5 个仓库
Processes all traffic directly in the operating system kernel for minimal latency and high throughput without user-space context switches.
Distinct from Packet Processing Frameworks: Distinct from Packet Processing Frameworks: focuses on the kernel-space execution model rather than the framework abstraction for packet processing.
Explore 5 awesome GitHub repositories matching networking & communication · Kernel-Space Packet Processing. Refine with filters or upvote what's useful.
itlwm is a macOS network driver and kernel extension that enables Intel wireless network adapters to function on macOS systems. It serves as a hardware-specific driver providing connectivity and stability for Intel Wi-Fi chips on non-native platforms. The project acts as a bridge between Intel hardware and native macOS wireless frameworks, allowing the system to recognize the adapter as a native device and utilize Apple Airport features and system settings. The driver manages hardware integration and network connectivity by interfacing with the operating system to support macOS hardware comp
Handles the transmission and reception of network frames directly within the OS kernel to minimize latency.
This is a structured reference that maps Linux kernel network sysctl parameters to their specific roles in the packet processing pipeline, from the network interface card through to the application layer. It covers the full path of packet traversal, including interrupt mitigation, buffer management, queuing disciplines, and TCP auto-tuning mechanisms. The reference details how to configure kernel parameters for connection handling, including TCP listen backlog, SYN queue management, FIN timeout, and syncookies. It also addresses queue management and bufferbloat control through egress queuing
Documents sysctl parameters for tuning the entire ingress packet processing pipeline from NIC to application.
Katran 是一个基于 eBPF 的四层负载均衡器,专为 Linux 内核内的高性能网络包转发而设计。它利用 XDP 网络包处理器绕过标准内核网络栈,从而最大限度地降低延迟并提高流量吞吐量。 该系统实现了加权一致性哈希,以分发网络负载并维持后端服务器之间的会话持久性。它通过直接服务器返回(DSR)网关进一步优化了返回路径,允许后端直接响应客户端,并将负载均衡器从出站流量流中移除。 其功能包括用于会话稳定性的无锁连接跟踪,以及消除跨节点延迟的 NUMA 感知核心绑定。该项目还支持后端流量加权、服务器排空,以及在负载均衡前执行防火墙逻辑的 XDP 程序链。 该实现为网络流量捕获、健康检查探测路由和封装流量过滤提供了可观测性工具。
Processes network packets directly within the Linux kernel to minimize latency and bypass the standard network stack.
This project provides an open-source VPN client that creates an unrestricted tunnel to bypass internet censorship without fees, registration, speed caps, or data limits. It offers free, unlimited VPN tunneling with no sign‑up required, enabling access to blocked websites and unrestricted browsing. The application uses the WireGuard protocol for encrypted traffic routing, backed by kernel‑level packet forwarding to minimize overhead. It includes automatic server discovery that selects the most responsive server based on real‑time latency, a connection keepalive mechanism to detect and restore
Creates a virtual network interface at the kernel level to intercept and redirect all device traffic into the encrypted tunnel.
Aya is a Rust-native framework for writing, compiling, and loading eBPF programs into the Linux kernel. It provides a complete development environment that eliminates the need for a C toolchain or libbpf, allowing developers to work entirely within the Rust ecosystem. The framework manages the full lifecycle of eBPF programs, including async runtime integration, CO-RE BTF resolution for kernel version portability, ELF-based program loading, and safe kernel memory access. The framework distinguishes itself through its pure Rust compilation pipeline, which compiles Rust source code directly int
Filters, redirects, and inspects network packets at the kernel level using XDP and socket filters.