Fork vs. Vfork: How Memory Sharing Changes Process Safety
The key difference is memory isolation: see how fork() uses copy-on-write, why vfork() suspends the parent, and when each call is safe in practice.
Tag: Linux kernel programming
How Kernel Executors Dispatch Compute to Hardware
Kernel executors route compiled kernels to devices and manage dependencies. Learn how OpenCL queues, PyTorch dispatch, and backend delegates actually work.
How Linux Maps Hardware to Kernel Drivers
Use lspci, lsusb, lsmod, modinfo, and sysfs to identify bound, loaded, built-in, and missing Linux drivers, then trace probe failures with confidence.
SIGUSR1 Default Action: Why Signals Terminate Processes
SIGUSR1 defaults to terminating processes with no core dump. Learn why POSIX chose this behavior, how to install handlers, and debug unexpected signal deaths.
6 Linux IPC Mechanisms: When to Use Each
Pick the right Linux IPC mechanism for your task: pipes for streaming, sockets for client-server, shared memory for throughput, signals for notifications.
Linux Process Life Cycle: States, Signals, and Diagnosis
Linux processes cycle through R, S, D, and T states. Learn to read /proc, use strace, and diagnose stuck tasks without guessing or killing blindly.
Writing a Daemon in C: The Double-Fork Explained
Master daemon creation in C with the double-fork and setsid technique. Verify detachment with ps and strace to ensure your service survives logout.
CONFIG_BPF_SYSCALL: The Kernel Flag eBPF Programs Need
CONFIG_BPF_SYSCALL compiles the bpf() syscall into your kernel, the single entry point every eBPF program load and firewall rule requires to work.
Preverify eBPF Programs in CI Before Kernel Load
Run the kernel verifier off-device during CI to catch eBPF rejections early. Learn verifier basics, log reading, BTF fixes, and troubleshooting methods.
Kernel Overhead: What It Is and How eBPF Cuts It
Kernel overhead drains CPU cycles through privilege transitions, context switching, and memory isolation. Learn what causes it and how to measure it yourself.
Mount Bpffs at /Sys/fs/bpf: CAP_SYS_ADMIN and Persistence
bpffs mount requires CAP_SYS_ADMIN for both the mount and bpf() syscall. Learn how to diagnose failures, make mounts persistent, and secure pins across reboots.
XDP Offload vs Native: Which Mode Runs on Your Hardware
XDP offload runs programs on the NIC processor, not your CPU. Learn which modes actually use hardware and how to verify your real attach mode.
How to Load Linux Kernel Modules With Modprobe
Use modprobe to load Linux kernel modules safely with dependency resolution. This guide covers the commands, diagnostics, and configuration you need.
BCC eBPF Compiler for Linux Kernel Tracing
BCC compiles eBPF C at runtime for deep Linux kernel tracing. Learn when to use BCC versus CO-RE, which tools solve real problems, and how to write safe probes.
Bpftool: BPF Program Inspection and Debugging Guide
Master bpftool for inspecting, loading, and debugging BPF programs. Learn map dumps, program verification, BTF headers, and kernel feature probes.
Katran XDP Load Balancer: Kernel-Native Packet Forwarding
Katran forwards packets at the NIC receive path using eBPF and XDP, eliminating userspace overhead and lock contention to hold at Facebook scale.
How to Contribute to Linux: Start With Documentation
Master Linux kernel contributions by starting small: documentation fixes and checkpatch cleanups teach the workflow without breaking production systems.
Linux Poll() vs Epoll: When to Use Each Syscall
poll() works for small file descriptor sets with POSIX compatibility; switch to epoll at scale. Learn when each syscall fits and common pitfalls to avoid.