Flamegraphs

mohitmishra786/low-level-dev-skills/skills/profilers/flamegraphs

作者 mohitmishra786bdc58472fa9f無授權條款253 個星標收錄於 2026年10月9日更新於 2026年10月9日儲存庫3 個月前更新

Flamegraph generation and interpretation skill. Use when converting perf, Valgrind Callgrind, or other profiler output into SVG flamegraphs using Brendan Gregg's FlameGraph tools, or when reading flamegraphs to identify performance bottlenecks. Activates on queries about flamegraphs, stackcollapse, flamegraph.svg, identifying hot frames, wide vs tall frames, or performance visualisation.

僅含說明Software DevelopmentData & Analytics
AI 產生的概覽

指導如何從效能分析工具輸出產生 SVG 火焰圖，並解讀火焰圖以找出效能瓶頸。

功能
此技能引導代理使用 Brendan Gregg 的 FlameGraph 工具，完成從效能分析資料到 SVG 火焰圖的流程，涵蓋 perf、Valgrind Callgrind、Go pprof、DTrace 與 async-profiler 等輸入。它也說明如何閱讀火焰圖，包括寬框架與高框架的意義、前後對比的差異火焰圖，以及 flamegraph.pl 的各種選項。它產出的是指令步驟與解讀指引，本身不產生檔案。
適用情境
適用於將 perf、Callgrind 或其他效能分析工具輸出轉換為 SVG 火焰圖，或閱讀既有火焰圖以找出熱點框架。也適合用差異火焰圖比較最佳化前後的效能剖析結果。
執行需求
需要 FlameGraph 工具指令碼（stackcollapse-perf.pl、flamegraph.pl、difffolded.pl 等），以及對應的效能分析工具，例如 perf、Valgrind Callgrind、Go pprof、DTrace 或 async-profiler。此技能本身不附帶指令碼，也不需要憑證，但複製 FlameGraph 儲存庫需要網路存取。

Flamegraphs

Purpose

Guide agents through the pipeline from profiler data to SVG flamegraph, and teach interpretation of flamegraphs to drive concrete optimisation decisions.

Triggers

  • "How do I generate a flamegraph from perf data?"
  • "How do I read a flamegraph?"
  • "The flamegraph shows a wide frame — what does that mean?"
  • "How do I generate a flamegraph from Callgrind?"
  • "I want to compare two flamegraphs (before/after)"

Workflow

1. Install FlameGraph tools

bash
git clone https://github.com/brendangregg/FlameGraph# No install needed; scripts are in the repoexport PATH=$PATH:/path/to/FlameGraph

2. perf → flamegraph (most common path)

bash
# Step 1: recordperf record -F 999 -g -o perf.data ./prog
# Step 2: generate script outputperf script -i perf.data > out.perf
# Step 3: collapse stacksstackcollapse-perf.pl out.perf > out.folded
# Step 4: generate SVGflamegraph.pl out.folded > flamegraph.svg
# Step 5: viewxdg-open flamegraph.svg     # Linuxopen flamegraph.svg          # macOS

One-liner:

bash
perf record -F 999 -g ./prog && perf script | stackcollapse-perf.pl | flamegraph.pl > fg.svg

3. Differential flamegraph (before/after)

bash
# Collect two profilesperf record -g -o before.data ./prog_oldperf record -g -o after.data ./prog_new
# Collapseperf script -i before.data | stackcollapse-perf.pl > before.foldedperf script -i after.data  | stackcollapse-perf.pl > after.folded
# Diff (red = regressed, blue = improved)difffolded.pl before.folded after.folded | flamegraph.pl > diff.svg

4. Callgrind → flamegraph

bash
valgrind --tool=callgrind --callgrind-out-file=cg.out ./progstackcollapse-callgrind.pl cg.out | flamegraph.pl > fg.svg

5. Other profiler inputs

bash
# Go pprofgo tool pprof -raw -output=prof.txt progstackcollapse-go.pl prof.txt | flamegraph.pl > fg.svg
# DTracedtrace -x ustackframes=100 -n 'profile-99 /execname=="prog"/ { @[ustack()] = count(); }' \  -o out.stacks sleep 10stackcollapse.pl out.stacks | flamegraph.pl > fg.svg
# Java (async-profiler)async-profiler -d 30 -f out.collapsed PIDflamegraph.pl out.collapsed > fg.svg

6. Reading flamegraphs

A flamegraph is a call-stack visualisation:

  • X axis: time on CPU (not time sequence) — wider = more time
  • Y axis: call stack depth — taller = deeper call chain
  • Color: random (no significance) — unless using differential mode

