avoid_ub_bank_conflict/avoid_ub_bank_conflict.md)
		27	+ - [概述](./memory_access/avoid_ub_bank_conflict/overview.md)
		28	+ - [避免bank冲突（NPU架构版本2201）](./memory_access/avoid_ub_bank_conflict/avoid_bank_conflict_npu_arch_2201.md)
		29	+ - [避免bank冲突（NPU架构版本3510）](./memory_access/avoid_ub_bank_conflict/avoid_bank_conflict_npu_arch_3510.md)
		30	+ - [L2 Cache切分](./memory_access/l2_cache_split.md)
		31	+- [矢量计算](./vector_compute/vector_compute.md)
		32	+ - [通过Unified Buffer融合实现连续vector计算](./vector_compute/ub_fusion_vector_compute.md)
		33	+ - [Vector算子灵活运用Counter模式](./vector_compute/vector_counter_mode.md)
		34	+ - [选择低延迟指令，优化归约操作性能](./vector_compute/low_latency_instruction.md)
		35	+ - [VF性能优化](./vector_compute/vf_optimization/vf_optimization.md)
		36	+ - [VF循环优化](./vector_compute/vf_optimization/vf_loop_optimization.md)
		37	+ - [指令双发优化](./vector_compute/vf_optimization/dual_issue_optimization.md)
		38	+ - [连续非对齐场景优化](./vector_compute/vf_optimization/continuous_unaligned_optimization.md)
		39	+ - [VF融合优化](./vector_compute/vf_optimization/vf_fusion_optimization.md)
		40	+ - [基于全局掩码复用的计算性能优化](./vector_compute/mask_reuse_optimization.md)
		41	+- [矩阵计算](./matrix_compute/matrix_compute.md)
		42	+ - [使用hifloat8_t实现Matmul性能优化](./matrix_compute/hifloat8_matmul_optimization.md)
		43	+ - [通过BT Buffer实现高效的bias计算](./matrix_compute/bt_buffer_bias_compute.md)
		44	+ - [通过FP Buffer存放量化参数实现高效随路量化](./matrix_compute/fp_buffer_quantization.md)
		45	+ - [通过L0C Buffer数据暂存实现高效的矩阵乘结果累加](./matrix_compute/l0c_buffer_matmul_accumulate.md)
		46	+ - [较小矩阵长驻L1 Buffer，仅分次搬运较大矩阵](./matrix_compute/small_matrix_l1_buffer.md)
		47	+ - [Matmul开启AtomicAdd选项](./matrix_compute/matmul_atomic_add.md)

Mdocs/zh/guide/operator_practice/simd_operator_optimization/tiling_strategy/tiling_strategy.md+1-1

	@@ -1,3 +1,3 @@
1	# Tiling策略	1	# Tiling策略
2		2
3	-本文件为文档发布时所需的索引文件。	3	+- [核间负载均衡](./inter_core_load_balance.md)

Mdocs/zh/guide/operator_practice/simd_operator_optimization/vector_compute/vector_compute.md+9-1

@@ @@ -1,3 +1,11 @@ @@
 # 矢量计算
++- [通过Unified Buffer融合实现连续vector计算](./ub_fusion_vector_compute.md)
++- [Vector算子灵活运用Counter模式](./vector_counter_mode.md)
++- [选择低延迟指令，优化归约操作性能](./low_latency_instruction.md)
++- [VF性能优化](./vf_optimization/vf_optimization.md)
++  - [VF循环优化](./vf_optimization/vf_loop_optimization.md)
++  - [指令双发优化](./vf_optimization/dual_issue_optimization.md)
++  - [连续非对齐场景优化](./vf_optimization/continuous_unaligned_optimization.md)
++  - [VF融合优化](./vf_optimization/vf_fusion_optimization.md)
++- [基于全局掩码复用的计算性能优化](./mask_reuse_optimization.md)

Adocs/zh/guide/operator_practice/simd_operator_optimization/vector_compute/vf_optimization/vf_optimization.md+6-0

@@ -0,0 +1,6 @@
	1	+# VF性能优化
	2	+
	3	+- [VF循环优化](./vf_loop_optimization.md)
	4	+- [指令双发优化](./dual_issue_optimization.md)
	5	+- [连续非对齐场景优化](./continuous_unaligned_optimization.md)
	6	+- [VF融合优化](./vf_fusion_optimization.md)

Mdocs/zh/guide/operator_practice/simd_simt_hybrid_operator_impl/simd_simt_hybrid_operator_impl.md+2-1

	@@ -1,3 +1,4 @@
1	# SIMD与SIMT混合算子实现	1	# SIMD与SIMT混合算子实现
2		2
3	-本文件为文档发布时所需的索引文件。	3	+- [基础知识](./basics.md)
		4	+- [算子实现](./operator_impl.md)

Adocs/zh/guide/operator_practice/simd_simt_hybrid_optimization/compute_optimization/compute_optimization.md+3-0

@@ -0,0 +1,3 @@
	1	+# 计算优化
	2	+
	3	+- [通过SIMT实现分支判断](./simt_branch_judgment.md)

Adocs/zh/guide/operator_practice/simd_simt_hybrid_optimization/execution_control/execution_control.md+3-0

@@ -0,0 +1,3 @@
	1	+# 执行控制
	2	+
	3	+- [线程块数量配置优化](./thread_block_config.md)

Adocs/zh/guide/operator_practice/simd_simt_hybrid_optimization/instruction_optimization/instruction_optimization.md+3-0

@@ -0,0 +1,3 @@
	1	+# 指令优化
	2	+
	3	+- [整数除法的快速算法](./fast_integer_division.md)

Adocs/zh/guide/operator_practice/simd_simt_hybrid_optimization/memory_access/memory_access.md+3-0

@@ -0,0 +1,3 @@
	1	+# 内存访问
	2	+
	3	+- [使用Unified Buffer提升内存访问效率](./ub_memory_access.md)

Mdocs/zh/guide/operator_practice/simd_simt_hybrid_optimization/simd_simt_hybrid_optimization.md+9-1

@@ @@ -1,3 +1,11 @@ @@
 # SIMD与SIMT混合算子性能优化
++- [概述](./overview.md)
++- [内存访问](./memory_access/memory_access.md)
++  - [使用Unified Buffer提升内存访问效率](./memory_access/ub_memory_access.md)
++- [指令优化](./instruction_optimization/instruction_optimization.md)
++  - [整数除法的快速算法](./instruction_optimization/fast_integer_division.md)
++- [计算优化](./compute_optimization/compute_optimization.md)
++  - [通过SIMT实现分支判断](./compute_optimization/simt_branch_judgment.md)
++- [执行控制](./execution_control/execution_control.md)
++  - [线程块数量配置优化](./execution_control/thread_block_config.md)

Mdocs/zh/guide/operator_practice/simt_operator_impl/simt_operator_impl.md+3-0

	@@ -1 +1,4 @@
1	# SIMT算子实现	1	# SIMT算子实现
		2	+
		3	+- [基础知识](basics.md)
		4	+- [算子实现](operator_impl.md)

Adocs/zh/guide/operator_practice/simt_operator_optimization/control_flow/control_flow.md+4-0

@@ -0,0 +1,4 @@
	1	+# 控制流
	2	+
	3	+- [同一Warp内避免分支跳转](./avoid_warp_branch.md)
	4	+- [分支处理](./branch_handing.md)

Adocs/zh/guide/operator_practice/simt_operator_optimization/execution_config/execution_config.md+4-0

@@ -0,0 +1,4 @@
	1	+# 执行配置
	2	+
	3	+- [线程块数量配置优化](./thread_block_config.md)
	4	+- [合理配置线程数避免寄存器溢出](./thread_count_register_overflow.md)

Adocs/zh/guide/operator_practice/simt_operator_optimization/instruction_optimization/instruction_optimization.md+4-0

@@ -0,0 +1,4 @@
	1	+# 指令优化
	2	+
	3	+- [原子操作指令优化](./atomic_instruction_optimization.md)
	4	+- [原地加法指令优化](./inplace_add_instruction_optimization.md)

Adocs/zh/guide/operator_practice/simt_operator_optimization/memory_access/memory_access.md+7-0

@@ -0,0 +1,7 @@
	1	+# 内存访问
	2	+
	3	+- [访存合并](./memory_access_merge.md)
	4	+- [避免UB的Bank冲突](./avoid_ub_bank_conflict_simt.md)
	5	+- [DCache访问优化](./dcache_access.md)
	6	+- [通过类型对齐提升搬运效率](./type_alignment_transfer.md)
	7	+- [使用短向量类型提升效率](./short_vector_efficiency.md)

Mdocs/zh/guide/operator_practice/simt_operator_optimization/simt_operator_optimization.md+16-1

	@@ -1,3 +1,18 @@
1	# SIMT算子性能优化	1	# SIMT算子性能优化
2		2
3	-本文件为文档发布时所需的索引文件。	3	+- [概述](./overview.md)
		4	+- [内存访问](./memory_access/memory_access.md)
		5	+ - [访存合并](./memory_access/memory_access_merge.md)
		6	+ - [避免UB的Bank冲突](./memory_access/avoid_ub_bank_conflict_simt.md)
		7	+ - [DCache访问优化](./memory_access/dcache_access.md)
		8	+ - [通过类型对齐提升搬运效率](./memory_access/type_alignment_transfer.md)
		9	+ - [使用短向量类型提升效率](./memory_access/short_vector_efficiency.md)
		10	+- [执行配置](./execution_config/execution_config.md)
		11	+ - [线程块数量配置优化](./execution_config/thread_block_config.md)
		12	+ - [合理配置线程数避免寄存器溢出](./execution_config/thread_count_register_overflow.md)
		13	+- [控制流](./control_flow/control_flow.md)
		14	+ - [同一Warp内避免分支跳转](./control_flow/avoid_warp_branch.md)
		15	+ - [分支处理](./control_flow/branch_handing.md)
		16	+- [指令优化](./instruction_optimization/instruction_optimization.md)
		17	+ - [原子操作指令优化](./instruction_optimization/atomic_instruction_optimization.md)
		18	+ - [原地加法指令优化](./instruction_optimization/inplace_add_instruction_optimization.md)

