Reading strace: errno, filters, descriptors, -f, -c, -T and -k. Distinguishing CPU time from waiting and understanding ptrace overhead.
strace as a Debugging Primitive — When the Stack Trace Lies Read Post »
Analyzing Linux segfaults: MMU, page faults, SIGSEGV, mappings and gdb core dumps. Null pointers, stack overflow and use-after-free without guesswork.
Anatomy of a Segfault — From MMU Through Kernel to gdb Core Dump Read Post »
Profiling Bash startup, lazy loading nvm and pyenv, cached completions and prompt overhead. Practical .bashrc changes guided by measurement.
.bashrc Optimization — Measuring and Speeding Up Bash Startup Read Post »
epoll reports readiness; io_uring returns operation results. Comparing I/O models, batching, SQPOLL and deployment constraints without universal RPS thresholds.
epoll vs io_uring — When the Event Loop Isn’t Enough Read Post »
DRY is about knowledge, not similar lines. Distinguishing shared rules from incidental duplication and recognizing the cost of premature abstraction.
The DRY Myth — When Code Duplication Is the Correct Architecture Read Post »
Limiting process resources with cgroups v2 and systemd-run: controller hierarchy, CPU and memory budgets, and launching workers from Python.
cgroups v2 Without Docker — CPU and Memory Limits with systemd Read Post »
Exploring an unfamiliar repository: entry points, call paths, grep and rg, LSP, Git history and running one concrete execution path.
Reading Other People’s Code — From Entry Points to Understanding the System Read Post »
What Linux copies after fork() and what stays shared: copy-on-write, file offsets, mutexes, exec and collecting child-process status.
Linux fork() — Copy-on-Write, File Descriptors and Child Processes Read Post »
Debugging methodically: describe the symptom, establish expected behavior, form a hypothesis and test it with a debugger, logs or git bisect.
Debugging Code — Hypotheses Instead of Guesswork Read Post »