Skip to content

Ascend/msagent / skills/operator/ops-performance-tuning

ops-performance-tuning

Verified

Compile, profile, diagnose, optimize, and compare Ascend NPU operators across CANN versions and A2/A3/A5 for Ascend C, CATLASS, Triton-Ascend, TileLang-Ascend, PyPTO, and SHMEM/MC2. Use when an operator already has a runnable implementation and the user asks for msOpProf collection, bottleneck analysis, source-level tuning, or a reproducible before/after performance report. Do not use as the primary workflow for operator creation, migration, or unresolved correctness failures.

FileKindSize
SKILL.mdskill md15.4 KB
evals/evals.jsonconfig3.6 KB
references/case-routing.mdmarkdown30.0 KB
references/cases/ascendc/pr_ascendc_allgathermm_scalar_align.mdmarkdown1.7 KB
references/cases/ascendc/pr_ascendc_cann_samples_perf_story.mdmarkdown3.9 KB
references/cases/ascendc/pr_ascendc_cgdr_vf_inverse.mdmarkdown1.6 KB
references/cases/ascendc/pr_ascendc_conv3ddx_tque_tbuf.mdmarkdown1.6 KB
references/cases/ascendc/pr_ascendc_layernorm_areg.mdmarkdown1.6 KB
references/cases/ascendc/pr_ascendc_logit_a5_microapi.mdmarkdown1.9 KB
references/cases/ascendc/pr_ascendc_matmul_case_18x.mdmarkdown1.5 KB
references/cases/ascendc/pr_ascendc_matmul_full_load_wiki.mdmarkdown1.6 KB
references/cases/ascendc/pr_ascendc_matmul_vcv_wiki.mdmarkdown1.8 KB
references/cases/ascendc/pr_ascendc_mc2_fusion.mdmarkdown1.4 KB
references/cases/ascendc/pr_ascendc_mhc_24x.mdmarkdown1.2 KB
references/cases/ascendc/pr_ascendc_msopprof_add_block_imbalance.mdmarkdown1.7 KB
references/cases/ascendc/pr_ascendc_quantize_a2_micro_opt.mdmarkdown2.0 KB
references/cases/ascendc/pr_ascendc_rotate_quant_blaze.mdmarkdown2.8 KB
references/cases/ascendc/pr_ascendc_tiling_l2_official.mdmarkdown1.6 KB
references/cases/catlass/pr_catlass_baseline.mdmarkdown1.3 KB
references/cases/catlass/pr_catlass_fixpipe_dualdst.mdmarkdown1.6 KB
references/cases/catlass/pr_catlass_grouped_matmul_tiling.mdmarkdown1.5 KB
references/cases/catlass/pr_catlass_mstuner_3p8x.mdmarkdown1.9 KB
references/cases/catlass/pr_catlass_swizzle_balance.mdmarkdown1.6 KB
references/cases/pypto/pr_pypto_gqa_decode_vector_merge.mdmarkdown2.4 KB
references/cases/pypto/pr_pypto_l2_hit_layout_s13.mdmarkdown1.9 KB
references/cases/pypto/pr_pypto_per_matmul_tile_shapes.mdmarkdown2.4 KB
references/cases/pypto/pr_pypto_small_shape_matmul.mdmarkdown2.3 KB
references/cases/pypto/pr_pypto_weight_none_cacheable.mdmarkdown2.6 KB
references/cases/shmem/pr_shmem_alltoallv_mr152.mdmarkdown2.4 KB
references/cases/shmem/pr_shmem_reduce_scatter_serial.mdmarkdown2.6 KB
references/cases/tilelang/pr_tilelang_benchmark_vs_ascendc.mdmarkdown1.7 KB
references/cases/tilelang/pr_tilelang_compiler_prs.mdmarkdown2.3 KB
references/cases/tilelang/pr_tilelang_fa_pr698.mdmarkdown1.5 KB
references/cases/tilelang/pr_tilelang_gqa_bwd_pr1494.mdmarkdown1.3 KB
references/cases/tilelang/pr_tilelang_persistent_mqa_decode.mdmarkdown2.2 KB
references/cases/tilelang/pr_tilelang_sfa_pr665.mdmarkdown914 B
references/cases/triton/pr_triton_atomic_simd_24x.mdmarkdown1.8 KB
references/cases/triton/pr_triton_autotune_background.mdmarkdown1.0 KB
references/cases/triton/pr_triton_constexpr_recompile.mdmarkdown1.1 KB
references/cases/triton/pr_triton_lgamma_simt.mdmarkdown1.6 KB
references/compile/compile-ascendc.mdmarkdown9.0 KB
references/compile/compile-catlass.mdmarkdown2.4 KB
references/compile/compile-pypto.mdmarkdown2.0 KB
references/compile/compile-shmem.mdmarkdown2.6 KB
references/compile/compile-tilelang.mdmarkdown2.7 KB
references/compile/compile-triton.mdmarkdown4.4 KB
references/diagnose/diagnose-l2-cache.mdmarkdown2.2 KB
references/diagnose/diagnose-occupancy.mdmarkdown4.0 KB
references/diagnose/diagnose-pipeline.mdmarkdown4.1 KB
references/diagnose/diagnose-roofline.mdmarkdown3.1 KB
references/optimize/optimize-api-usage.mdmarkdown1.7 KB
references/optimize/optimize-ascendc-regbase-vf.mdmarkdown5.9 KB
references/optimize/optimize-ascendc-tiling.mdmarkdown935 B
references/optimize/optimize-catlass.mdmarkdown3.0 KB
references/optimize/optimize-data-copy.mdmarkdown870 B
references/optimize/optimize-index.mdmarkdown3.6 KB
references/optimize/optimize-memory-hierarchy.mdmarkdown599 B
references/optimize/optimize-pipeline.mdmarkdown1.0 KB
references/optimize/optimize-pypto.mdmarkdown1.5 KB
references/optimize/optimize-shmem.mdmarkdown5.1 KB
references/optimize/optimize-tilelang.mdmarkdown1.7 KB
references/optimize/optimize-triton.mdmarkdown3.8 KB
references/profile/benchmark-harness.mdmarkdown2.2 KB
references/profile/profile-msopprof.mdmarkdown7.4 KB
references/troubleshooting.mdmarkdown5.8 KB
scripts/bound_analyzer.pyscript9.0 KB
scripts/delta_report.pyscript4.8 KB
scripts/env_check.shscript5.0 KB
scripts/result_saver.pyscript5.0 KB

