Chapter 10. Performance Hunt 1: A CPU-Bound Application (GIMP)

This chapter contains an example of how to use the Linux performance tools to find and fix performance problems in a CPU-bound application.

After reading this chapter, you should be able to

  • Figure out which source lines are using all the CPU in a CPU-bound application.

  • Use ltrace and oprofile to figure out how often an application is calling various internal and external functions.

  • Look for patterns in the applications source, and search online for information about how an application behaves and possible solutions.

  • Use this chapter as a template for tracking down a CPU-related performance problem.

CPU-Bound Application

In this chapter, we investigate an application that is CPU-bound. It ...

Get Optimizing Linux® Performance: A Hands-On Guide to Linux® Performance Tools now with the O’Reilly learning platform.

O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.