xref: /llvm-project/mlir/docs/Tutorials/Toy/_index.md (revision 31d1ae79752da75503a4f1deb3ef2fd0b7868682)
1# Toy Tutorial
2
3This tutorial runs through the implementation of a basic toy language on top of
4MLIR. The goal of this tutorial is to introduce the concepts of MLIR; in
5particular, how [dialects](../../LangRef.md/#dialects) can help easily support
6language specific constructs and transformations while still offering an easy
7path to lower to LLVM or other codegen infrastructure. This tutorial is based on
8the model of the
9[LLVM Kaleidoscope Tutorial](https://llvm.org/docs/tutorial/MyFirstLanguageFrontend/index.html).
10
11Another good source of introduction is the online [recording](https://www.youtube.com/watch?v=Y4SvqTtOIDk)
12from the 2020 LLVM Dev Conference ([slides](https://llvm.org/devmtg/2020-09/slides/MLIR_Tutorial.pdf)).
13
14This tutorial assumes you have cloned and built MLIR; if you have not yet done
15so, see
16[Getting started with MLIR](../../../getting_started/).
17
18This tutorial is divided in the following chapters:
19
20-   [Chapter #1](Ch-1.md): Introduction to the Toy language and the definition
21    of its AST.
22-   [Chapter #2](Ch-2.md): Traversing the AST to emit a dialect in MLIR,
23    introducing base MLIR concepts. Here we show how to start attaching
24    semantics to our custom operations in MLIR.
25-   [Chapter #3](Ch-3.md): High-level language-specific optimization using
26    pattern rewriting system.
27-   [Chapter #4](Ch-4.md): Writing generic dialect-independent transformations
28    with Interfaces. Here we will show how to plug dialect specific information
29    into generic transformations like shape inference and inlining.
30-   [Chapter #5](Ch-5.md): Partially lowering to lower-level dialects. We'll
31    convert some of our high level language specific semantics towards a generic
32    affine oriented dialect for optimization.
33-   [Chapter #6](Ch-6.md): Lowering to LLVM and code generation. Here we'll
34    target LLVM IR for code generation, and detail more of the lowering
35    framework.
36-   [Chapter #7](Ch-7.md): Extending Toy: Adding support for a composite type.
37    We'll demonstrate how to add a custom type to MLIR, and how it fits in the
38    existing pipeline.
39
40The [first chapter](Ch-1.md) will introduce the Toy language and AST.
41