Adocs/zh/guide/programming_guide/advanced_programming/aclnn_operator_development/aclnn_operator_development.md+31-0

@@ -0,0 +1,31 @@
	1	+# Aclnn算子工程化开发
	2	+
	3	+- [概述](./overview.md)
	4	+- [快速入门](./aclnn_quick_start.md)
	5	+- [设计与实现](./design_and_implementation/design_and_implementation.md)
	6	+ - [算子功能设计](./design_and_implementation/operator_function_design.md)
	7	+ - [算子原型定义](./design_and_implementation/operator_prototype_definition.md)
	8	+ - [Host侧Tiling实现](./design_and_implementation/host_tiling_implementation.md)
	9	+ - [核函数（Kernel）侧算子实现](./design_and_implementation/kernel_operator_implementation.md)
	10	+ - [多分支策略](./design_and_implementation/multi_branch_strategy.md)
	11	+ - [GenSimplifiedKey回调实现](./design_and_implementation/gen_simplified_key.md)
	12	+ - [通过TilingData传递属性信息](./design_and_implementation/tiling_data_attributes.md)
	13	+ - [使用高阶API时配套的Tiling实现](./design_and_implementation/tiling_with_advanced_api.md)
	14	+ - [核函数（Kernel）侧输出shape依赖计算](./design_and_implementation/kernel_output_shape_computation.md)
	15	+- [编译与部署](./compilation_and_deployment/compilation_and_deployment.md)
	16	+ - [基本流程](./compilation_and_deployment/basic_process.md)
	17	+ - [交叉编译](./compilation_and_deployment/cross_compilation.md)
	18	+ - [多算子包组织](./compilation_and_deployment/multi_operator_package.md)
	19	+ - [算子动态库和静态库编译](./compilation_and_deployment/dynamic_static_lib_compilation.md)
	20	+ - [编译加速](./compilation_and_deployment/compilation_acceleration.md)
	21	+ - [编译过程调试](./compilation_and_deployment/compilation_debug.md)
	22	+ - [simplified key二进制配置JSON](./compilation_and_deployment/simplified_key_json_configuration.md)
	23	+- [调用](./invocation/invocation.md)
	24	+ - [单算子API调用](./invocation/single_operator_api_call.md)
	25	+ - [运行时加载机制](./invocation/runtime_loading_mechanism.md)
	26	+ - [基于simplified key的运行时选择流程](./invocation/simplified_key_runtime_flow.md)
	27	+- [附录](./appendix/appendix.md)
	28	+ - [aclnn工程化算子开发与核函数（Kernel）直调对比](./appendix/kernel_direct_call_comparison.md)
	29	+ - [命名转换规则对照表](./appendix/naming_conversion_table.md)
	30	+ - [CMake函数参考](./appendix/cmake_function_reference.md)
	31	+ - [外部样例链接汇总](./appendix/external_sample_links.md)

Adocs/zh/guide/programming_guide/advanced_programming/aclnn_operator_development/compilation_and_deployment/compilation_and_deployment.md+9-0

@@ -0,0 +1,9 @@
	1	+# 编译与部署
	2	+
	3	+- [基本流程](./basic_process.md)
	4	+- [交叉编译](./cross_compilation.md)
	5	+- [多算子包组织](./multi_operator_package.md)
	6	+- [算子动态库和静态库编译](./dynamic_static_lib_compilation.md)
	7	+- [编译加速](./compilation_acceleration.md)
	8	+- [编译过程调试](./compilation_debug.md)
	9	+- [simplified key二进制配置JSON](./simplified_key_json_configuration.md)

Adocs/zh/guide/programming_guide/advanced_programming/aclnn_operator_development/design_and_implementation/design_and_implementation.md+11-0

@@ -0,0 +1,11 @@
	1	+# 设计与实现
	2	+
	3	+- [算子功能设计](./operator_function_design.md)
	4	+- [算子原型定义](./operator_prototype_definition.md)
	5	+- [Host侧Tiling实现](./host_tiling_implementation.md)
	6	+- [核函数（Kernel）侧算子实现](./kernel_operator_implementation.md)
	7	+- [多分支策略](./multi_branch_strategy.md)
	8	+- [GenSimplifiedKey回调实现](./gen_simplified_key.md)
	9	+- [通过TilingData传递属性信息](./tiling_data_attributes.md)
	10	+- [使用高阶API时配套的Tiling实现](./tiling_with_advanced_api.md)
	11	+- [核函数（Kernel）侧输出shape依赖计算](./kernel_output_shape_computation.md)

Adocs/zh/guide/programming_guide/advanced_programming/aclnn_operator_development/invocation/invocation.md+5-0

@@ -0,0 +1,5 @@
	1	+# 调用
	2	+
	3	+- [单算子API调用](./single_operator_api_call.md)
	4	+- [运行时加载机制](./runtime_loading_mechanism.md)
	5	+- [基于simplified key的运行时选择流程](./simplified_key_runtime_flow.md)

Mdocs/zh/guide/programming_guide/advanced_programming/advanced_ai_core_programming_model/advanced_ai_core_programming_model.md+8-1

	@@ -1 +1,8 @@
1	-# 高级AI-Core编程模型	1	+# 高级AI-Core编程模型
		2	+
		3	+- [SIMD与SIMT混合编程](./simd_simt_hybrid_programming/simd_simt_hybrid_programming.md)
		4	+ - [概述](./simd_simt_hybrid_programming/overview.md)
		5	+ - [抽象硬件架构](./simd_simt_hybrid_programming/abstract_hardware_architecture.md)
		6	+ - [核函数（Kernel）与VF函数](./simd_simt_hybrid_programming/kernel_and_vf_functions.md)
		7	+ - [内存层级](./simd_simt_hybrid_programming/memory_hierarchy.md)
		8	+ - [编程示例](./simd_simt_hybrid_programming/programming_examples.md)

Adocs/zh/guide/programming_guide/advanced_programming/advanced_ai_core_programming_model/simd_simt_hybrid_programming/simd_simt_hybrid_programming.md+7-0

@@ -0,0 +1,7 @@
	1	+# SIMD与SIMT混合编程
	2	+
	3	+- [概述](./overview.md)
	4	+- [抽象硬件架构](./abstract_hardware_architecture.md)
	5	+- [核函数（Kernel）与VF函数](./kernel_and_vf_functions.md)
	6	+- [内存层级](./memory_hierarchy.md)
	7	+- [编程示例](./programming_examples.md)

Mdocs/zh/guide/programming_guide/advanced_programming/advanced_kernel_programming/advanced_kernel_programming.md+2-1

	@@ -1,3 +1,4 @@
1	# 高级核函数编程	1	# 高级核函数编程
2		2
3	-本文件为文档发布时所需的索引文件。	3	+- [核内同步](./intra_core_sync/intra_core_sync.md)
		4	+ - [基于Mutex实现多流水同步](./intra_core_sync/mutex_multi_pipeline_sync.md)

Mdocs/zh/guide/programming_guide/advanced_programming/advanced_kernel_programming/intra_core_sync/intra_core_sync.md+1-1

	@@ -1,3 +1,3 @@
1	# 核内同步	1	# 核内同步
2		2
3	-本文件为文档发布时所需的索引文件。	3	+- [基于Mutex实现多流水同步](./mutex_multi_pipeline_sync.md)

Mdocs/zh/guide/programming_guide/advanced_programming/advanced_programming.md+75-1

	@@ -1,3 +1,77 @@
1	# 高级编程	1	# 高级编程
2		2
3	-本文件为文档发布时所需的索引文件。	3	+- [硬件实现](./hardware_implementation/hardware_implementation.md)
		4	+ - [基本架构](./hardware_implementation/basic_architecture.md)
		5	+ - [架构规格](./hardware_implementation/architecture_spec/architecture_spec.md)
		6	+ - [NPU架构版本2002](./hardware_implementation/architecture_spec/npu_arch_2002.md)
		7	+ - [NPU架构版本2201](./hardware_implementation/architecture_spec/npu_arch_2201.md)
		8	+ - [NPU架构版本3002](./hardware_implementation/architecture_spec/npu_arch_3002.md)
		9	+ - [NPU架构版本3510](./hardware_implementation/architecture_spec/npu_arch_3510.md)
		10	+ - [硬件约束](./hardware_implementation/hardware_constraints/hardware_constraints.md)
		11	+ - [NPU架构版本2002](./hardware_implementation/hardware_constraints/npu_arch_2002.md)
		12	+ - [NPU架构版本2201](./hardware_implementation/hardware_constraints/npu_arch_2201.md)
		13	+ - [NPU架构版本3510](./hardware_implementation/hardware_constraints/npu_arch_3510.md)
		14	+- [内存模型](./memory_model/memory_model.md)
		15	+ - [缓存一致性](./memory_model/cache_coherence.md)
		16	+ - [内存一致性](./memory_model/memory_consistency.md)
		17	+- [核间通信](./inter_core_communication/inter_core_communication.md)
		18	+ - [使用SSBuffer实现核间通信](./inter_core_communication/ssbuffer_inter_core_memory_feature.md)
		19	+- [高级核函数编程](./advanced_kernel_programming/advanced_kernel_programming.md)
		20	+ - [核内同步](./advanced_kernel_programming/intra_core_sync/intra_core_sync.md)
		21	+ - [基于Mutex实现多流水同步](./advanced_kernel_programming/intra_core_sync/mutex_multi_pipeline_sync.md)
		22	+- [高级AI Core编程模型](./advanced_ai_core_programming_model/advanced_ai_core_programming_model.md)
		23	+ - [SIMD与SIMT混合编程](./advanced_ai_core_programming_model/simd_simt_hybrid_programming/simd_simt_hybrid_programming.md)
		24	+ - [概述](./advanced_ai_core_programming_model/simd_simt_hybrid_programming/overview.md)
		25	+ - [抽象硬件架构](./advanced_ai_core_programming_model/simd_simt_hybrid_programming/abstract_hardware_architecture.md)
		26	+ - [核函数（Kernel）与VF函数](./advanced_ai_core_programming_model/simd_simt_hybrid_programming/kernel_and_vf_functions.md)
		27	+ - [内存层级](./advanced_ai_core_programming_model/simd_simt_hybrid_programming/memory_hierarchy.md)
		28	+ - [编程示例](./advanced_ai_core_programming_model/simd_simt_hybrid_programming/programming_examples.md)
		29	+- [SuperKernel](./super_kernel/super_kernel.md)
		30	+ - [原理介绍](./super_kernel/principles.md)
		31	+ - [算子适配说明](./super_kernel/operator_adaptation.md)
		32	+ - [核函数（Kernel）直调算子额外适配说明](./super_kernel/kernel_direct_call_adaptation.md)
		33	+ - [算子自验证说明](./super_kernel/operator_self_verification.md)
		34	+- [Aclnn算子工程化开发](./aclnn_operator_development/aclnn_operator_development.md)
		35	+ - [概述](./aclnn_operator_development/overview.md)
		36	+ - [快速入门](./aclnn_operator_development/aclnn_quick_start.md)
		37	+ - [设计与实现](./aclnn_operator_development/design_and_implementation/design_and_implementation.md)
		38	+ - [算子功能设计](./aclnn_operator_development/design_and_implementation/operator_function_design.md)
		39	+ - [算子原型定义](./aclnn_operator_development/design_and_implementation/operator_prototype_definition.md)
		40	+ - [Host侧Tiling实现](./aclnn_operator_development/design_and_implementation/host_tiling_implementation.md)
		41	+ - [核函数（Kernel）侧算子实现](./aclnn_operator_development/design_and_implementation/kernel_operator_implementation.md)
		42	+ - [多分支策略](./aclnn_operator_development/design_and_implementation/multi_branch_strategy.md)
		43	+ - [GenSimplifiedKey回调实现](./aclnn_operator_development/design_and_implementation/gen_simplified_key.md)
		44	+ - [通过TilingData传递属性信息](./aclnn_operator_development/design_and_implementation/tiling_data_attributes.md)
		45	+ - [使用高阶API时配套的Tiling实现](./aclnn_operator_development/design_and_implementation/tiling_with_advanced_api.md)
		46	+ - [核函数（Kernel）侧输出shape依赖计算](./aclnn_operator_development/design_and_implementation/kernel_output_shape_computation.md)
		47	+ - [编译与部署](./aclnn_operator_development/compilation_and_deployment/compilation_and_deployment.md)
		48	+ - [基本流程](./aclnn_operator_development/compilation_and_deployment/basic_process.md)
		49	+ - [交叉编译](./aclnn_operator_development/compilation_and_deployment/cross_compilation.md)
		50	+ - [多算子包组织](./aclnn_operator_development/compilation_and_deployment/multi_operator_package.md)
		51	+ - [算子动态库和静态库编译](./aclnn_operator_development/compilation_and_deployment/dynamic_static_lib_compilation.md)
		52	+ - [编译加速](./aclnn_operator_development/compilation_and_deployment/compilation_acceleration.md)
		53	+ - [编译过程调试](./aclnn_operator_development/compilation_and_deployment/compilation_debug.md)
		54	+ - [simplified key二进制配置JSON](./aclnn_operator_development/compilation_and_deployment/simplified_key_json_configuration.md)
		55	+ - [调用](./aclnn_operator_development/invocation/invocation.md)
		56	+ - [单算子API调用](./aclnn_operator_development/invocation/single_operator_api_call.md)
		57	+ - [运行时加载机制](./aclnn_operator_development/invocation/runtime_loading_mechanism.md)
		58	+ - [基于simplified key的运行时选择流程](./aclnn_operator_development/invocation/simplified_key_runtime_flow.md)
		59	+ - [附录](./aclnn_operator_development/appendix/appendix.md)
		60	+ - [aclnn工程化算子开发与核函数（Kernel）直调对比](./aclnn_operator_development/appendix/kernel_direct_call_comparison.md)
		61	+ - [命名转换规则对照表](./aclnn_operator_development/appendix/naming_conversion_table.md)
		62	+ - [CMake函数参考](./aclnn_operator_development/appendix/cmake_function_reference.md)
		63	+ - [外部样例链接汇总](./aclnn_operator_development/appendix/external_sample_links.md)
		64	+- [算子入图开发](./operator_graph_development/operator_graph_development.md)
		65	+ - [概述](./operator_graph_development/overview.md)
		66	+ - [基本开发流程](./operator_graph_development/basic_development_flow.md)
		67	+ - [开启Tiling下沉](./operator_graph_development/enable_tiling_sink.md)
		68	+ - [图编译和图执行](./operator_graph_development/graph_compilation_and_execution.md)
		69	+- [AI框架算子适配](./ai_framework_adaptation/ai_framework_adaptation.md)
		70	+ - [概述](./ai_framework_adaptation/overview.md)
		71	+ - [PyTorch框架](./ai_framework_adaptation/pytorch_framework.md)
		72	+ - [ONNX框架](./ai_framework_adaptation/onnx_framework/onnx_framework.md)
		73	+ - [适配插件开发](./ai_framework_adaptation/onnx_framework/adaptation_plugin_development.md)
		74	+ - [调用样例](./ai_framework_adaptation/onnx_framework/call_examples.md)
		75	+ - [TensorFlow框架](./ai_framework_adaptation/tensorflow_framework.md)
		76	+- [SIMT协作组](./simt_cooperative_groups.md)
		77	+- [AOT编译优化](./aot_compilation_optimization.md)

Adocs/zh/guide/programming_guide/advanced_programming/ai_framework_adaptation/ai_framework_adaptation.md+8-0

@@ -0,0 +1,8 @@
	1	+# AI框架算子适配
	2	+
	3	+- [概述](./overview.md)
	4	+- [PyTorch框架](./pytorch_framework.md)
	5	+- [ONNX框架](./onnx_framework/onnx_framework.md)
	6	+ - [适配插件开发](./onnx_framework/adaptation_plugin_development.md)
	7	+ - [调用样例](./onnx_framework/call_examples.md)
	8	+- [TensorFlow框架](./tensorflow_framework.md)

Adocs/zh/guide/programming_guide/advanced_programming/ai_framework_adaptation/onnx_framework/onnx_framework.md+4-0

@@ -0,0 +1,4 @@
	1	+# ONNX框架
	2	+
	3	+- [适配插件开发](./adaptation_plugin_development.md)
	4	+- [调用样例](./call_examples.md)

Mdocs/zh/guide/programming_guide/advanced_programming/hardware_implementation/architecture_spec/architecture_spec.md+4-1

	@@ -1,3 +1,6 @@
1	# 架构规格	1	# 架构规格
2		2
3	-本文件为文档发布时所需的索引文件。	3	+- [NPU架构版本2002](./npu_arch_2002.md)
		4	+- [NPU架构版本2201](./npu_arch_2201.md)
		5	+- [NPU架构版本3002](./npu_arch_3002.md)
		6	+- [NPU架构版本3510](./npu_arch_3510.md)

Adocs/zh/guide/programming_guide/advanced_programming/hardware_implementation/hardware_constraints/hardware_constraints.md+5-0

@@ -0,0 +1,5 @@
	1	+# 硬件约束
	2	+
	3	+- [NPU架构版本2002](./npu_arch_2002.md)
	4	+- [NPU架构版本2201](./npu_arch_2201.md)
	5	+- [NPU架构版本3510](./npu_arch_3510.md)

Mdocs/zh/guide/programming_guide/advanced_programming/hardware_implementation/hardware_implementation.md+10-1

@@ @@ -1,3 +1,12 @@ @@
 # 硬件实现
++- [基本架构](./basic_architecture.md)
++- [架构规格](./architecture_spec/architecture_spec.md)
++  - [NPU架构版本2002](./architecture_spec/npu_arch_2002.md)
++  - [NPU架构版本2201](./architecture_spec/npu_arch_2201.md)
++  - [NPU架构版本3002](./architecture_spec/npu_arch_3002.md)
++  - [NPU架构版本3510](./architecture_spec/npu_arch_3510.md)
++- [硬件约束](./hardware_constraints/hardware_constraints.md)
++  - [NPU架构版本2002](./hardware_constraints/npu_arch_2002.md)
++  - [NPU架构版本2201](./hardware_constraints/npu_arch_2201.md)
++  - [NPU架构版本3510](./hardware_constraints/npu_arch_3510.md)

Mdocs/zh/guide/programming_guide/advanced_programming/inter_core_communication/inter_core_communication.md+1-1

	@@ -1,3 +1,3 @@
1	# 核间通信	1	# 核间通信
2		2
3	-本文件为文档发布时所需的索引文件。	3	+- [使用SSBuffer实现核间通信](./ssbuffer_inter_core_memory_feature.md)

Mdocs/zh/guide/programming_guide/advanced_programming/memory_model/memory_model.md+2-1

	@@ -1,3 +1,4 @@
1	# 内存模型	1	# 内存模型
2		2
3	-本文件为文档发布时所需的索引文件。	3	+- [缓存一致性](./cache_coherence.md)
		4	+- [内存一致性](./memory_consistency.md)

Adocs/zh/guide/programming_guide/advanced_programming/operator_graph_development/operator_graph_development.md+6-0

@@ -0,0 +1,6 @@
	1	+# 算子入图开发
	2	+
	3	+- [概述](./overview.md)
	4	+- [基本开发流程](./basic_development_flow.md)
	5	+- [开启Tiling下沉](./enable_tiling_sink.md)
	6	+- [图编译和图执行](./graph_compilation_and_execution.md)

Adocs/zh/guide/programming_guide/advanced_programming/super_kernel/super_kernel.md+6-0

@@ -0,0 +1,6 @@
	1	+# SuperKernel
	2	+
	3	+- [原理介绍](./principles.md)
	4	+- [算子适配说明](./operator_adaptation.md)
	5	+- [核函数（Kernel）直调算子额外适配说明](./kernel_direct_call_adaptation.md)
	6	+- [算子自验证说明](./operator_self_verification.md)

Adocs/zh/guide/programming_guide/appendix/appendix.md+21-0

@@ -0,0 +1,21 @@
	1	+# 附录
	2	+
	3	+- [show_kernel_debug_data工具](./show_kernel_debug_data_tool.md)
	4	+- [msobjdump工具](./msobjdump_tool.md)
	5	+- [optype_collector工具](./optype_collector_tool.md)
	6	+- [基于样例工程完成核函数（Kernel）直调](./kernel_direct_call_from_sample.md)
	7	+- [常用操作](./common_operations/common_operations.md)
	8	+ - [如何开发动态输入算子](./common_operations/develop_dynamic_input_operator.md)
	9	+ - [如何在矢量编程时启用Vector Core](./common_operations/enable_vector_core.md)
	10	+ - [如何使用workspace](./common_operations/how_to_use_workspace.md)
	11	+ - [如何进行Tiling调测](./common_operations/tiling_debug.md)
	12	+ - [如何使用Tensor原地操作提升算子性能](./common_operations/tensor_inplace_performance.md)
	13	+- [FAQ](./faq/faq.md)
	14	+ - [核函数（Kernel）运行验证时算子存在精度问题](./faq/kernel_precision_issue.md)
	15	+ - [运行验证时AllocTensor/FreeTensor失败](./faq/alloc_free_tensor_failure.md)
	16	+ - [kernel侧获取Tiling信息不正确](./faq/kernel_tiling_info_incorrect.md)
	17	+ - [核函数（Kernel）编译时报错"error: out of jump/jumpc imm range"](./faq/kernel_compile_jump_range_error.md)
	18	+ - [含有Matmul高阶API的算子精度问题](./faq/matmul_advanced_api_precision.md)
	19	+ - [算子工程编译时出现文件名过长报错](./faq/filename_too_long_error.md)
	20	+ - [调用算子时出现无法打开config.ini的报错](./faq/config_ini_open_error.md)
	21	+ - [算子包部署时出现权限不足报错](./faq/operator_deploy_permission_error.md)

Adocs/zh/guide/programming_guide/appendix/common_operations/common_operations.md+7-0

@@ -0,0 +1,7 @@
	1	+# 常用操作
	2	+
	3	+- [如何开发动态输入算子](./develop_dynamic_input_operator.md)
	4	+- [如何在矢量编程时启用Vector Core](./enable_vector_core.md)
	5	+- [如何使用workspace](./how_to_use_workspace.md)
	6	+- [如何进行Tiling调测](./tiling_debug.md)
	7	+- [如何使用Tensor原地操作提升算子性能](./tensor_inplace_performance.md)

Adocs/zh/guide/programming_guide/appendix/faq/faq.md+10-0

@@ -0,0 +1,10 @@
	1	+# FAQ
	2	+
	3	+- [核函数（Kernel）运行验证时算子存在精度问题](./kernel_precision_issue.md)
	4	+- [运行验证时AllocTensor/FreeTensor失败](./alloc_free_tensor_failure.md)
	5	+- [kernel侧获取Tiling信息不正确](./kernel_tiling_info_incorrect.md)
	6	+- [核函数（Kernel）编译时报错"error: out of jump/jumpc imm range"](./kernel_compile_jump_range_error.md)
	7	+- [含有Matmul高阶API的算子精度问题](./matmul_advanced_api_precision.md)
	8	+- [算子工程编译时出现文件名过长报错](./filename_too_long_error.md)
	9	+- [调用算子时出现无法打开config.ini的报错](./config_ini_open_error.md)
	10	+- [算子包部署时出现权限不足报错](./operator_deploy_permission_error.md)

Mdocs/zh/guide/programming_guide/compilation_and_execution/compilation_and_execution.md+7-1

	@@ -1,3 +1,9 @@
1	# 编译与运行	1	# 编译与运行
2		2
3	-本文件为文档发布时所需的索引文件。	3	+- [异步执行](./async_execution.md)
		4	+- [算子编译](./operator_compilation/operator_compilation.md)
		5	+ - [毕昇编译器](./operator_compilation/bisheng_compiler.md)
		6	+ - [AI Core算子编译基本用法](./operator_compilation/ai_core_operator_compilation.md)
		7	+ - [AI CPU算子编译基本用法](./operator_compilation/ai_cpu_operator_compilation.md)
		8	+ - [RTC运行时编译](./operator_compilation/rtc_runtime_compilation.md)
		9	+ - [约束说明](./operator_compilation/constraints.md)

Adocs/zh/guide/programming_guide/compilation_and_execution/operator_compilation/operator_compilation.md+7-0

@@ -0,0 +1,7 @@
	1	+# 算子编译
	2	+
	3	+- [毕昇编译器](./bisheng_compiler.md)
	4	+- [AI Core算子编译基本用法](./ai_core_operator_compilation.md)
	5	+- [AI CPU算子编译基本用法](./ai_cpu_operator_compilation.md)
	6	+- [RTC运行时编译](./rtc_runtime_compilation.md)
	7	+- [约束说明](./constraints.md)

Mdocs/zh/guide/programming_guide/debug_and_tuning/debug_and_tuning.md+5-1

	@@ -1,3 +1,7 @@
1	# 调试调优	1	# 调试调优
2		2
3	-本文件为文档发布时所需的索引文件。	3	+- [概述](./overview.md)
		4	+- [功能调试](./functional_debug/functional_debug.md)
		5	+ - [CPU域孪生调试](./functional_debug/cpu_twin_debug.md)
		6	+ - [NPU域上板调试](./functional_debug/npu_board_debug.md)
		7	+- [性能调优](./performance_tuning.md)

Mdocs/zh/guide/programming_guide/debug_and_tuning/functional_debug/functional_debug.md+2-1

	@@ -1,3 +1,4 @@
1	# 功能调试	1	# 功能调试
2		2
3	-本文件为文档发布时所需的索引文件。	3	+- [CPU域孪生调试](./cpu_twin_debug.md)
		4	+- [NPU域上板调试](./npu_board_debug.md)

Mdocs/zh/guide/programming_guide/language_extension/language_extension.md+5-1

	@@ -1,3 +1,7 @@
1	# 语言扩展层	1	# 语言扩展层
2		2
3	-本文件为文档发布时所需的索引文件。	3	+- [SIMD BuiltIn关键字](./simd_builtin_keywords.md)
		4	+- [SIMT BuiltIn关键字](./simt_builtin_keywords.md)
		5	+- [SIMD与SIMT混合编程BuiltIn关键字](./simd_simt_hybrid_builtin_keywords.md)
		6	+- [SIMD语言扩展层C API](./simd_language_extension_c_api.md)
		7	+- [SIMT语言扩展层C API](./simt_language_extension_c_api.md)

Adocs/zh/guide/programming_guide/library_api/advanced_api/advanced_api.md+6-0

@@ -0,0 +1,6 @@
	1	+# 高阶API
	2	+
	3	+- [概述](./overview.md)
	4	+- [常用操作速查指导](./quick_reference/quick_reference.md)
	5	+ - [如何使用Tiling依赖的头文件](./quick_reference/how_to_use_tiling_headers.md)
	6	+ - [如何使用核函数（Kernel）侧临时空间](./quick_reference/how_to_use_kernel_temp_space.md)

Adocs/zh/guide/programming_guide/library_api/advanced_api/quick_reference/quick_reference.md+4-0

@@ -0,0 +1,4 @@
	1	+# 常用操作速查指导
	2	+
	3	+- [如何使用Tiling依赖的头文件](./how_to_use_tiling_headers.md)
	4	+- [如何使用核函数（Kernel）侧临时空间](./how_to_use_kernel_temp_space.md)

Adocs/zh/guide/programming_guide/library_api/basic_api/basic_api.md+9-0

@@ -0,0 +1,9 @@
	1	+# 基础API
	2	+
	3	+- [概述](./overview.md)
	4	+- [接口分类说明](./interface_classification/interface_classification.md)
	5	+ - [连续计算API](./interface_classification/continuous_compute_api.md)
	6	+ - [高维切分API](./interface_classification/high_dim_split_api.md)
	7	+- [常用操作速查指导](./quick_reference/quick_reference.md)
	8	+ - [如何使用掩码操作API](./quick_reference/how_to_use_mask_api.md)
	9	+ - [如何使用归约计算API](./quick_reference/how_to_use_reduction_api.md)

Adocs/zh/guide/programming_guide/library_api/basic_api/interface_classification/interface_classification.md+4-0

@@ -0,0 +1,4 @@
	1	+# 接口分类说明
	2	+
	3	+- [连续计算API](./continuous_compute_api.md)
	4	+- [高维切分API](./high_dim_split_api.md)

Adocs/zh/guide/programming_guide/library_api/basic_api/quick_reference/quick_reference.md+4-0

@@ -0,0 +1,4 @@
	1	+# 常用操作速查指导
	2	+
	3	+- [如何使用掩码操作API](./how_to_use_mask_api.md)
	4	+- [如何使用归约计算API](./how_to_use_reduction_api.md)

