Pgo

Walks you through the full two-stage PGO workflow for C/C++ projects: instrument your binary, run representative workloads to collect profile data, then rebuild with optimizations tuned to actual runtime behavior. Covers both GCC and Clang approaches, plus BOLT for post-link binary layout optimization when you need that extra 5-15% after PGO. Most useful when standard O3 optimization has plateaued and you're working with branch-heavy code or large binaries like compilers and databases. The CMake integration examples are solid, and the SamplePGO section is handy if you need production profiling without instrumentation overhead. This is what you reach for when performance actually matters and you've exhausted the easy wins.

skills.sh installs: 610
GitHub stars: 279
last updated: Aug 21, 2026

View on GitHub

Install

$ npx -y skills add mohitmishra786/low-level-dev-skills --skill pgo --agent claude-code

Installs into .claude/skills of the current project.

Files

PGO (Profile-Guided Optimisation)

Purpose

Guide agents through the full PGO workflow: instrument build → representative workload → collect profile → optimised build, covering both GCC and Clang, plus BOLT for post-link optimisation.

Triggers

"How do I use PGO to speed up my binary?"
"What is profile-guided optimization and when should I use it?"
"How do I use -fprofile-generate and -fprofile-use?"
"My -O3 build isn't fast enough — what next?"
"How does BOLT differ from PGO?"
"How do I collect representative profile data?"

Workflow

1. When to use PGO

Is -O3 -march=native already applied?
  no  → apply standard optimisation first
  yes → is workload branch-heavy or has irregular call patterns?
          yes → PGO will likely help 5-30%
          no  → PGO may not help; profile first with linux-perf

PGO helps most with: