Skip to content

kali20gakki/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.pyscript9.8 KB
scripts/delta_report.pyscript5.5 KB
scripts/env_check.shscript5.7 KB
scripts/result_saver.pyscript5.8 KB

references/profile/benchmark-harness.md

View on GitHub
# 工程无 benchmark 时的计时 harness 模板

> SKILL Step 3 要求固定 warmup/repeat 与设备 event 计时,但 cann-samples 类样例普遍只有"单次 launch + 精度 PASS"。本文档给出合规的最小补齐方式。

## 归类原则:测量设施 ≠ 优化变量

- 在**产物目录的工程副本**上加计时,不直接改用户基线源码;改动前先把原始源码备份到 `baseline/`。
- harness 改动(aclrtEvent 创建/记录、warmup 循环、打印耗时)属于测量设施:前后两轮使用**同一份 harness**,不得把 harness 改动写进优化 diff。
- kernel 源码一行不动;若必须改(如 SHMEMI_PROF 插桩、phase skip 位),单独归类为"诊断插桩",同样不算优化变量,并在报告中注明。

## aclrtEvent 模板(host 侧)

```cpp
// 在 kernel launch 调用处包一层;WARMUP/REPEAT 编译期或环境变量固定,前后一致
aclrtEvent start, stop;
aclrtCreateEvent(&start);
aclrtCreateEvent(&stop);
for (int i = 0; i < WARMUP; ++i) { /* kernel launch */ }
aclrtSynchronizeStream(stream);
aclrtRecordEvent(start, stream);
for (int i = 0; i < REPEAT; ++i) { /* kernel launch */ }
aclrtRecordEvent(stop, stream);
aclrtSynchronizeStream(stream);
float ms = 0.f;
aclrtEventElapsedTime(&ms, start, stop);
printf("[perf] kernel avg: %.2f us over %d iters (warmup %d)\n",
       ms * 1000.0f / REPEAT, REPEAT, WARMUP);
```

- warmup 剔除冷跑(首跑含初始化/缓存预热,实测可差 25%);稳态多轮取中位。
- 通信/多 rank 算子:跨 rank 同步后再计时,报告每 rank 最大值与均值(通信性能由最慢 rank 决定);e2e wall 含进程派生与 SHMEM 握手,与 kernel 耗时分开记录。
- 进程级复跑(≥3 次)消除单次噪声;run 间噪声 >5% 时,小于噪声的收益不得归因。

## 何时允许用 profiler Task Duration 作基线

- kernel 为 ms 级长耗时、工程无 event 设施、且插桩开销占比可忽略时:允许(`result_saver.py --timing-method msprof_task_duration`),报告注明口径。
- µs 级短 kernel:禁止——插桩耗时不能替代 event;用"stream 内多 launch 单 event 对"放大可测时长,或以 device duration 作辅助口径并注明局限。
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