Ltrace

ltrace

If strace shows what the kernel saw, ltrace shows what userland asked its libraries to do.

When to use ltrace

Use ltrace when you need to:

• watch libc/API usage such as strcmp, fopen, memcpy, or system
• infer control-flow choices from arguments and return values
• identify likely string checks, password comparisons, or parsing logic
• complement strace with a friendlier view of dynamic library behavior

Quick Start

# Trace library calls
ltrace ./chall

# Follow children and save output
ltrace -f -o ltrace.log ./chall

# Show longer strings
ltrace -f -s 256 ./chall

High-Value Workflows

Focus on suspicious calls

ltrace -e strcmp+strncmp+memcmp ./chall
ltrace -e malloc+free+realloc ./chall
ltrace -e fopen+fread+fwrite+system ./chall

Attach to an already running process

ltrace -p 1234

Practical Notes

• Start here when you suspect a challenge is hiding comparisons or parser behavior in libc calls.
• Combine with strings to guess a secret check, then use ltrace -e strcmp+memcmp to confirm it.
• Pair with strace when you need both the high-level call and the final syscall effect.

Caveats

• Static binaries, direct syscalls, or stripped custom runtimes reduce usefulness sharply.
• Some optimized calls inline away the exact function you hoped to observe.
• Anti-debug protections can interfere just as they do with other ptrace-based tools.

Resources

No bundled scripts/, references/, or assets/. Use the ltrace man page for expression filters, prototype handling, and timing output.