SMOL-V: like Vulkan/Khronos SPIR-V, but smaller.

TL;DR: makes your SPIR-V shaders about 3x smaller, and decodes very fast. If you like to compress them with regular compressor like zlib/lz4/zstd, then smol-v before the compression can still make them 2x smaller.

Overview

SMOL-V encodes Vulkan/Khronos SPIR-V format programs into a form that is smoller, and is more compressible. Normally no changes to the programs are done; they decode into exactly same program as what was encoded. Optionally, debug information can be removed too.

SPIR-V is a very verbose format, several times larger than same programs expressed in other shader formats (e.g. DX11 bytecode, GLSL, DX9 bytecode etc.). The SSA-form with ever increasing IDs is not very appreciated by regular data compressors either. SMOL-V does several things to improve this:

• Many words, especially ones that most often have small values, are encoded using "varint" scheme (1-5 bytes per word, with just one byte for values in 0..127 range).
• Some IDs used in the program are delta-encoded, relative to previously seen IDs (e.g. Result IDs). Often instructions reference things that were computed just before, so this results in small deltas. These values are also encoded using "varint" scheme.
• Reordering instruction opcodes so that the most common ones are the smallest values, for smaller varint encoding.
• Encoding several instructions in a more compact form, e.g. the "typical <=4 component swizzle" shape of a VectorShuffle instruction, or sequences of MemberDecorate instructions.

A somewhat similar utility is spirv-remap from glslang.

See this blog post for more information about how I did SMOL-V.

Usage

Add source/smolv.h and source/smolv.cpp to your C++ project build. It might require C++11 or somesuch. There are Github Actions set up for this project, that build on Windows (VS2022), macOS (Xcode 15) and Linux (Ubuntu 22 / gcc 11).

smolv::Encode and smolv::Decode is the basic functionality. See smolv.h.

SPIR-V versions up to and including 1.6 are supported currently.

Other functions are for development/statistics purposes, to figure out frequencies and distributions of the instructions.

There's a test + compression benchmarking suite in testing/testmain.cpp, using that needs adding other files under testing/external to the build too (3rd party code: glslang remapper 14.3.0, Zstd 1.5.6, LZ4 1.10, miniz).

Changelog

See Changelog.

Limitations / TODO

• SPIR-V where the words got stored in big-endian layout is not supported yet.
• The whole thing might not work on Big-Endian CPUs. It might, but I'm not 100% sure.
• Not much prevention is done against malformed/corrupted inputs, TODO.
• Out of memory cases are not handled. The code will either throw exception or crash, depending on your compilation flags.

License

Code itself: Public Domain or MIT, pick whichever works better for you.

There is 3rd party code under the testing framework (testing/external); it is not required for using SMOL-V. Most of that code (glslang, LZ4, Zstd, sokol_time.h) is BSD or zlib-licensed, and taken from github repositories of the respective projects. miniz is public domain.

There are SPIR-V binary shader dumps under tests/spirv-dumps for compression testing; these are not required for using SMOL-V. Not sure how to appropriately "license" them (but hey they are kinda useless by themselves out of context), so I'll go with this: "Binary shader dumps under 'tests' folder are only to be used for SMOL-V testing". Details on them:

• tests/spirv-dumps/dota2 - some shaders from DOTA2, Copyright Valve Corporation, all rights reserved.
• tests/spirv-dumps/shadertoy - several most popular shaders from Shadertoy, converted to Vulkan SPIR-V via glslang. Copyrights by their individual authors (filename matches last component of shadertoy URL).
• tests/spirv-dumps/talos - some shaders from The Talos Principle, Copyright (c) 2002-2016 Croteam All rights reserved.
• tests/spirv-dumps/unity - various Unity shaders, produced through a HLSL -> DX11 bytecode -> HLSLcc -> glslang toolchain.

Results

As of 2024 Sep 23, results on 383 shaders (under tests/spirv-dumps) are:

Compressed with <none>:
Raw        5948.7KB 100.0%
Remapper   5602.9KB  94.2%
SmolV      2020.9KB  34.0%
Compressed with zlib:
Raw        1550.6KB  26.1%
Remapper   1419.2KB  23.9%
SmolV       790.6KB  13.3%
Compressed with LZ4 HC:
Raw        1769.6KB  29.7%
Remapper   1551.1KB  26.1%
SmolV       823.1KB  13.8%
Compressed with Zstandard:
Raw        1125.5KB  18.9%
Remapper    887.4KB  14.9%
SmolV       573.1KB   9.6%
Compressed with Zstandard 20:
Raw         834.5KB  14.0%
Remapper    666.7KB  11.2%
SmolV       497.4KB   8.4%

Decoding these 383 shaders from SMOL-V back into SPIR-V takes 10.1ms (VS2022, x64 Release, AMD Ryzen 5950X, one thread).

• "Raw" is just raw SPIR-V, with no extra processing.
• "Remapper" is spirv-remap from glslang, with debug info stripping.
• SmolV is what you're looking at, with debug info stripping too.
• zlib, LZ4HC and Zstd are general compression algorithms at default settings (Zstd20 is Zstd compression with almost max setting of 20).
• Compression is done on the whole blob of all the test programs (not individually for each program).