references/compile/compile-tilelang.md

View on GitHub
# TileLang-Ascend 算子工程:编译、运行、验证、单测方法

> 本文档目的:整理 tilelang-ascend 框架的算子工程方法(简介与仓库地址、依赖版本、编译命令、运行/验证方法、单测方式)。多框架总览与环境速查见 [算子类型路由](../case-routing.md)。

## 3. tilelang-ascend

### 3.1 简介与仓库

基于 TVM 的 Python tile DSL 的昇腾后端,两条路线:

- Ascend C & PTO 后端:<https://github.com/tile-ai/tilelang-ascend>(ascendc_pto 分支;GitCode 镜像 <https://gitcode.com/runnee-wwl/tilelang-ascend>);
- AscendNPU IR (MLIR) 后端:<https://github.com/tile-ai/tilelang-mlir-ascend>(npuir 路线)。
- 上游 tile-ai/tilelang 已于 2025-09-29 宣布支持 AscendC 与 Ascend NPU IR 两条后端分支(PyPI 页面 Latest News:<https://pypi.org/project/tilelang/>)。

### 3.2 依赖与后端选择

先确定 `ascendc_pto` 或 `npuir` 路线,再读取该分支的 CANN、torch_npu、Python 和硬件兼容要求。两条路线的工具链不能混用;历史最低版本不等同于当前分支完整兼容关系。

```bash
git branch --show-current && git rev-parse HEAD
rg -n "CANN|torch.npu|torch_npu|Python|ascendc_pto|npuir" README* docs/ 2>/dev/null
python3 -c 'import torch_npu; print(torch_npu.__version__)'
```

### 3.3 安装(编译)

```bash
export ASCEND_HOME_PATH=<target-cann-path>/ascend-toolkit/latest
# 方式一:预构建 wheel
pip install tilelang-*.whl
# 方式二:源码构建 wheel
git clone --recursive https://github.com/tile-ai/tilelang-ascend.git && cd tilelang-ascend
./build_wheel_ascend.sh [--enable-llvm]
pip install dist/tilelang-*.whl
# 方式三:源码直接安装
bash install_ascend.sh && source set_env.sh
# npuir 路线:bash install_npuir.sh
```

### 3.4 运行/验证方法

```bash
cd examples/gemm
python example_gemm.py    # 成功打印 "Kernel Output Match!"
```

kernel 用 Python DSL + `@tilelang.jit` 编写;`generate_source()` 可 lower 成 Ascend C 源码(xLLM 等项目即按此集成,见 <https://docs.xllm-ai.com/zh/dev_guide/tilelang_ascend_kernel_dev/>)。

### 3.5 模式与编程模型

- 环境变量 `TILELANG_ASCEND_MODE`:`Developer`(自动 C/V 作用域分离、自动插同步)/ `Expert`(手动 `T.Scope("C")/T.Scope("V")` + `T.set_cross_flag/wait_cross_flag`)。
- 向量化两种风格:`T.Parallel(M, N)` 自动向量化;`T.tile.add(...)` 显式 tile 原语。
- 软件流水 `T.Pipelined`(2026-01-15 引入);片上内存 `alloc_L1/alloc_ub/...`;`T.printf`/`T.dump_tensor` 调试。

### 3.6 单测方式

examples 下 `test_*.py`(精度 + `--level perf` 性能档,见 PR #1494);CI 有 ci_performance.py。
planning-and-task-breakdownaddyosmani/agent-skillsBreaks work into ordered tasks. Use when you have a spec or clear requirements and need to break work into implementable tasks. Use when a task feels too large to start, when you need to estimate scope, or when parallel work is possible.Stars: 119kUpdated Claude CodeCodexVerifiedInstall
screenpipethedotmack/claude-memSearch screen and audio history recorded by Screenpipe and optionally save selected context into claude-mem. Use when the user asks to recall desktop activity, a meeting, or something seen in another app through Screenpipe.Stars: 118kUpdated Claude CodeCodexVerifiedInstall
contact-widgetnexu-io/open-designSelf-contained floating chat widget with welcome screen, social links, meeting button, and message input. Single HTML file, zero dependencies.Stars: 108kUpdated Claude CodeCodexVerifiedInstall
meeting-notesnexu-io/open-designMeeting notes page — title bar with attendees, agenda checklist, decisions block, action items table with owners + dates, and a "next meeting" footer. Use when the brief mentions "meeting notes", "minutes", "1:1 notes", "all-hands recap", or "会议纪要".Stars: 108kUpdated Claude CodeCodexVerifiedInstall