What to look for:

PatternMeaningAction
Wide frame near bottomFunction itself is hotOptimise that function
Wide frame with tall narrow towersCalling many different calleesHot dispatch; reduce call overhead
Very tall stack with wide baseDeep recursionCheck recursion depth; consider iterative approach
Plateau at the topLeaf function with no calleesThis leaf is the actual hotspot
Many narrow identical stacksMany threads doing the same workConsider parallelism or batching

Identifying the actionable hotspot:

  1. Find the widest top frame (a frame with no or narrow children above it)
  2. That is where CPU time is actually spent
  3. Trace down to understand what called it and why

Differential flamegraph:

  • Red frames: more time in new profile (regression)
  • Blue frames: less time in new profile (improvement)
  • Frames only in one profile appear solid colored

7. flamegraph.pl options

bash
flamegraph.pl --title "My App" \              --subtitle "Release build, workload X" \              --width 1600 \              --height 16 \              --minwidth 0.5 \              --colors java \              out.folded > fg.svg
OptionEffect
--titleSVG title
--widthWidth in pixels
--heightFrame height in pixels
--minwidthOmit frames < N% (reduces clutter)
--colorsPalette: hot (default), mem, io, java, js, perl, red, green, blue
--invertedIcicle chart (roots at top)
--reverseReverse stacks
--cpConsistent palette (same frame = same color across SVGs)

References

For tool installation, stackcollapse scripts, and palette options, see references/tools.md [blocked].

Related skills

  • Use skills/profilers/linux-perf to collect perf data
  • Use skills/profilers/valgrind to collect Callgrind data
  • Use skills/compilers/clang for LLVM PGO from sampling profiles

來源與署名

來源：mohitmishra786/low-level-dev-skills位於skills/profilers/flamegraphs提交bdc5847

授權條款： 無授權條款

內容歸原作者所有。SourceWeft 從公開儲存庫中收錄這些內容。

檢舉或申請下架

更多來自 mohitmishra786/low-level-dev-skills 的技能

Zig Debugging

mohitmishra786

指導使用 GDB、LLDB 偵錯 Zig 程式，解讀 panic、錯誤返回追蹤、日誌記錄及 IDE 設定。

Software Development2533 個月前更新

Simd Intrinsics

mohitmishra786

指導使用 x86 SSE/AVX 與 ARM NEON 內建函式撰寫及診斷 SIMD 向量化。

Software Development2533 個月前更新

Zig Cross

mohitmishra786

指導 Zig 交叉編譯：目標三元組、CPU 特性、以 zig cc 編譯 C、嵌入式與 WASM 目標。

Software Development2533 個月前更新

Rustc Basics

mohitmishra786

指導 Rust 編譯器設定：RUSTFLAGS、Cargo 設定檔、建置調校，以及檢視 MIR 或組譯輸出。

Software Development2533 個月前更新

Zig Cinterop

mohitmishra786

指導 Zig 與 C 互通：@cImport、translate-c、extern 與 packed 結構，以及 C/Zig 混合建置。

Software Development2533 個月前更新

Zig Build System

mohitmishra786

指導撰寫與除錯 Zig 的 build.zig 檔案，涵蓋目標、C 原始碼、選項、測試與套件清單。

Software Development2533 個月前更新

更多Software Development技能

Playground

anthropics

精選

建立自成一體的互動式 HTML 遊樂場，包含控制項、即時預覽與可複製的提示詞輸出。

Software Development37K今天更新

M5 Onboard

anthropics

精選

透過 USB 偵測 M5Stack ESP32 開發板，燒錄 UIFlow 2.0 韌體並安裝 MicroPython 應用程式套件。

Software Development37K今天更新

Cardputer Buddy

anthropics

精選

指導迭代 Cardputer-Adv 的 MicroPython 應用程式套件：新增應用程式、透過 USB 序列埠推送檔案、查看日誌並執行 REPL 指令。

Software Development37K今天更新

Web Design Guidelines

vercel-labs

精選

依據從遠端來源取得的 Web Interface Guidelines 審查 UI 程式碼，並以 file:line 形式回報問題。

Software Development32K5 週前更新

Vercel React Native Skills

vercel-labs

精選

React Native 與 Expo 最佳實務規則，涵蓋清單效能、動畫、導覽、UI 模式、狀態與 monorepo 設定。

Software Development32K5 週前更新

Vercel React Best Practices

vercel-labs

精選

Vercel 的 React 與 Next.js 效能指南，以 70 條規則的形式用於撰寫、審查與重構程式碼。

Software Development32K5 週前更新