learning-path.md

layout	default
title	Learning Path
nav_order	3
permalink	/docs/learning-path

Learning Path

{: .fs-8 }

Follow the optimization ladder in the order the repository was designed to teach it {: .fs-6 .fw-300 }

---

Recommended order

Step	Kernel	Why it comes here
1	Naive	Establish the baseline cost model
2	Tiled	Introduce shared-memory reuse
3	Bank-Free	Show why shared-memory layout still matters
4	Double Buffer	Add staging and overlap concepts
5	Tensor Core	Move to WMMA and mixed-precision hardware

---

What each stage teaches

Naive -> Tiled

• Thread/block mapping
• Memory coalescing
• Shared-memory reuse

Tiled -> Bank-Free

• 32-bank shared-memory behavior
• Why [32][33] matters

Bank-Free -> Double Buffer

• Pipeline thinking
• Tile staging and latency hiding

Double Buffer -> Tensor Core

• WMMA fragments
• Mixed precision
• Safe fallback behavior for unsupported shapes

---

Suggested reading rhythm

1. Build and run the project first
2. Read the kernel page for one stage
3. Run the benchmark again
4. Compare the code with the previous stage
5. Move to the next optimization only after the current one is clear

---

Before you start

• Make sure your environment follows Getting Started
• Use the Architecture page if you want the repository-level map first
• Keep the Specifications Index nearby if you want the normative requirements