<div align="center">
# `SPIR-🇹`
**⋯🢒 🇹arget 🠆 🇹ransform 🠆 🇹ranslate ⋯🢒**
[![Crates.io](https://img.shields.io/crates/v/spirt.svg)](https://crates.io/crates/spirt)
[![Docs](https://docs.rs/spirt/badge.svg)](https://docs.rs/spirt)
[![Git Docs](https://img.shields.io/badge/git%20main%20docs-published-blue)](https://rust-gpu.github.io/spirt/spirt/index.html)
[![Build status](https://github.com/rust-gpu/spirt/workflows/CI/badge.svg)](https://github.com/rust-gpu/spirt/actions)
</div>
**SPIR-🇹** is a research project aimed at exploring shader-oriented IR designs derived from SPIR-V, and producing a framework around such an IR to facilitate advanced compilation pipelines, beyond what existing SPIR-V tooling allows for.
Such a need arose in the [Rust-GPU] project, which requires a variety of legalization passes to turn general-purpose (Rust<sup>1</sup>) code operating on *untyped* memory, into GPU-friendly direct data-flow.
Our goal is to replace the existing [Rust-GPU] SPIR-V legalizations passes with **SPIR-🇹** equivalents - but even more imporantly, **SPIR-🇹** should allow writing much more powerful legalization/optimization passes, that would've been unfathomable<sup>2</sup> for direct SPIR-V manipulation.
---
<sub><sup>1</sup> Rust is not unique in its needs here, and more languages (or IRs) could eventually make use of such a framework, but the initial design and implementation work has focused on [Rust-GPU]</sub>
<sub><sup>2</sup> not outright impossible, but requiring excessive development/maintenance cost, having to constantly balance correctness and power (more conservative passes are easier to trust), etc.</sub>
## Disclaimer
This project is not affiliated, associated, authorized, endorsed by, or in any way officially connected with Khronos Group Inc., or any of its subsidiaries or its affiliates. The official Khronos Group Inc. website can be found at https://www.khronos.org.
The names SPIR, SPIR-V, as well as related names, marks, emblems and images are trademarks of their respective owners.
<sub><i>Additional context: the name of this project is a pun on SPIR-V, and entirely unrelated to SPIR (the older IR standard).</i></sub>
## Status
🚧 *This project is in active design and development, many details can and will change* 🚧
If you're interested in using **SPIR-🇹** yourself, you may want to first take a look at [the issue tracker](https://github.com/rust-gpu/spirt/issues) for relevant issues, and even open new ones describing your usecase.
With the initial focus being on [Rust-GPU]'s usecase, various (otherwise desirable) functionality/APIs/docs may be lacking, or rapidly changing - at the same time, discussions around widening the scope and usability of **SPIR-🇹** _in the long term_ are still welcome.
### Non-goals (at least in the short term)
* supporting the ("OpenCL") `Kernel` dialect of SPIR-V
* `Kernel` SPIR-V is much closer to LLVM IR, than `Shader` SPIR-V, and
as such tooling oriented around LLVM is more likely to be a better fit
* textual syntax that can be parsed back
* i.e. the pretty-printer output is purely a visualization
### Designed and implemented so far
<table>
<tr><td width="50%">
**IR data types**:
* allowing near-arbitrary SPIR-V instructions for any unrecognized opcodes
* IDs are replaced with interned/"entity" handles (see below)
* interning for attributes (decorations & similar), types and constants
* i.e. automatic deduplication, efficient indexing, and no concept of "definition"
(only uses of interned handles can lead to a module being considered to contain a specific type/constant)
* "entity" system for e.g. definitions in a module, instructions in a function, etc.