Mdocs/zh/guide/programming_guide/library_api/library_api.md+15-1

	@@ -1,3 +1,17 @@
1	# 类库API	1	# 类库API
2		2
3	-本文件为文档发布时所需的索引文件。	3	+- [编程接口概述](./programming_interface_overview.md)
		4	+- [基础API](./basic_api/basic_api.md)
		5	+ - [概述](./basic_api/overview.md)
		6	+ - [接口分类说明](./basic_api/interface_classification/interface_classification.md)
		7	+ - [连续计算API](./basic_api/interface_classification/continuous_compute_api.md)
		8	+ - [高维切分API](./basic_api/interface_classification/high_dim_split_api.md)
		9	+ - [常用操作速查指导](./basic_api/quick_reference/quick_reference.md)
		10	+ - [如何使用掩码操作API](./basic_api/quick_reference/how_to_use_mask_api.md)
		11	+ - [如何使用归约计算API](./basic_api/quick_reference/how_to_use_reduction_api.md)
		12	+- [高阶API](./advanced_api/advanced_api.md)
		13	+ - [概述](./advanced_api/overview.md)
		14	+ - [常用操作速查指导](./advanced_api/quick_reference/quick_reference.md)
		15	+ - [如何使用Tiling依赖的头文件](./advanced_api/quick_reference/how_to_use_tiling_headers.md)
		16	+ - [如何使用核函数（Kernel）侧临时空间](./advanced_api/quick_reference/how_to_use_kernel_temp_space.md)
		17	+- [Utils API](./utils_api.md)

