jlfwong/speedscope

Format Support: concatenated output from many calls to jstack

Open

#173 创建于 2018年10月7日

在 GitHub 查看
 (3 评论) (0 反应) (0 负责人)TypeScript (4,040 star) (181 fork)batch import
charity donation bountyhelp wantedimport source

描述

Poor man's profiling of a JVM:

$ for i in `seq 1 300`;do jstack $(pgrep java) > $i.jstack;sleep 1;done;zip profile.zip *.jstack;rm *.jstack

...and then uploading profile.zip. I'm tired of opening up traffic for profiling to production and/or building lots of tooling around it. jstack (and shell) is always availalable.

贡献者指南

技术栈
typescript
领域
performancedata
议题类型
feature
难度面向新贡献者的预计实现难度，1 表示很小改动，5 表示专家级工作。
3
预计时间有经验贡献者完成调查、实现、测试并准备 pull request 的粗略时间范围。
1-3 hours
活动状态议题当前的可参与程度：新鲜、活跃、陈旧、阻塞或等待维护者输入。
stale
清晰度议题是否清楚说明期望改动、验收标准和下一步。
clear
前置要求
TypeScriptspeedscope import architectureJVM stack trace format
新手友好度1-100 的估计分数，表示该议题对首次贡献者的友好程度。
60
研究方向
First, examine speedscope's existing import system in the src/import/ directory to understand how other profile formats (e.g., Chrome trace) are parsed. Then, implement a new importer that reads a zip file containing multiple jstack outputs, extracts each file, parses the thread dumps, and concatenates them into a single profile. Test the new format by creating sample jstack outputs and verifying the viewer displays them correctly. Check issue comments for any additional maintainer guidance.