* disallows iteration in favor of/forcing the use of efficient indexing
* structured control-flow "regions" inspired by RVSDG, stricter than SPIR-V
(see `ControlRegionDef`'s docs for more details)
</td><td>
**Framework utilities**:
* `visit`/`transform`: immutable/mutable IR traversal
* `print`: pretty-printer with (styled and hyperlinked) HTML output
**Passes (to/from/on SPIR-🇹)**:
* `spv::lower`: "lowering" from SPIR-V, normalizing away many irrelevant details
* lossy for some relevant details (these are bugs, though many are non-semantic so lower priority)
* `spv::lift`: "lifting" back up to SPIR-V, making arbitrary choices where necessary
* comparable to e.g. generating GLSL syntax from SPIR-V, just one level down
* `cfg::Structurizer`: (re)structurization, from arbitrary control-flow to the stricter structured "regions"
* `passes::link`: mapping (linkage) imports to relevant exports
</td></tr></table>
## Simple example (with non-trivial control-flow)
<table>
<tr><td>
<div align="center">
**GLSL** ([`for-loop.vert.glsl`](tests/data/for-loop.vert.glsl))</div>
<sup>
```glsl
#version 450
out int output0;
void main() {
int o = 1;
for(int i = 1; i < 10; i++)
o *= i;
output0 = o;
}
```
</sup>
<div align="center">
**WGSL** ([`for-loop.wgsl`](tests/data/for-loop.wgsl))</div>
<sup>
```glsl
@vertex
fn main() -> @location(0) i32 {
var o: i32 = 1;
for(var i: i32 = 1; i < 10; i++) {
o *= i;
}
return o;
}
```
</sup>
</td><td>
<div align="center">
**SPIR-🇹**</div>
<sup>
```cxx
#[spv.Decoration.Flat]
#[spv.Decoration.Location(Location: 0)]
global_var GV0 in spv.StorageClass.Output: s32
func F0() -> spv.OpTypeVoid {
loop(v0: s32 <- 1s32, v1: s32 <- 1s32) {
v2 = spv.OpSLessThan(v1, 10s32): bool
(v3: s32, v4: s32) = if v2 {
v5 = spv.OpIMul(v0, v1): s32
v6 = spv.OpIAdd(v1, 1s32): s32
(v5, v6)
} else {
(spv.OpUndef: s32, spv.OpUndef: s32)
}
(v3, v4) -> (v0, v1)
} while v2
spv.OpStore(Pointer: &GV0, Object: v0)
}
```
</sup>
</td><td>
<div align="center">
**SPIR-V** ([`for-loop.wgsl.spvasm`](tests/data/for-loop.wgsl.spvasm))</div>
<sup>
```llvm
%typeof_output0 = OpTypePointer Output %i32
%output0 = OpVariable %typeof_output0 Output
%typeof_main = OpTypeFunction %void
%main = OpFunction %void None %typeof_main
%entry = OpLabel
OpBranch %bb0
%bb0 = OpLabel
OpBranch %bb1
%bb1 = OpLabel
%o = OpPhi %i32 %1_i32 %bb0 %o_next %bb5
%i = OpPhi %i32 %0_i32 %bb0 %i_next %bb5
OpLoopMerge %bb6 %bb5 None
OpBranch %bb2
%bb2 = OpLabel
%cond = OpSLessThan %bool %i %10_i32
OpSelectionMerge %bb4 None
OpBranchConditional %cond %bb4 %bb3
%bb3 = OpLabel
OpBranch %bb6
%bb4 = OpLabel
%o_next = OpIMul %i32 %o %i
OpBranch %bb5
%bb5 = OpLabel
%i_next = OpIAdd %i32 %i %1_i32
OpBranch %bb1
%bb6 = OpLabel
OpStore %output0 %o
OpReturn
OpFunctionEnd
```
</sup>
</td></tr></table>
## GPU (shader) IR landscape overview
*(and the vision of how **SPIR-🇹** fits into it)*
![](docs/landscape.svg)
The distinction being made here is between:
* **Interchange IRs** (standards that many tools can use to interoperate)
* SPIR-V was very much intended as such a standard
(outside of the GPU space, wasm is also a great example)
* they only need to encode the right concepts, not straying too far away from what tools understand, but the design effort is often oriented around being a "serialization" format
* **Compiler IRs** (non-standard implementation details of compilers)
* LLVM is quite well-known, but Mesa's NIR is even closer to **SPIR-🇹**
(both being shader-oriented, and having similar specialized choices of e.g. handling control-flow)
* these _have to_ handle legalization/optimization passes quite well, and in general a lot of on-the-fly transformations - as their main purpose is to _expedite_ such operations
* this is where **SPIR-🇹** sits, as a kind of "relative"/dialect of SPIR-V, but making trade-offs in favor of the "intra-compiler" usage
## Contribution
[![Contributor Covenant](https://img.shields.io/badge/contributor%20covenant-v1.4-ff69b4.svg)](CODE_OF_CONDUCT.md)
We welcome community contributions to this project.
Please read our [Contributor Guide](CONTRIBUTING.md) for more information on how to get started.
Please also read our [Contributor Terms](CONTRIBUTING.md#contributor-terms) before you make any contributions.
Any contribution intentionally submitted for inclusion shall comply with the Rust standard licensing model (MIT OR Apache 2.0) and therefore be dual licensed as described below, without any additional terms or conditions:
### License
This contribution is dual licensed under EITHER OF
- Apache License, Version 2.0, ([LICENSE-APACHE](LICENSE-APACHE) or <http://www.apache.org/licenses/LICENSE-2.0>)
- MIT license ([LICENSE-MIT](LICENSE-MIT) or <http://opensource.org/licenses/MIT>)
at your option.
For clarity, "your" refers to any licensee/user of the contribution.
[Rust-GPU]: https://github.com/rust-gpu/rust-gpu