Adocs/zh/guide/programming_guide/programming_guide.md+166-0

@@ -0,0 +1,166 @@
	1	+# 编程指南
	2	+
	3	+- [本文档组织结构](./document_structure.md)
	4	+- [编程模型](./programming_model/programming_model.md)
	5	+ - [异构系统](./programming_model/heterogeneous_system.md)
	6	+ - [编程模型概述](./programming_model/programming_model_overview.md)
	7	+ - [AI Core SIMD编程](./programming_model/ai_core_simd_programming/ai_core_simd_programming.md)
	8	+ - [概述](./programming_model/ai_core_simd_programming/overview.md)
	9	+ - [抽象硬件架构](./programming_model/ai_core_simd_programming/abstract_hardware_architecture.md)
	10	+ - [核函数（Kernel）](./programming_model/ai_core_simd_programming/kernel_function.md)
	11	+ - [基于指针的C语言编程](./programming_model/ai_core_simd_programming/c_pointer_programming/c_pointer_programming.md)
	12	+ - [C语言编程概述](./programming_model/ai_core_simd_programming/c_pointer_programming/c_programming_overview.md)
	13	+ - [Memory矢量计算编程](./programming_model/ai_core_simd_programming/c_pointer_programming/memory_vector_computation.md)
	14	+ - [Reg矢量计算编程](./programming_model/ai_core_simd_programming/c_pointer_programming/reg_vector_computation.md)
	15	+ - [Cube矩阵计算编程](./programming_model/ai_core_simd_programming/c_pointer_programming/cube_matrix_computation.md)
	16	+ - [基于Tensor的C++编程](./programming_model/ai_core_simd_programming/cpp_tensor_programming/cpp_tensor_programming.md)
	17	+ - [C++ Tensor编程概述](./programming_model/ai_core_simd_programming/cpp_tensor_programming/cpp_tensor_programming_overview.md)
	18	+ - [Memory矢量计算编程](./programming_model/ai_core_simd_programming/cpp_tensor_programming/memory_vector_computation.md)
	19	+ - [Reg矢量计算编程](./programming_model/ai_core_simd_programming/cpp_tensor_programming/reg_vector_computation.md)
	20	+ - [Cube矩阵计算编程](./programming_model/ai_core_simd_programming/cpp_tensor_programming/cube_matrix_computation.md)
	21	+ - [静态Tensor编程](./programming_model/ai_core_simd_programming/cpp_tensor_programming/static_tensor_programming.md)
	22	+ - [基于TPipe和TQue编程](./programming_model/ai_core_simd_programming/tpipe_tque_programming/tpipe_tque_programming.md)
	23	+ - [TPipe-TQue框架编程原理](./programming_model/ai_core_simd_programming/tpipe_tque_programming/tpipe_tque_principles.md)
	24	+ - [TPipe-TQue框架编程范式](./programming_model/ai_core_simd_programming/tpipe_tque_programming/tpipe_tque_paradigm.md)
	25	+ - [AI Core SIMT编程](./programming_model/ai_core_simt_programming/ai_core_simt_programming.md)
	26	+ - [概述](./programming_model/ai_core_simt_programming/overview.md)
	27	+ - [抽象硬件架构](./programming_model/ai_core_simt_programming/abstract_hardware_architecture.md)
	28	+ - [线程架构](./programming_model/ai_core_simt_programming/thread_architecture.md)
	29	+ - [内存层级](./programming_model/ai_core_simt_programming/memory_hierarchy.md)
	30	+ - [核函数（Kernel）](./programming_model/ai_core_simt_programming/kernel_function.md)
	31	+ - [同步机制](./programming_model/ai_core_simt_programming/synchronization.md)
	32	+ - [原子操作](./programming_model/ai_core_simt_programming/atomic_operations.md)
	33	+ - [编程示例](./programming_model/ai_core_simt_programming/programming_examples.md)
	34	+ - [AI CPU编程](./programming_model/ai_cpu_programming.md)
	35	+- [编译与运行](./compilation_and_execution/compilation_and_execution.md)
	36	+ - [异步执行](./compilation_and_execution/async_execution.md)
	37	+ - [算子编译](./compilation_and_execution/operator_compilation/operator_compilation.md)
	38	+ - [毕昇编译器](./compilation_and_execution/operator_compilation/bisheng_compiler.md)
	39	+ - [AI Core算子编译基本用法](./compilation_and_execution/operator_compilation/ai_core_operator_compilation.md)
	40	+ - [AI CPU算子编译基本用法](./compilation_and_execution/operator_compilation/ai_cpu_operator_compilation.md)
	41	+ - [RTC运行时编译](./compilation_and_execution/operator_compilation/rtc_runtime_compilation.md)
	42	+ - [约束说明](./compilation_and_execution/operator_compilation/constraints.md)
	43	+- [语言扩展层](./language_extension/language_extension.md)
	44	+ - [SIMD BuiltIn关键字](./language_extension/simd_builtin_keywords.md)
	45	+ - [SIMT BuiltIn关键字](./language_extension/simt_builtin_keywords.md)
	46	+ - [SIMD与SIMT混合编程BuiltIn关键字](./language_extension/simd_simt_hybrid_builtin_keywords.md)
	47	+ - [SIMD语言扩展层C API](./language_extension/simd_language_extension_c_api.md)
	48	+ - [SIMT语言扩展层C API](./language_extension/simt_language_extension_c_api.md)
	49	+- [C++类库API](./library_api/library_api.md)
	50	+ - [编程接口概述](./library_api/programming_interface_overview.md)
	51	+ - [基础API](./library_api/basic_api/basic_api.md)
	52	+ - [概述](./library_api/basic_api/overview.md)
	53	+ - [接口分类说明](./library_api/basic_api/interface_classification/interface_classification.md)
	54	+ - [连续计算API](./library_api/basic_api/interface_classification/continuous_compute_api.md)
	55	+ - [高维切分API](./library_api/basic_api/interface_classification/high_dim_split_api.md)
	56	+ - [常用操作速查指导](./library_api/basic_api/quick_reference/quick_reference.md)
	57	+ - [如何使用掩码操作API](./library_api/basic_api/quick_reference/how_to_use_mask_api.md)
	58	+ - [如何使用归约计算API](./library_api/basic_api/quick_reference/how_to_use_reduction_api.md)
	59	+ - [高阶API](./library_api/advanced_api/advanced_api.md)
	60	+ - [概述](./library_api/advanced_api/overview.md)
	61	+ - [常用操作速查指导](./library_api/advanced_api/quick_reference/quick_reference.md)
	62	+ - [如何使用Tiling依赖的头文件](./library_api/advanced_api/quick_reference/how_to_use_tiling_headers.md)
	63	+ - [如何使用核函数（Kernel）侧临时空间](./library_api/advanced_api/quick_reference/how_to_use_kernel_temp_space.md)
	64	+ - [Utils API](./library_api/utils_api.md)
	65	+- [调试调优](./debug_and_tuning/debug_and_tuning.md)
	66	+ - [概述](./debug_and_tuning/overview.md)
	67	+ - [功能调试](./debug_and_tuning/functional_debug/functional_debug.md)
	68	+ - [CPU域孪生调试](./debug_and_tuning/functional_debug/cpu_twin_debug.md)
	69	+ - [NPU域上板调试](./debug_and_tuning/functional_debug/npu_board_debug.md)
	70	+ - [性能调优](./debug_and_tuning/performance_tuning.md)
	71	+- [高级编程](./advanced_programming/advanced_programming.md)
	72	+ - [硬件实现](./advanced_programming/hardware_implementation/hardware_implementation.md)
	73	+ - [基本架构](./advanced_programming/hardware_implementation/basic_architecture.md)
	74	+ - [架构规格](./advanced_programming/hardware_implementation/architecture_spec/architecture_spec.md)
	75	+ - [NPU架构版本2002](./advanced_programming/hardware_implementation/architecture_spec/npu_arch_2002.md)
	76	+ - [NPU架构版本2201](./advanced_programming/hardware_implementation/architecture_spec/npu_arch_2201.md)
	77	+ - [NPU架构版本3002](./advanced_programming/hardware_implementation/architecture_spec/npu_arch_3002.md)
	78	+ - [NPU架构版本3510](./advanced_programming/hardware_implementation/architecture_spec/npu_arch_3510.md)
	79	+ - [硬件约束](./advanced_programming/hardware_implementation/hardware_constraints/hardware_constraints.md)
	80	+ - [NPU架构版本2002](./advanced_programming/hardware_implementation/hardware_constraints/npu_arch_2002.md)
	81	+ - [NPU架构版本2201](./advanced_programming/hardware_implementation/hardware_constraints/npu_arch_2201.md)
	82	+ - [NPU架构版本3510](./advanced_programming/hardware_implementation/hardware_constraints/npu_arch_3510.md)
	83	+ - [内存模型](./advanced_programming/memory_model/memory_model.md)
	84	+ - [缓存一致性](./advanced_programming/memory_model/cache_coherence.md)
	85	+ - [内存一致性](./advanced_programming/memory_model/memory_consistency.md)
	86	+ - [核间通信](./advanced_programming/inter_core_communication/inter_core_communication.md)
	87	+ - [使用SSBuffer实现核间通信](./advanced_programming/inter_core_communication/ssbuffer_inter_core_memory_feature.md)
	88	+ - [高级核函数编程](./advanced_programming/advanced_kernel_programming/advanced_kernel_programming.md)
	89	+ - [核内同步](./advanced_programming/advanced_kernel_programming/intra_core_sync/intra_core_sync.md)
	90	+ - [基于Mutex实现多流水同步](./advanced_programming/advanced_kernel_programming/intra_core_sync/mutex_multi_pipeline_sync.md)
	91	+ - [高级AI Core编程模型](./advanced_programming/advanced_ai_core_programming_model/advanced_ai_core_programming_model.md)
	92	+ - [SIMD与SIMT混合编程](./advanced_programming/advanced_ai_core_programming_model/simd_simt_hybrid_programming/simd_simt_hybrid_programming.md)
	93	+ - [概述](./advanced_programming/advanced_ai_core_programming_model/simd_simt_hybrid_programming/overview.md)
	94	+ - [抽象硬件架构](./advanced_programming/advanced_ai_core_programming_model/simd_simt_hybrid_programming/abstract_hardware_architecture.md)
	95	+ - [核函数（Kernel）与VF函数](./advanced_programming/advanced_ai_core_programming_model/simd_simt_hybrid_programming/kernel_and_vf_functions.md)
	96	+ - [内存层级](./advanced_programming/advanced_ai_core_programming_model/simd_simt_hybrid_programming/memory_hierarchy.md)
	97	+ - [编程示例](./advanced_programming/advanced_ai_core_programming_model/simd_simt_hybrid_programming/programming_examples.md)
	98	+ - [SuperKernel](./advanced_programming/super_kernel/super_kernel.md)
	99	+ - [原理介绍](./advanced_programming/super_kernel/principles.md)
	100	+ - [算子适配说明](./advanced_programming/super_kernel/operator_adaptation.md)
	101	+ - [核函数（Kernel）直调算子额外适配说明](./advanced_programming/super_kernel/kernel_direct_call_adaptation.md)
	102	+ - [算子自验证说明](./advanced_programming/super_kernel/operator_self_verification.md)
	103	+ - [Aclnn算子工程化开发](./advanced_programming/aclnn_operator_development/aclnn_operator_development.md)
	104	+ - [概述](./advanced_programming/aclnn_operator_development/overview.md)
	105	+ - [快速入门](./advanced_programming/aclnn_operator_development/aclnn_quick_start.md)
	106	+ - [设计与实现](./advanced_programming/aclnn_operator_development/design_and_implementation/design_and_implementation.md)
	107	+ - [算子功能设计](./advanced_programming/aclnn_operator_development/design_and_implementation/operator_function_design.md)
	108	+ - [算子原型定义](./advanced_programming/aclnn_operator_development/design_and_implementation/operator_prototype_definition.md)
	109	+ - [Host侧Tiling实现](./advanced_programming/aclnn_operator_development/design_and_implementation/host_tiling_implementation.md)
	110	+ - [核函数（Kernel）侧算子实现](./advanced_programming/aclnn_operator_development/design_and_implementation/kernel_operator_implementation.md)
	111	+ - [多分支策略](./advanced_programming/aclnn_operator_development/design_and_implementation/multi_branch_strategy.md)
	112	+ - [GenSimplifiedKey回调实现](./advanced_programming/aclnn_operator_development/design_and_implementation/gen_simplified_key.md)
	113	+ - [通过TilingData传递属性信息](./advanced_programming/aclnn_operator_development/design_and_implementation/tiling_data_attributes.md)
	114	+ - [使用高阶API时配套的Tiling实现](./advanced_programming/aclnn_operator_development/design_and_implementation/tiling_with_advanced_api.md)
	115	+ - [核函数（Kernel）侧输出shape依赖计算](./advanced_programming/aclnn_operator_development/design_and_implementation/kernel_output_shape_computation.md)
	116	+ - [编译与部署](./advanced_programming/aclnn_operator_development/compilation_and_deployment/compilation_and_deployment.md)
	117	+ - [基本流程](./advanced_programming/aclnn_operator_development/compilation_and_deployment/basic_process.md)
	118	+ - [交叉编译](./advanced_programming/aclnn_operator_development/compilation_and_deployment/cross_compilation.md)
	119	+ - [多算子包组织](./advanced_programming/aclnn_operator_development/compilation_and_deployment/multi_operator_package.md)
	120	+ - [算子动态库和静态库编译](./advanced_programming/aclnn_operator_development/compilation_and_deployment/dynamic_static_lib_compilation.md)
	121	+ - [编译加速](./advanced_programming/aclnn_operator_development/compilation_and_deployment/compilation_acceleration.md)
	122	+ - [编译过程调试](./advanced_programming/aclnn_operator_development/compilation_and_deployment/compilation_debug.md)
	123	+ - [simplified key二进制配置JSON](./advanced_programming/aclnn_operator_development/compilation_and_deployment/simplified_key_json_configuration.md)
	124	+ - [调用](./advanced_programming/aclnn_operator_development/invocation/invocation.md)
	125	+ - [单算子API调用](./advanced_programming/aclnn_operator_development/invocation/single_operator_api_call.md)
	126	+ - [运行时加载机制](./advanced_programming/aclnn_operator_development/invocation/runtime_loading_mechanism.md)
	127	+ - [基于simplified key的运行时选择流程](./advanced_programming/aclnn_operator_development/invocation/simplified_key_runtime_flow.md)
	128	+ - [附录](./advanced_programming/aclnn_operator_development/appendix/appendix.md)
	129	+ - [aclnn工程化算子开发与核函数（Kernel）直调对比](./advanced_programming/aclnn_operator_development/appendix/kernel_direct_call_comparison.md)
	130	+ - [命名转换规则对照表](./advanced_programming/aclnn_operator_development/appendix/naming_conversion_table.md)
	131	+ - [CMake函数参考](./advanced_programming/aclnn_operator_development/appendix/cmake_function_reference.md)
	132	+ - [外部样例链接汇总](./advanced_programming/aclnn_operator_development/appendix/external_sample_links.md)
	133	+ - [算子入图开发](./advanced_programming/operator_graph_development/operator_graph_development.md)
	134	+ - [概述](./advanced_programming/operator_graph_development/overview.md)
	135	+ - [基本开发流程](./advanced_programming/operator_graph_development/basic_development_flow.md)
	136	+ - [开启Tiling下沉](./advanced_programming/operator_graph_development/enable_tiling_sink.md)
	137	+ - [图编译和图执行](./advanced_programming/operator_graph_development/graph_compilation_and_execution.md)
	138	+ - [AI框架算子适配](./advanced_programming/ai_framework_adaptation/ai_framework_adaptation.md)
	139	+ - [概述](./advanced_programming/ai_framework_adaptation/overview.md)
	140	+ - [PyTorch框架](./advanced_programming/ai_framework_adaptation/pytorch_framework.md)
	141	+ - [ONNX框架](./advanced_programming/ai_framework_adaptation/onnx_framework/onnx_framework.md)
	142	+ - [适配插件开发](./advanced_programming/ai_framework_adaptation/onnx_framework/adaptation_plugin_development.md)
	143	+ - [调用样例](./advanced_programming/ai_framework_adaptation/onnx_framework/call_examples.md)
	144	+ - [TensorFlow框架](./advanced_programming/ai_framework_adaptation/tensorflow_framework.md)
	145	+ - [SIMT协作组](./advanced_programming/simt_cooperative_groups.md)
	146	+ - [AOT编译优化](./advanced_programming/aot_compilation_optimization.md)
	147	+- [附录](./appendix/appendix.md)
	148	+ - [show_kernel_debug_data工具](./appendix/show_kernel_debug_data_tool.md)
	149	+ - [msobjdump工具](./appendix/msobjdump_tool.md)
	150	+ - [optype_collector工具](./appendix/optype_collector_tool.md)
	151	+ - [基于样例工程完成核函数（Kernel）直调](./appendix/kernel_direct_call_from_sample.md)
	152	+ - [常用操作](./appendix/common_operations/common_operations.md)
	153	+ - [如何开发动态输入算子](./appendix/common_operations/develop_dynamic_input_operator.md)
	154	+ - [如何在矢量编程时启用Vector Core](./appendix/common_operations/enable_vector_core.md)
	155	+ - [如何使用workspace](./appendix/common_operations/how_to_use_workspace.md)
	156	+ - [如何进行Tiling调测](./appendix/common_operations/tiling_debug.md)
	157	+ - [如何使用Tensor原地操作提升算子性能](./appendix/common_operations/tensor_inplace_performance.md)
	158	+ - [FAQ](./appendix/faq/faq.md)
	159	+ - [核函数（Kernel）运行验证时算子存在精度问题](./appendix/faq/kernel_precision_issue.md)
	160	+ - [运行验证时AllocTensor/FreeTensor失败](./appendix/faq/alloc_free_tensor_failure.md)
	161	+ - [kernel侧获取Tiling信息不正确](./appendix/faq/kernel_tiling_info_incorrect.md)
	162	+ - [核函数（Kernel）编译时报错"error: out of jump/jumpc imm range"](./appendix/faq/kernel_compile_jump_range_error.md)
	163	+ - [含有Matmul高阶API的算子精度问题](./appendix/faq/matmul_advanced_api_precision.md)
	164	+ - [算子工程编译时出现文件名过长报错](./appendix/faq/filename_too_long_error.md)
	165	+ - [调用算子时出现无法打开config.ini的报错](./appendix/faq/config_ini_open_error.md)
	166	+ - [算子包部署时出现权限不足报错](./appendix/faq/operator_deploy_permission_error.md)

Mdocs/zh/guide/programming_guide/programming_model/ai_core_simd_programming/ai_core_simd_programming.md+18-0

	@@ -1 +1,19 @@
1	# AI-Core-SIMD编程	1	# AI-Core-SIMD编程
		2	+
		3	+- [概述](./overview.md)
		4	+- [抽象硬件架构](./abstract_hardware_architecture.md)
		5	+- [核函数（Kernel）](./kernel_function.md)
		6	+- [基于指针的C语言编程](./c_pointer_programming/c_pointer_programming.md)
		7	+ - [C语言编程概述](./c_pointer_programming/c_programming_overview.md)
		8	+ - [Memory矢量计算编程](./c_pointer_programming/memory_vector_computation.md)
		9	+ - [Reg矢量计算编程](./c_pointer_programming/reg_vector_computation.md)
		10	+ - [Cube矩阵计算编程](./c_pointer_programming/cube_matrix_computation.md)
		11	+- [基于Tensor的C++编程](./cpp_tensor_programming/cpp_tensor_programming.md)
		12	+ - [C++ Tensor编程概述](./cpp_tensor_programming/cpp_tensor_programming_overview.md)
		13	+ - [Memory矢量计算编程](./cpp_tensor_programming/memory_vector_computation.md)
		14	+ - [Reg矢量计算编程](./cpp_tensor_programming/reg_vector_computation.md)
		15	+ - [Cube矩阵计算编程](./cpp_tensor_programming/cube_matrix_computation.md)
		16	+ - [静态Tensor编程](./cpp_tensor_programming/static_tensor_programming.md)
		17	+- [基于TPipe和TQue编程](./tpipe_tque_programming/tpipe_tque_programming.md)
		18	+ - [TPipe-TQue框架编程原理](./tpipe_tque_programming/tpipe_tque_principles.md)
		19	+ - [TPipe-TQue框架编程范式](./tpipe_tque_programming/tpipe_tque_paradigm.md)

Adocs/zh/guide/programming_guide/programming_model/ai_core_simd_programming/c_pointer_programming/c_pointer_programming.md+6-0

@@ -0,0 +1,6 @@
	1	+# 基于指针的C语言编程
	2	+
	3	+- [C语言编程概述](./c_programming_overview.md)
	4	+- [Memory矢量计算编程](./memory_vector_computation.md)
	5	+- [Reg矢量计算编程](./reg_vector_computation.md)
	6	+- [Cube矩阵计算编程](./cube_matrix_computation.md)

Adocs/zh/guide/programming_guide/programming_model/ai_core_simd_programming/cpp_tensor_programming/cpp_tensor_programming.md+7-0

@@ -0,0 +1,7 @@
	1	+# 基于Tensor的C++编程
	2	+
	3	+- [C++ Tensor编程概述](./cpp_tensor_programming_overview.md)
	4	+- [Memory矢量计算编程](./memory_vector_computation.md)
	5	+- [Reg矢量计算编程](./reg_vector_computation.md)
	6	+- [Cube矩阵计算编程](./cube_matrix_computation.md)
	7	+- [静态Tensor编程](./static_tensor_programming.md)

Adocs/zh/guide/programming_guide/programming_model/ai_core_simd_programming/tpipe_tque_programming/tpipe_tque_programming.md+4-0

@@ -0,0 +1,4 @@
	1	+# 基于TPipe和TQue编程
	2	+
	3	+- [TPipe-TQue框架编程原理](./tpipe_tque_principles.md)
	4	+- [TPipe-TQue框架编程范式](./tpipe_tque_paradigm.md)

Mdocs/zh/guide/programming_guide/programming_model/ai_core_simt_programming/ai_core_simt_programming.md+8-1

	@@ -1,3 +1,10 @@
1	# AI-Core-SIMT编程	1	# AI-Core-SIMT编程
2		2
3	-本文件为文档发布时所需的索引文件。	3	+- [概述](./overview.md)
		4	+- [抽象硬件架构](./abstract_hardware_architecture.md)
		5	+- [线程架构](./thread_architecture.md)
		6	+- [内存层级](./memory_hierarchy.md)
		7	+- [核函数（Kernel）](./kernel_function.md)
		8	+- [同步机制](./synchronization.md)
		9	+- [原子操作](./atomic_operations.md)
		10	+- [编程示例](./programming_examples.md)

Mdocs/zh/guide/programming_guide/programming_model/programming_model.md+30-1

	@@ -1,3 +1,32 @@
1	# 编程模型	1	# 编程模型
2		2
3	-本文件为文档发布时所需的索引文件。	3	+- [异构系统](./heterogeneous_system.md)
		4	+- [编程模型概述](./programming_model_overview.md)
		5	+- [AI Core SIMD编程](./ai_core_simd_programming/ai_core_simd_programming.md)
		6	+ - [概述](./ai_core_simd_programming/overview.md)
		7	+ - [抽象硬件架构](./ai_core_simd_programming/abstract_hardware_architecture.md)
		8	+ - [核函数（Kernel）](./ai_core_simd_programming/kernel_function.md)
		9	+ - [基于指针的C语言编程](./ai_core_simd_programming/c_pointer_programming/c_pointer_programming.md)
		10	+ - [C语言编程概述](./ai_core_simd_programming/c_pointer_programming/c_programming_overview.md)
		11	+ - [Memory矢量计算编程](./ai_core_simd_programming/c_pointer_programming/memory_vector_computation.md)
		12	+ - [Reg矢量计算编程](./ai_core_simd_programming/c_pointer_programming/reg_vector_computation.md)
		13	+ - [Cube矩阵计算编程](./ai_core_simd_programming/c_pointer_programming/cube_matrix_computation.md)
		14	+ - [基于Tensor的C++编程](./ai_core_simd_programming/cpp_tensor_programming/cpp_tensor_programming.md)
		15	+ - [C++ Tensor编程概述](./ai_core_simd_programming/cpp_tensor_programming/cpp_tensor_programming_overview.md)
		16	+ - [Memory矢量计算编程](./ai_core_simd_programming/cpp_tensor_programming/memory_vector_computation.md)
		17	+ - [Reg矢量计算编程](./ai_core_simd_programming/cpp_tensor_programming/reg_vector_computation.md)
		18	+ - [Cube矩阵计算编程](./ai_core_simd_programming/cpp_tensor_programming/cube_matrix_computation.md)
		19	+ - [静态Tensor编程](./ai_core_simd_programming/cpp_tensor_programming/static_tensor_programming.md)
		20	+ - [基于TPipe和TQue编程](./ai_core_simd_programming/tpipe_tque_programming/tpipe_tque_programming.md)
		21	+ - [TPipe-TQue框架编程原理](./ai_core_simd_programming/tpipe_tque_programming/tpipe_tque_principles.md)
		22	+ - [TPipe-TQue框架编程范式](./ai_core_simd_programming/tpipe_tque_programming/tpipe_tque_paradigm.md)
		23	+- [AI Core SIMT编程](./ai_core_simt_programming/ai_core_simt_programming.md)
		24	+ - [概述](./ai_core_simt_programming/overview.md)
		25	+ - [抽象硬件架构](./ai_core_simt_programming/abstract_hardware_architecture.md)
		26	+ - [线程架构](./ai_core_simt_programming/thread_architecture.md)
		27	+ - [内存层级](./ai_core_simt_programming/memory_hierarchy.md)
		28	+ - [核函数（Kernel）](./ai_core_simt_programming/kernel_function.md)
		29	+ - [同步机制](./ai_core_simt_programming/synchronization.md)
		30	+ - [原子操作](./ai_core_simt_programming/atomic_operations.md)
		31	+ - [编程示例](./ai_core_simt_programming/programming_examples.md)
		32	+- [AI CPU编程](./ai_cpu_programming.md)

Adocs/zh/guide/technical_appendix/concepts_and_terms/concepts_and_terms.md+10-0

@@ -0,0 +1,10 @@
	1	+# 概念原理和术语
	2	+
	3	+- [术语表](./glossary.md)
	4	+- [神经网络和算子](./neural_networks_and_operators/neural_networks_and_operators.md)
	5	+ - [算子基本概念](./neural_networks_and_operators/operator_basics.md)
	6	+ - [数据排布格式](./neural_networks_and_operators/data_layout.md)
	7	+- [内存访问原理](./memory_access/memory_access.md)
	8	+ - [Scalar读写数据](./memory_access/scalar_read_write.md)
	9	+- [性能优化技术原理](./performance_optimization/performance_optimization.md)
	10	+ - [Double Buffer](./performance_optimization/double_buffer.md)

Adocs/zh/guide/technical_appendix/concepts_and_terms/memory_access/memory_access.md+3-0

@@ -0,0 +1,3 @@
	1	+# 内存访问原理
	2	+
	3	+- [Scalar读写数据](./scalar_read_write.md)

Adocs/zh/guide/technical_appendix/concepts_and_terms/neural_networks_and_operators/neural_networks_and_operators.md+4-0

@@ -0,0 +1,4 @@
	1	+# 神经网络和算子
	2	+
	3	+- [算子基本概念](./operator_basics.md)
	4	+- [数据排布格式](./data_layout.md)

Adocs/zh/guide/technical_appendix/concepts_and_terms/performance_optimization/performance_optimization.md+3-0

@@ -0,0 +1,3 @@
	1	+# 性能优化技术原理
	2	+
	3	+- [Double Buffer](./double_buffer.md)

Adocs/zh/guide/technical_appendix/cpp_standard_support/cpp_standard_support.md+6-0

