Analyze Kernel Bottleneck
- `pipeline-gpu-kernel` -- implement software pipelining with cp.async when analysis identifies a memory-bound kernel with low compute/load ratio
LLM Evaluation
Evaluated by: xiaomi/mimo-v2-flash:free
Last evaluated: May 17, 2026
Prompt Preview
---
name: analyze-kernel-bottleneck
description: >
Systematically identify whether a GPU kernel is compute-bound, memory-bound,
or latency-bound using roofline analysis, occupancy calculations, compute/load
ratio per tile, and SASS instruction inspection. Produces a decision matrix
for optimization strategy selection (cp.async, warp interleaving, tiling,
double-buffering, or CuAssembler hand-tuning).
license: MIT
allowed-tools: Read Grep Glob Bash
metadata:
author: Philipp Thoss
ve...
Full prompt length: 17023 characters