Systems programming as a swiss army knife
summary
Demos run three mysteries — missing configuration, a slow program and a French website — and resolve them with concrete observations: estrace shows which file is opened so editing the wrong rc file is obvious, one program spends 5% CPU waiting for TCP while another uses 99% CPU adding 14 million numbers, a third writes ~300MB to disk and hits ecrypt Fs overhead revealed by perf top and perf stat, and HTTP language negotiation appears in packets captured by tcpdump / wire shark.
AI-written summary · about
chapters (16)
- 0:00Start
- 1:23all bugs are easy (with the right tools)
- 2:04don't be scared to go deeper
- 3:25missing @ configuration file
- 5:33demo
- 6:10strace can make your program ron 50x slower
- 9:12mystery program #1
- 9:45what is it waiting for?
- 10:50mystery program #2
- 11:20USE A PYTHON PROPILER
- 11:39mystery program #3
- 23:17LATENCY NUMBERS EVERY PROGRAMMER SHOULD KNOW
- 30:40save network traffic to analyze later
- 32:31topdump -A print packets to your Screen!
- 35:05learn your operating system tools
- 35:53you can be a wizard
related talks
47mMinimum Viable Linux
32mListen to your Engine
26mExposing Hidden Exploitable Behaviors In Programming Languages Using Differential Fuzzing
52mUnlocking secrets of proprietary software using Frida
55mHow hacking works
53mHow hacking works - Web edition
29mAre Our Systems Using up-to-Date Cryptography?
02h 27mHow hacking works (Workshop edition)