@@ -0,0 +1,6 @@
	1	+# C++标准支持
	2	+
	3	+- [概述](./overview.md)
	4	+- [Device侧C/C++语法限制](./syntax_restrictions/syntax_restrictions.md)
	5	+ - [特性](./syntax_restrictions/features.md)
	6	+ - [函数](./syntax_restrictions/functions.md)

Mdocs/zh/guide/technical_appendix/cpp_standard_support/syntax_restrictions/syntax_restrictions.md+3-0

	@@ -1 +1,4 @@
1	# 语法限制	1	# 语法限制
		2	+
		3	+- [特性](./features.md)
		4	+- [函数](./functions.md)

Adocs/zh/guide/technical_appendix/technical_appendix.md+16-0

@@ -0,0 +1,16 @@
	1	+# 技术附录
	2	+
	3	+- [概念原理和术语](./concepts_and_terms/concepts_and_terms.md)
	4	+ - [术语表](./concepts_and_terms/glossary.md)
	5	+ - [神经网络和算子](./concepts_and_terms/neural_networks_and_operators/neural_networks_and_operators.md)
	6	+ - [算子基本概念](./concepts_and_terms/neural_networks_and_operators/operator_basics.md)
	7	+ - [数据排布格式](./concepts_and_terms/neural_networks_and_operators/data_layout.md)
	8	+ - [内存访问原理](./concepts_and_terms/memory_access/memory_access.md)
	9	+ - [Scalar读写数据](./concepts_and_terms/memory_access/scalar_read_write.md)
	10	+ - [性能优化技术原理](./concepts_and_terms/performance_optimization/performance_optimization.md)
	11	+ - [Double Buffer](./concepts_and_terms/performance_optimization/double_buffer.md)
	12	+- [C++标准支持](./cpp_standard_support/cpp_standard_support.md)
	13	+ - [概述](./cpp_standard_support/overview.md)
	14	+ - [Device侧C/C++语法限制](./cpp_standard_support/syntax_restrictions/syntax_restrictions.md)
	15	+ - [特性](./cpp_standard_support/syntax_restrictions/features.md)
	16	+ - [函数](./cpp_standard_support/syntax_restrictions/functions.md)

	@@ -1,9 +1,9 @@
1	# Ascend C算子开发指南	1	# Ascend C算子开发指南
2		2
3	-- [入门教程](getting_started/ascend_c_overview_and_learning_path.md)	3	+- [入门教程](getting_started/getting_started.md)
4	- [Ascend C概述与学习路径](getting_started/ascend_c_overview_and_learning_path.md)	4	- [Ascend C概述与学习路径](getting_started/ascend_c_overview_and_learning_path.md)
5	- [环境准备](getting_started/environment_setup.md)	5	- [环境准备](getting_started/environment_setup.md)
6	- - [快速入门](getting_started/quick_start/heterogeneous_system_and_programming_model.md)	6	+ - [快速入门](getting_started/quick_start/quick_start.md)
7	- [异构系统与编程模型](getting_started/quick_start/heterogeneous_system_and_programming_model.md)	7	- [异构系统与编程模型](getting_started/quick_start/heterogeneous_system_and_programming_model.md)
8	- [基于SIMD编程](getting_started/quick_start/simd_programming/simd_programming.md)	8	- [基于SIMD编程](getting_started/quick_start/simd_programming/simd_programming.md)
9	- [HelloWorld](getting_started/quick_start/simd_programming/hello_world.md)	9	- [HelloWorld](getting_started/quick_start/simd_programming/hello_world.md)
	@@ -13,7 +13,7 @@
13	- [HelloWorld](getting_started/quick_start/simt_programming/hello_world.md)	13	- [HelloWorld](getting_started/quick_start/simt_programming/hello_world.md)
14	- [Gather算子快速入门](getting_started/quick_start/simt_programming/gather_op_quick_start.md)	14	- [Gather算子快速入门](getting_started/quick_start/simt_programming/gather_op_quick_start.md)
15		15
16	-- [编程指南](programming_guide/document_structure.md)	16	+- [编程指南](programming_guide/programming_guide.md)
17	- [本文档组织结构](programming_guide/document_structure.md)	17	- [本文档组织结构](programming_guide/document_structure.md)
18	- [编程模型](programming_guide/programming_model/programming_model.md)	18	- [编程模型](programming_guide/programming_model/programming_model.md)
19	- [异构系统](programming_guide/programming_model/heterogeneous_system.md)	19	- [异构系统](programming_guide/programming_model/heterogeneous_system.md)
	@@ -22,18 +22,18 @@
22	- [概述](programming_guide/programming_model/ai_core_simd_programming/overview.md)	22	- [概述](programming_guide/programming_model/ai_core_simd_programming/overview.md)
23	- [抽象硬件架构](programming_guide/programming_model/ai_core_simd_programming/abstract_hardware_architecture.md)	23	- [抽象硬件架构](programming_guide/programming_model/ai_core_simd_programming/abstract_hardware_architecture.md)
24	- [核函数（Kernel）](programming_guide/programming_model/ai_core_simd_programming/kernel_function.md)	24	- [核函数（Kernel）](programming_guide/programming_model/ai_core_simd_programming/kernel_function.md)
25	- - [基于指针的C语言编程](programming_guide/programming_model/ai_core_simd_programming/c_pointer_programming/c_programming_overview.md)	25	+ - [基于指针的C语言编程](programming_guide/programming_model/ai_core_simd_programming/c_pointer_programming/c_pointer_programming.md)
26	- [C语言编程概述](programming_guide/programming_model/ai_core_simd_programming/c_pointer_programming/c_programming_overview.md)	26	- [C语言编程概述](programming_guide/programming_model/ai_core_simd_programming/c_pointer_programming/c_programming_overview.md)
27	- [Memory矢量计算编程](programming_guide/programming_model/ai_core_simd_programming/c_pointer_programming/memory_vector_computation.md)	27	- [Memory矢量计算编程](programming_guide/programming_model/ai_core_simd_programming/c_pointer_programming/memory_vector_computation.md)
28	- [Reg矢量计算编程](programming_guide/programming_model/ai_core_simd_programming/c_pointer_programming/reg_vector_computation.md)	28	- [Reg矢量计算编程](programming_guide/programming_model/ai_core_simd_programming/c_pointer_programming/reg_vector_computation.md)
29	- [Cube矩阵计算编程](programming_guide/programming_model/ai_core_simd_programming/c_pointer_programming/cube_matrix_computation.md)	29	- [Cube矩阵计算编程](programming_guide/programming_model/ai_core_simd_programming/c_pointer_programming/cube_matrix_computation.md)
30	- - [基于Tensor的C++编程](programming_guide/programming_model/ai_core_simd_programming/cpp_tensor_programming/cpp_tensor_programming_overview.md)	30	+ - [基于Tensor的C++编程](programming_guide/programming_model/ai_core_simd_programming/cpp_tensor_programming/cpp_tensor_programming.md)
31	- [C++ Tensor编程概述](programming_guide/programming_model/ai_core_simd_programming/cpp_tensor_programming/cpp_tensor_programming_overview.md)	31	- [C++ Tensor编程概述](programming_guide/programming_model/ai_core_simd_programming/cpp_tensor_programming/cpp_tensor_programming_overview.md)
32	- [Memory矢量计算编程](programming_guide/programming_model/ai_core_simd_programming/cpp_tensor_programming/memory_vector_computation.md)	32	- [Memory矢量计算编程](programming_guide/programming_model/ai_core_simd_programming/cpp_tensor_programming/memory_vector_computation.md)
33	- [Reg矢量计算编程](programming_guide/programming_model/ai_core_simd_programming/cpp_tensor_programming/reg_vector_computation.md)	33	- [Reg矢量计算编程](programming_guide/programming_model/ai_core_simd_programming/cpp_tensor_programming/reg_vector_computation.md)
34	- [Cube矩阵计算编程](programming_guide/programming_model/ai_core_simd_programming/cpp_tensor_programming/cube_matrix_computation.md)	34	- [Cube矩阵计算编程](programming_guide/programming_model/ai_core_simd_programming/cpp_tensor_programming/cube_matrix_computation.md)
35	- [静态Tensor编程](programming_guide/programming_model/ai_core_simd_programming/cpp_tensor_programming/static_tensor_programming.md)	35	- [静态Tensor编程](programming_guide/programming_model/ai_core_simd_programming/cpp_tensor_programming/static_tensor_programming.md)
36	- - [基于TPipe和TQue编程](programming_guide/programming_model/ai_core_simd_programming/tpipe_tque_programming/tpipe_tque_principles.md)	36	+ - [基于TPipe和TQue编程](programming_guide/programming_model/ai_core_simd_programming/tpipe_tque_programming/tpipe_tque_programming.md)
37	- [TPipe-TQue框架编程原理](programming_guide/programming_model/ai_core_simd_programming/tpipe_tque_programming/tpipe_tque_principles.md)	37	- [TPipe-TQue框架编程原理](programming_guide/programming_model/ai_core_simd_programming/tpipe_tque_programming/tpipe_tque_principles.md)
38	- [TPipe-TQue框架编程范式](programming_guide/programming_model/ai_core_simd_programming/tpipe_tque_programming/tpipe_tque_paradigm.md)	38	- [TPipe-TQue框架编程范式](programming_guide/programming_model/ai_core_simd_programming/tpipe_tque_programming/tpipe_tque_paradigm.md)
39	- [AI Core SIMT编程](programming_guide/programming_model/ai_core_simt_programming/ai_core_simt_programming.md)	39	- [AI Core SIMT编程](programming_guide/programming_model/ai_core_simt_programming/ai_core_simt_programming.md)
	@@ -50,7 +50,7 @@
50		50
51	- [编译与运行](programming_guide/compilation_and_execution/compilation_and_execution.md)	51	- [编译与运行](programming_guide/compilation_and_execution/compilation_and_execution.md)