Evaluate using Profile-Guided Optimization (PGO) and Post Link Optimization (PLO) for the Inko tooling ecosystem

[zamazan4ik]

Description

Recently I did many Profile-Guided Optimization (PGO) benchmarks on multiple projects (including static analysis tools and compilers like Rustc, Clang, Clangd, Clang Tidy, and many others) - the results are available here. I think it's worth trying to apply PGO to the Inko tooling ecosystem like the compiler (and maybe something else like code formatters, LSP server, etc.).

I can suggest the following things to do:

• Evaluate PGO's applicability to Inko tooling.
• If PGO helps to achieve better performance - add a note to Inko's documentation about that. In this case, users and maintainers will be aware of another optimization opportunity for Inko tools.
• Provide PGO integration into the build scripts. It can help users and maintainers easily apply PGO for their workloads.
• Optimize prebuilt binaries with PGO.

Here are some examples of how PGO is already integrated into other projects' build scripts:

• Rustc: a CI script for the multi-stage build
• GCC:
  • Official docs, section "Building with profile feedback" (even AutoFDO build is supported)
  • A part in a "wonderful" configure script
• Clang: Docs
• Python:
  • CPython: README
  • Pyston: README
• Go: Bash script
• V8: Bazel flag
• ChakraCore: Scripts
• Chromium: Script
• Firefox: Docs
  • Thunderbird has PGO support too
• PHP - Makefile command and old Centminmod scripts
• MySQL: CMake script
• YugabyteDB: GitHub commit
• FoundationDB: Script
• Zstd: Makefile
• Foot: Scripts
• Windows Terminal: GitHub PR
• Pydantic-core: GitHub PR

Some PGO documentation examples in various projects:

• ClickHouse: https://clickhouse.com/docs/en/operations/optimizing-performance/profile-guided-optimization
• Databend: https://databend.rs/doc/contributing/pgo
• Vector: https://vector.dev/docs/administration/tuning/pgo/
• Nebula: https://docs.nebula-graph.io/3.5.0/8.service-tuning/enable_autofdo_for_nebulagraph/
• GCC: Official docs, section "Building with profile feedback" (even AutoFDO build is supported)
• Clang:
  • https://llvm.org/docs/HowToBuildWithPGO.html
  • https://llvm.org/docs/AdvancedBuilds.html
• tsv-utils: https://github.com/eBay/tsv-utils/blob/master/docs/BuildingWithLTO.md

After PGO, I can suggest evaluating LLVM BOLT as an additional optimization step after PGO - Post Link Optimization (PLO). But I suggest starting with PGO - it's a more stable optimization than PLO in the general case.

For the Rust projects, I recommend starting with cargo-pgo.

Related work

The issue is not the same as #650 . In #650 I propose to add a possibility to compile Inko programs with PGO. In this issue, I suggest compiling Inko tools like the compiler with PGO and PLO.

[yorickpeterse]

PGO is something that indeed can be useful at some point in the future, though it will likely also require us to host pre-built binaries so others can take advantage of it (e.g. Linux distributions aren't going to run PGO while building I think).

[zamazan4ik]

though it will likely also require us to host pre-built binaries so others can take advantage of it (e.g. Linux distributions aren't going to run PGO while building I think).

Not, that it's not a mandatory requirement :) Users can benefit from having PGO too since some people rebuild binary for their localhost machines (Gentoo-style, you know). Somewhere rebuilding is a mandatory requirement (usually due to some security policies).

Regarding Linux distributions - PGO is often used in Linux-based repositories too. GCC, Clang, CPython, Chromium, Firefox - are the most popular packages with enabled PGO in the Linux repositories.

For all of these use-cases having an ability to easily enable PGO - is a very nice feature to have.

[yorickpeterse]

We're not going to host any PGO data any time soon, nor will we start hosting binaries. This means it's up to end users/packagers to apply PGO data.