More GDB wizardry and 8 other essential Linux application debugging tools
summary
Demos show modern GDB features such as reverse debug (full record/replay to step back to the exact instruction and inspect locals), dynamic printf and Python-driven pretty‑printers plus the TUI/dashboard. Hardware watchpoints use CPU support (typically up to four) and GDB falls back to slow software watchpoints. Other tools shown: ftrace to trace signals, strace/ltrace for syscalls and libc calls, valgrind catching uninitialized stack data in a customer binary, rr/Undo for fast recording and ASan/perf/Intel PT for memory and branch tracing.
AI-written summary · about
chapters (19)
- 0:00Introduction to debugging
- 8:34GDB basics and TUI mode
- 14:11GDB Python scripting and pretty printers
- 23:22GDB internals and P trace
- 28:44Watchpoints and hardware debugging
- 35:04Thread debugging and dynamic printf
- 42:10Multi-process debugging
- 45:26GDB customization and hooks
- 47:51GCC debug flags and optimization
- 49:18Valgrind for memory debugging
- 53:16Cachegrind and Memcheck
- 55:44AddressSanitizer and fortification
- 59:46F trace for kernel debugging
- 1:05:02Strace L trace and perf
- 1:08:02Intel PT and processor tracing
- 1:10:49Reversible debugging with gdb
- 1:19:11UndoDB and live recording
- 1:23:32JetBrains CLion and undo plugin
- 1:28:42Q&A and closing remarks
related talks
18mFavourite GDB developers tricks
01h 29mModern Linux C++ Debugging Tools - Under the Covers
58m“Debugging Linux C++”
01h 10mCool New Stuff in GDB 9, 10 and 11
01h 07mCool New Stuff in Gdb 9 and Gdb 10
25mDebugging memory issues with Valgrind and GDB
38mProgrammable debuggers and the Linux kernel (drg...
01h 00mInvestigating C++ Applications in Production on Linux and Windows