Xv6 Kernel Optimization

View GitHub View Documentation

This project implements a collection of advanced system calls for the XV6 kernel. The goal is to expand on XV6’s core functionality with file system traversal, inter-process communication, and optimized I/O operations. It is based on MIT’s 6.1810 (formerly 6.828) Operating Systems course.

This project accomplishes the following objectives:

• Implements standard Unix user-level utilities: sleep, which pauses execution for a set number of ticks; find, which recursively searches directories for specific filenames; and xargs, which executes commands from standard input.
• Improves memory allocation by replacing statically declared arrays with a buddy allocator and implementing lazy page allocation for user-space heap memory.
• Optimizes the fork() system call by using a copy-on-write method that initially shares physical memory pages between parent and child processes, as opposed to directly duplicating them.
• Adds to the capabilities of the existing Xv6 file system by supporting much larger file sizes and implementing symbolic links.
• Implements the mmap and munmap system calls, allowing processes to dynamically map files directly into memory and share memory with other processes.